2025-09-09T14:02:53.2581017Z Current runner version: '2.328.0' 2025-09-09T14:02:53.2587055Z Runner name: 'i-00b63bd9592e830d3' 2025-09-09T14:02:53.2587791Z Runner group name: 'default' 2025-09-09T14:02:53.2588616Z Machine name: 'ip-10-0-56-61' 2025-09-09T14:02:53.2591452Z ##[group]GITHUB_TOKEN Permissions 2025-09-09T14:02:53.2593726Z Contents: read 2025-09-09T14:02:53.2594252Z Metadata: read 2025-09-09T14:02:53.2594751Z Packages: read 2025-09-09T14:02:53.2595284Z ##[endgroup] 2025-09-09T14:02:53.2597187Z Secret source: Actions 2025-09-09T14:02:53.2598061Z Prepare workflow directory 2025-09-09T14:02:53.3149611Z Prepare all required actions 2025-09-09T14:02:53.3187286Z Getting action download info 2025-09-09T14:02:53.6510803Z Download action repository 'actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683' (SHA:11bd71901bbe5b1630ceea73d27597364c9af683) 2025-09-09T14:02:53.9970850Z Download action repository 'pytorch/pytorch@main' (SHA:4dd73e659a8fd4872e5f49cfd72e420fa7c4e6c9) 2025-09-09T14:03:08.9524696Z Download action repository 'actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093' (SHA:d3f86a106a0bac45b974a628896c90dbdf5c8093) 2025-09-09T14:03:09.3243880Z Download action repository 'pmeier/pytest-results-action@a2c1430e2bddadbad9f49a6f9b879f062c6b19b1' (SHA:a2c1430e2bddadbad9f49a6f9b879f062c6b19b1) 2025-09-09T14:03:09.4775619Z Download action repository 'actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02' (SHA:ea165f8d65b6e75b540449e92b4886f43607fa02) 2025-09-09T14:03:09.9978922Z Getting action download info 2025-09-09T14:03:10.2207515Z Uses: pytorch/test-infra/.github/workflows/linux_job_v2.yml@refs/heads/main (e502b6d9079a2a411c68046e8a7694b851c5df33) 2025-09-09T14:03:10.2211076Z ##[group] Inputs 2025-09-09T14:03:10.2212982Z script: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:10.2215311Z timeout: 180 2025-09-09T14:03:10.2215573Z runner: linux.g5.12xlarge.nvidia.gpu 2025-09-09T14:03:10.2215977Z upload-artifact: 2025-09-09T14:03:10.2216656Z upload-artifact-to-s3: false 2025-09-09T14:03:10.2216963Z download-artifact: 2025-09-09T14:03:10.2217207Z repository: 2025-09-09T14:03:10.2217461Z fetch-depth: 1 2025-09-09T14:03:10.2217682Z submodules: recursive 2025-09-09T14:03:10.2217923Z ref: 2025-09-09T14:03:10.2218176Z test-infra-repository: pytorch/test-infra 2025-09-09T14:03:10.2218487Z test-infra-ref: 2025-09-09T14:03:10.2218744Z use-custom-docker-registry: true 2025-09-09T14:03:10.2219068Z docker-image: pytorch/almalinux-builder 2025-09-09T14:03:10.2219389Z docker-build-dir: .ci/docker 2025-09-09T14:03:10.2219660Z gpu-arch-type: cuda 2025-09-09T14:03:10.2219911Z gpu-arch-version: 12.6 2025-09-09T14:03:10.2220156Z job-name: linux-job 2025-09-09T14:03:10.2220405Z continue-on-error: false 2025-09-09T14:03:10.2220705Z binary-matrix: 2025-09-09T14:03:10.2220932Z run-with-docker: true 2025-09-09T14:03:10.2221181Z secrets-env: 2025-09-09T14:03:10.2221423Z no-sudo: false 2025-09-09T14:03:10.2221653Z ##[endgroup] 2025-09-09T14:03:10.2222061Z Complete job name: test (CUDA 2.7, linux.g5.12xlarge.nvidia.gpu, torch==2.7.0, cuda, 12.6) / linux-job 2025-09-09T14:03:10.3050412Z A job started hook has been configured by the self-hosted runner administrator 2025-09-09T14:03:10.3190411Z ##[group]Run '/home/ec2-user/runner-scripts/before_job.sh' 2025-09-09T14:03:10.3208432Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:03:10.3209015Z ##[endgroup] 2025-09-09T14:03:11.7508356Z Runner Type: linux.g5.12xlarge.nvidia.gpu 2025-09-09T14:03:11.7508879Z Instance Type: g5.12xlarge 2025-09-09T14:03:11.7509128Z AMI Name: unknown 2025-09-09T14:03:11.7553941Z AMI ID: ami-05ffe3c48a9991133 2025-09-09T14:03:17.4428978Z ##[group]Run set -euxo pipefail 2025-09-09T14:03:17.4429350Z set -euxo pipefail 2025-09-09T14:03:17.4429656Z if [[ "${NO_SUDO}" == "false" ]]; then 2025-09-09T14:03:17.4430019Z  echo "::group::Cleanup with-sudo debug output" 2025-09-09T14:03:17.4430392Z  sudo rm -rfv "${GITHUB_WORKSPACE}" 2025-09-09T14:03:17.4430685Z else 2025-09-09T14:03:17.4430950Z  echo "::group::Cleanup no-sudo debug output" 2025-09-09T14:03:17.4431290Z  rm -rfv "${GITHUB_WORKSPACE}" 2025-09-09T14:03:17.4431579Z fi 2025-09-09T14:03:17.4431782Z  2025-09-09T14:03:17.4432006Z mkdir -p "${GITHUB_WORKSPACE}" 2025-09-09T14:03:17.4432322Z echo "::endgroup::" 2025-09-09T14:03:17.4447041Z shell: /usr/bin/bash -e {0} 2025-09-09T14:03:17.4447314Z env: 2025-09-09T14:03:17.4447569Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:17.4447932Z REPOSITORY: pytorch/ao 2025-09-09T14:03:17.4448206Z PR_NUMBER: 2963 2025-09-09T14:03:17.4450018Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:17.4451868Z NO_SUDO: false 2025-09-09T14:03:17.4452086Z ##[endgroup] 2025-09-09T14:03:17.4486722Z + [[ false == \f\a\l\s\e ]] 2025-09-09T14:03:17.4499543Z ##[group]Cleanup with-sudo debug output 2025-09-09T14:03:17.4502484Z + echo '::group::Cleanup with-sudo debug output' 2025-09-09T14:03:17.4502878Z + sudo rm -rfv /home/ec2-user/actions-runner/_work/ao/ao 2025-09-09T14:03:17.5565577Z removed directory '/home/ec2-user/actions-runner/_work/ao/ao' 2025-09-09T14:03:17.5588869Z + mkdir -p /home/ec2-user/actions-runner/_work/ao/ao 2025-09-09T14:03:17.5607248Z + echo ::endgroup:: 2025-09-09T14:03:17.5607810Z ##[endgroup] 2025-09-09T14:03:17.5731929Z ##[group]Run actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 2025-09-09T14:03:17.5732359Z with: 2025-09-09T14:03:17.5732650Z repository: pytorch/test-infra 2025-09-09T14:03:17.5732994Z path: test-infra 2025-09-09T14:03:17.5733285Z submodules: recursive 2025-09-09T14:03:17.5733809Z token: *** 2025-09-09T14:03:17.5734092Z ssh-strict: true 2025-09-09T14:03:17.5734360Z ssh-user: git 2025-09-09T14:03:17.5734653Z persist-credentials: true 2025-09-09T14:03:17.5734958Z clean: true 2025-09-09T14:03:17.5735225Z sparse-checkout-cone-mode: true 2025-09-09T14:03:17.5735497Z fetch-depth: 1 2025-09-09T14:03:17.5735716Z fetch-tags: false 2025-09-09T14:03:17.5736055Z show-progress: true 2025-09-09T14:03:17.5736306Z lfs: false 2025-09-09T14:03:17.5736518Z set-safe-directory: true 2025-09-09T14:03:17.5736771Z env: 2025-09-09T14:03:17.5737027Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:17.5737357Z REPOSITORY: pytorch/ao 2025-09-09T14:03:17.5737631Z PR_NUMBER: 2963 2025-09-09T14:03:17.5739441Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:17.5741264Z ##[endgroup] 2025-09-09T14:03:17.7143449Z Syncing repository: pytorch/test-infra 2025-09-09T14:03:17.7144156Z ##[group]Getting Git version info 2025-09-09T14:03:17.7144977Z Working directory is '/home/ec2-user/actions-runner/_work/ao/ao/test-infra' 2025-09-09T14:03:17.7145582Z [command]/usr/bin/git version 2025-09-09T14:03:17.7145841Z git version 2.47.1 2025-09-09T14:03:17.7152138Z ##[endgroup] 2025-09-09T14:03:17.7177084Z Temporarily overriding HOME='/home/ec2-user/actions-runner/_work/_temp/a02f20c6-bb56-4208-a6fd-35e72b0034a0' before making global git config changes 2025-09-09T14:03:17.7177991Z Adding repository directory to the temporary git global config as a safe directory 2025-09-09T14:03:17.7182625Z [command]/usr/bin/git config --global --add safe.directory /home/ec2-user/actions-runner/_work/ao/ao/test-infra 2025-09-09T14:03:17.7221404Z ##[group]Initializing the repository 2025-09-09T14:03:17.7225950Z [command]/usr/bin/git init /home/ec2-user/actions-runner/_work/ao/ao/test-infra 2025-09-09T14:03:17.7274919Z hint: Using 'master' as the name for the initial branch. This default branch name 2025-09-09T14:03:17.7275508Z hint: is subject to change. To configure the initial branch name to use in all 2025-09-09T14:03:17.7276067Z hint: of your new repositories, which will suppress this warning, call: 2025-09-09T14:03:17.7276451Z hint: 2025-09-09T14:03:17.7276744Z hint: git config --global init.defaultBranch 2025-09-09T14:03:17.7277081Z hint: 2025-09-09T14:03:17.7277408Z hint: Names commonly chosen instead of 'master' are 'main', 'trunk' and 2025-09-09T14:03:17.7277940Z hint: 'development'. The just-created branch can be renamed via this command: 2025-09-09T14:03:17.7278352Z hint: 2025-09-09T14:03:17.7278582Z hint: git branch -m 2025-09-09T14:03:17.7279074Z Initialized empty Git repository in /home/ec2-user/actions-runner/_work/ao/ao/test-infra/.git/ 2025-09-09T14:03:17.7287984Z [command]/usr/bin/git remote add origin https://github.com/pytorch/test-infra 2025-09-09T14:03:17.7323917Z ##[endgroup] 2025-09-09T14:03:17.7324335Z ##[group]Disabling automatic garbage collection 2025-09-09T14:03:17.7328161Z [command]/usr/bin/git config --local gc.auto 0 2025-09-09T14:03:17.7571446Z ##[endgroup] 2025-09-09T14:03:17.7571823Z ##[group]Setting up auth 2025-09-09T14:03:17.7577167Z [command]/usr/bin/git config --local --name-only --get-regexp core\.sshCommand 2025-09-09T14:03:17.7611254Z [command]/usr/bin/git submodule foreach --recursive sh -c "git config --local --name-only --get-regexp 'core\.sshCommand' && git config --local --unset-all 'core.sshCommand' || :" 2025-09-09T14:03:17.8069524Z [command]/usr/bin/git config --local --name-only --get-regexp http\.https\:\/\/github\.com\/\.extraheader 2025-09-09T14:03:17.8104382Z [command]/usr/bin/git submodule foreach --recursive sh -c "git config --local --name-only --get-regexp 'http\.https\:\/\/github\.com\/\.extraheader' && git config --local --unset-all 'http.https://github.com/.extraheader' || :" 2025-09-09T14:03:17.8524155Z [command]/usr/bin/git config --local http.https://github.com/.extraheader AUTHORIZATION: basic *** 2025-09-09T14:03:17.8576400Z ##[endgroup] 2025-09-09T14:03:17.8576819Z ##[group]Determining the default branch 2025-09-09T14:03:17.8579508Z Retrieving the default branch name 2025-09-09T14:03:18.1095591Z Default branch 'main' 2025-09-09T14:03:18.1096692Z ##[endgroup] 2025-09-09T14:03:18.1097737Z ##[group]Fetching the repository 2025-09-09T14:03:18.1101219Z [command]/usr/bin/git -c protocol.version=2 fetch --no-tags --prune --no-recurse-submodules --depth=1 origin +refs/heads/main:refs/remotes/origin/main 2025-09-09T14:03:18.4900343Z From https://github.com/pytorch/test-infra 2025-09-09T14:03:18.4900739Z * [new branch] main -> origin/main 2025-09-09T14:03:18.4930603Z ##[endgroup] 2025-09-09T14:03:18.4931013Z ##[group]Determining the checkout info 2025-09-09T14:03:18.4932105Z ##[endgroup] 2025-09-09T14:03:18.4936922Z [command]/usr/bin/git sparse-checkout disable 2025-09-09T14:03:18.4981123Z [command]/usr/bin/git config --local --unset-all extensions.worktreeConfig 2025-09-09T14:03:18.5017802Z ##[group]Checking out the ref 2025-09-09T14:03:18.5021244Z [command]/usr/bin/git checkout --progress --force -B main refs/remotes/origin/main 2025-09-09T14:03:18.6625722Z Switched to a new branch 'main' 2025-09-09T14:03:18.6628296Z branch 'main' set up to track 'origin/main'. 2025-09-09T14:03:18.6640539Z ##[endgroup] 2025-09-09T14:03:18.6640967Z ##[group]Setting up auth for fetching submodules 2025-09-09T14:03:18.6646612Z [command]/usr/bin/git config --global http.https://github.com/.extraheader AUTHORIZATION: basic *** 2025-09-09T14:03:18.6700658Z [command]/usr/bin/git config --global --unset-all url.https://github.com/.insteadOf 2025-09-09T14:03:18.6745796Z [command]/usr/bin/git config --global --add url.https://github.com/.insteadOf git@github.com: 2025-09-09T14:03:18.6786213Z [command]/usr/bin/git config --global --add url.https://github.com/.insteadOf org-21003710@github.com: 2025-09-09T14:03:18.6822757Z ##[endgroup] 2025-09-09T14:03:18.6823183Z ##[group]Fetching submodules 2025-09-09T14:03:18.6825990Z [command]/usr/bin/git submodule sync --recursive 2025-09-09T14:03:18.7247000Z [command]/usr/bin/git -c protocol.version=2 submodule update --init --force --depth=1 --recursive 2025-09-09T14:03:18.7654928Z [command]/usr/bin/git submodule foreach --recursive git config --local gc.auto 0 2025-09-09T14:03:18.8065296Z ##[endgroup] 2025-09-09T14:03:18.8065750Z ##[group]Persisting credentials for submodules 2025-09-09T14:03:18.8070445Z [command]/usr/bin/git submodule foreach --recursive sh -c "git config --local --name-only --get-regexp 'url\.https\:\/\/github\.com\/\.insteadOf' && git config --local --unset-all 'url.https://github.com/.insteadOf' || :" 2025-09-09T14:03:18.8469936Z [command]/usr/bin/git submodule foreach --recursive sh -c "git config --local 'http.https://github.com/.extraheader' 'AUTHORIZATION: basic ***' && git config --local --show-origin --name-only --get-regexp remote.origin.url" 2025-09-09T14:03:18.8876217Z [command]/usr/bin/git submodule foreach --recursive git config --local --add 'url.https://github.com/.insteadOf' 'git@github.com:' 2025-09-09T14:03:18.9289474Z [command]/usr/bin/git submodule foreach --recursive git config --local --add 'url.https://github.com/.insteadOf' 'org-21003710@github.com:' 2025-09-09T14:03:18.9702939Z ##[endgroup] 2025-09-09T14:03:18.9750396Z [command]/usr/bin/git log -1 --format=%H 2025-09-09T14:03:18.9784233Z e502b6d9079a2a411c68046e8a7694b851c5df33 2025-09-09T14:03:18.9996211Z Prepare all required actions 2025-09-09T14:03:18.9996646Z Getting action download info 2025-09-09T14:03:19.1403726Z Download action repository 'pytorch/test-infra@main' (SHA:e502b6d9079a2a411c68046e8a7694b851c5df33) 2025-09-09T14:03:21.2287166Z Getting action download info 2025-09-09T14:03:21.3512830Z Download action repository 'nick-fields/retry@3e91a01664abd3c5cd539100d10d33b9c5b68482' (SHA:3e91a01664abd3c5cd539100d10d33b9c5b68482) 2025-09-09T14:03:21.5207438Z ##[group]Run ./test-infra/.github/actions/setup-linux 2025-09-09T14:03:21.5207772Z env: 2025-09-09T14:03:21.5208023Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:21.5208361Z REPOSITORY: pytorch/ao 2025-09-09T14:03:21.5208622Z PR_NUMBER: 2963 2025-09-09T14:03:21.5210444Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:21.5212248Z ##[endgroup] 2025-09-09T14:03:21.5297982Z ##[group]Run set -euo pipefail 2025-09-09T14:03:21.5298307Z set -euo pipefail 2025-09-09T14:03:21.5298591Z function get_ec2_metadata() { 2025-09-09T14:03:21.5298943Z  # Pulled from instance metadata endpoint for EC2 2025-09-09T14:03:21.5299547Z  # see https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/instancedata-data-retrieval.html 2025-09-09T14:03:21.5300256Z  category=$1 2025-09-09T14:03:21.5301073Z  curl -H "X-aws-ec2-metadata-token: $(curl -s -X PUT "http://169.254.169.254/latest/api/token" -H "X-aws-ec2-metadata-token-ttl-seconds: 30")" -fsSL "http://169.254.169.254/latest/meta-data/${category}" 2025-09-09T14:03:21.5301894Z } 2025-09-09T14:03:21.5302146Z echo "ami-id: $(get_ec2_metadata ami-id)" 2025-09-09T14:03:21.5302539Z echo "instance-id: $(get_ec2_metadata instance-id)" 2025-09-09T14:03:21.5302975Z echo "instance-type: $(get_ec2_metadata instance-type)" 2025-09-09T14:03:21.5303366Z echo "system info $(uname -a)" 2025-09-09T14:03:21.5312805Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:03:21.5313151Z env: 2025-09-09T14:03:21.5313396Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:21.5313725Z REPOSITORY: pytorch/ao 2025-09-09T14:03:21.5313963Z PR_NUMBER: 2963 2025-09-09T14:03:21.5315771Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:21.5317569Z ##[endgroup] 2025-09-09T14:03:21.5485650Z ami-id: ami-05ffe3c48a9991133 2025-09-09T14:03:21.5636948Z instance-id: i-00b63bd9592e830d3 2025-09-09T14:03:21.5762532Z instance-type: g5.12xlarge 2025-09-09T14:03:21.5779214Z system info Linux ip-10-0-56-61.ec2.internal 6.1.141-155.222.amzn2023.x86_64 #1 SMP PREEMPT_DYNAMIC Tue Jun 17 10:29:47 UTC 2025 x86_64 x86_64 x86_64 GNU/Linux 2025-09-09T14:03:21.5837741Z ##[group]Run echo "IN_CONTAINER_RUNNER=$(if [ -f /.inarc ] || [ -f /.incontainer ]; then echo true ; else echo false; fi)" >> "$GITHUB_OUTPUT" 2025-09-09T14:03:21.5838604Z echo "IN_CONTAINER_RUNNER=$(if [ -f /.inarc ] || [ -f /.incontainer ]; then echo true ; else echo false; fi)" >> "$GITHUB_OUTPUT" 2025-09-09T14:03:21.5848238Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:03:21.5848601Z env: 2025-09-09T14:03:21.5848859Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:21.5849208Z REPOSITORY: pytorch/ao 2025-09-09T14:03:21.5849454Z PR_NUMBER: 2963 2025-09-09T14:03:21.5851239Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:21.5853066Z ##[endgroup] 2025-09-09T14:03:21.6125519Z ##[group]Run if systemctl is-active --quiet docker; then 2025-09-09T14:03:21.6125955Z if systemctl is-active --quiet docker; then 2025-09-09T14:03:21.6126338Z  echo "Docker daemon is running..."; 2025-09-09T14:03:21.6126653Z else 2025-09-09T14:03:21.6126986Z  echo "Starting docker deamon..." && sudo systemctl start docker; 2025-09-09T14:03:21.6127375Z fi 2025-09-09T14:03:21.6137075Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:03:21.6137424Z env: 2025-09-09T14:03:21.6137680Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:21.6138017Z REPOSITORY: pytorch/ao 2025-09-09T14:03:21.6138270Z PR_NUMBER: 2963 2025-09-09T14:03:21.6140037Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:21.6142043Z ##[endgroup] 2025-09-09T14:03:21.6240313Z Docker daemon is running... 2025-09-09T14:03:21.6275864Z ##[group]Run AWS_ACCOUNT_ID=$(aws sts get-caller-identity|grep Account|cut -f4 -d\") 2025-09-09T14:03:21.6276473Z AWS_ACCOUNT_ID=$(aws sts get-caller-identity|grep Account|cut -f4 -d\") 2025-09-09T14:03:21.6276964Z retry () { "$@" || (sleep 1 && "$@") || (sleep 2 && "$@") } 2025-09-09T14:03:21.6277524Z retry aws ecr get-login-password --region "$AWS_DEFAULT_REGION" | docker login --username AWS \ 2025-09-09T14:03:21.6278188Z  --password-stdin "$AWS_ACCOUNT_ID.dkr.ecr.$AWS_DEFAULT_REGION.amazonaws.com" 2025-09-09T14:03:21.6288072Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:03:21.6288424Z env: 2025-09-09T14:03:21.6288682Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:21.6289029Z REPOSITORY: pytorch/ao 2025-09-09T14:03:21.6289321Z PR_NUMBER: 2963 2025-09-09T14:03:21.6291086Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:21.6292893Z AWS_RETRY_MODE: standard 2025-09-09T14:03:21.6293180Z AWS_MAX_ATTEMPTS: 5 2025-09-09T14:03:21.6293441Z AWS_DEFAULT_REGION: us-east-1 2025-09-09T14:03:21.6293703Z ##[endgroup] 2025-09-09T14:03:22.7142393Z WARNING! Your password will be stored unencrypted in /home/ec2-user/.docker/config.json. 2025-09-09T14:03:22.7142940Z Configure a credential helper to remove this warning. See 2025-09-09T14:03:22.7143790Z https://docs.docker.com/engine/reference/commandline/login/#credentials-store 2025-09-09T14:03:22.7144156Z 2025-09-09T14:03:22.7144418Z Login Succeeded 2025-09-09T14:03:22.7208110Z ##[group]Run env | grep '^GITHUB' >> "${RUNNER_TEMP}/github_env_${GITHUB_RUN_ID}" 2025-09-09T14:03:22.7208645Z env | grep '^GITHUB' >> "${RUNNER_TEMP}/github_env_${GITHUB_RUN_ID}" 2025-09-09T14:03:22.7209118Z env | grep '^CI' >> "${RUNNER_TEMP}/github_env_${GITHUB_RUN_ID}" 2025-09-09T14:03:22.7219275Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:03:22.7219741Z env: 2025-09-09T14:03:22.7220030Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:22.7220372Z REPOSITORY: pytorch/ao 2025-09-09T14:03:22.7220621Z PR_NUMBER: 2963 2025-09-09T14:03:22.7222419Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:22.7224233Z ##[endgroup] 2025-09-09T14:03:22.7332984Z ##[group]Run RUNNER_ARTIFACT_DIR="${RUNNER_TEMP}/artifacts" 2025-09-09T14:03:22.7333437Z RUNNER_ARTIFACT_DIR="${RUNNER_TEMP}/artifacts" 2025-09-09T14:03:22.7333804Z sudo rm -rf "${RUNNER_ARTIFACT_DIR}" 2025-09-09T14:03:22.7334137Z mkdir -p "${RUNNER_ARTIFACT_DIR}" 2025-09-09T14:03:22.7334593Z echo "RUNNER_ARTIFACT_DIR=${RUNNER_ARTIFACT_DIR}" >> "${GITHUB_ENV}" 2025-09-09T14:03:22.7335176Z  2025-09-09T14:03:22.7335479Z RUNNER_TEST_RESULTS_DIR="${RUNNER_TEMP}/test-results" 2025-09-09T14:03:22.7335969Z sudo rm -rf "${RUNNER_TEST_RESULTS_DIR}" 2025-09-09T14:03:22.7336320Z mkdir -p "${RUNNER_TEST_RESULTS_DIR}" 2025-09-09T14:03:22.7336781Z echo "RUNNER_TEST_RESULTS_DIR=${RUNNER_TEST_RESULTS_DIR}" >> "${GITHUB_ENV}" 2025-09-09T14:03:22.7337193Z  2025-09-09T14:03:22.7337430Z RUNNER_DOCS_DIR="${RUNNER_TEMP}/docs" 2025-09-09T14:03:22.7337747Z sudo rm -rf "${RUNNER_DOCS_DIR}" 2025-09-09T14:03:22.7338050Z mkdir -p "${RUNNER_DOCS_DIR}" 2025-09-09T14:03:22.7338417Z echo "RUNNER_DOCS_DIR=${RUNNER_DOCS_DIR}" >> "${GITHUB_ENV}" 2025-09-09T14:03:22.7348994Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:03:22.7349324Z env: 2025-09-09T14:03:22.7349562Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:22.7349889Z REPOSITORY: pytorch/ao 2025-09-09T14:03:22.7350131Z PR_NUMBER: 2963 2025-09-09T14:03:22.7351894Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:22.7353686Z ##[endgroup] 2025-09-09T14:03:23.2480566Z ##[group]Run needs=0 2025-09-09T14:03:23.2480817Z needs=0 2025-09-09T14:03:23.2481164Z if lspci -v | grep -e 'controller.*NVIDIA' >/dev/null 2>/dev/null; then 2025-09-09T14:03:23.2481565Z  needs=1 2025-09-09T14:03:23.2481786Z fi 2025-09-09T14:03:23.2482024Z echo "does=${needs}" >> $GITHUB_OUTPUT 2025-09-09T14:03:23.2491158Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:03:23.2491517Z env: 2025-09-09T14:03:23.2491772Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:23.2492101Z REPOSITORY: pytorch/ao 2025-09-09T14:03:23.2492358Z PR_NUMBER: 2963 2025-09-09T14:03:23.2494340Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:23.2496703Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:03:23.2497252Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:03:23.2498006Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:03:23.2498361Z ##[endgroup] 2025-09-09T14:03:23.2853805Z ##[group]Run pytorch/test-infra/.github/actions/setup-nvidia@main 2025-09-09T14:03:23.2854181Z with: 2025-09-09T14:03:23.2854408Z driver-version: 580.65.06 2025-09-09T14:03:23.2854686Z env: 2025-09-09T14:03:23.2854958Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:23.2855294Z REPOSITORY: pytorch/ao 2025-09-09T14:03:23.2855546Z PR_NUMBER: 2963 2025-09-09T14:03:23.2857445Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:23.2859565Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:03:23.2860116Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:03:23.2860620Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:03:23.2860984Z ##[endgroup] 2025-09-09T14:03:23.2904668Z ##[group]Run nick-fields/retry@3e91a01664abd3c5cd539100d10d33b9c5b68482 2025-09-09T14:03:23.2905046Z with: 2025-09-09T14:03:23.2905249Z timeout_minutes: 10 2025-09-09T14:03:23.2905476Z max_attempts: 3 2025-09-09T14:03:23.2929707Z command: # Is it disgusting to have a full shell script here in this github action? Sure # But is it the best way to make it so that this action relies on nothing else? Absolutely set -eou pipefail DISTRIBUTION=$(. /etc/os-release;echo $ID$VERSION_ID) DRIVER_FN="NVIDIA-Linux-x86_64-${DRIVER_VERSION}.run" install_nvidia_docker2_amzn2() { ( set -x # Needed for yum-config-manager sudo yum install -y yum-utils if [[ "${DISTRIBUTION}" == "amzn2023" ]] ; then YUM_REPO_URL="https://nvidia.github.io/libnvidia-container/stable/rpm/nvidia-container-toolkit.repo" else # Amazon Linux 2 YUM_REPO_URL="https://nvidia.github.io/nvidia-docker/${DISTRIBUTION}/nvidia-docker.repo" fi sudo yum-config-manager --add-repo "${YUM_REPO_URL}" sudo yum install -y \ nvidia-container-toolkit-1.17.8 \ libnvidia-container-tools-1.17.8 \ libnvidia-container1-1.17.8 \ nvidia-container-toolkit-base-1.17.8 sudo systemctl restart docker ) } install_nvidia_docker2_ubuntu20() { ( set -x # Install nvidia-driver package if not installed status="$(dpkg-query -W --showformat='${db:Status-Status}' nvidia-docker2 2>&1)" if [ ! $? = 0 ] || [ ! "$status" = installed ]; then sudo apt-get install -y nvidia-container-toolkit-1.17.8 sudo systemctl restart docker fi ) } pre_install_nvidia_driver_amzn2() { ( # Purge any nvidia driver installed from RHEL repo sudo yum remove -y nvidia-driver-latest-dkms ) } install_nvidia_driver_common() { ( # Try to gather more information about the runner and its existing NVIDIA driver if any echo "Before installing NVIDIA driver" lspci lsmod modinfo nvidia || true HAS_NVIDIA_DRIVER=0 # Check if NVIDIA driver has already been installed if [ -x "$(command -v nvidia-smi)" ]; then set +e # The driver exists, check its version next. Also check only the first GPU if there are more than one of them # so that the same driver version is not print over multiple lines INSTALLED_DRIVER_VERSION=$(nvidia-smi --query-gpu=driver_version --format=csv,noheader --id=0) NVIDIA_SMI_STATUS=$? if [ "$NVIDIA_SMI_STATUS" -ne 0 ] && [ "$NVIDIA_SMI_STATUS" -ne 14 ]; then echo "Failed to get NVIDIA driver version ($INSTALLED_DRIVER_VERSION). Continuing" elif [ "$INSTALLED_DRIVER_VERSION" != "$DRIVER_VERSION" ]; then echo "NVIDIA driver ($INSTALLED_DRIVER_VERSION) has been installed, but we expect to have $DRIVER_VERSION instead. Continuing" # Turn off persistent mode so that the installation script can unload the kernel module sudo killall nvidia-persistenced || true else HAS_NVIDIA_DRIVER=1 echo "NVIDIA driver ($INSTALLED_DRIVER_VERSION) has already been installed. Skipping NVIDIA driver installation" fi set -e fi if [ "$HAS_NVIDIA_DRIVER" -eq 0 ]; then # CAUTION: this may need to be updated in future if [ "${DISTRIBUTION}" != ubuntu20.04 ]; then sudo yum groupinstall -y "Development Tools" # ensure our kernel install is the same as our underlying kernel, # groupinstall "Development Tools" has a habit of mismatching kernel headers sudo yum install -y "kernel-devel-uname-r == $(uname -r)" sudo modprobe backlight fi sudo curl -fsL -o /tmp/nvidia_driver "https://s3.amazonaws.com/ossci-linux/nvidia_driver/$DRIVER_FN" set +e sudo /bin/bash /tmp/nvidia_driver -s --no-drm NVIDIA_INSTALLATION_STATUS=$? RESET_GPU=0 if [ "$NVIDIA_INSTALLATION_STATUS" -ne 0 ]; then sudo cat /var/log/nvidia-installer.log # Fail to install NVIDIA driver, try to reset the GPU RESET_GPU=1 elif [ -x "$(command -v nvidia-smi)" ]; then # Check again if nvidia-smi works even if the driver installation completes successfully INSTALLED_DRIVER_VERSION=$(nvidia-smi --query-gpu=driver_version --format=csv,noheader --id=0) NVIDIA_SMI_STATUS=$? if [ "$NVIDIA_SMI_STATUS" -ne 0 ] && [ "$NVIDIA_SMI_STATUS" -ne 14 ]; then RESET_GPU=1 fi fi if [ "$RESET_GPU" -eq 1 ]; then NVIDIA_DEVICES=$(lspci -D | grep -i NVIDIA | cut -d' ' -f1) # The GPU can get stuck in a failure state if somehow the test crashs the GPU microcode. When this # happens, we'll try to reset all NVIDIA devices https://github.com/pytorch/pytorch/issues/88388 for PCI_ID in $NVIDIA_DEVICES; do DEVICE_ENABLED=$(cat /sys/bus/pci/devices/$PCI_ID/enable) echo "Reseting $PCI_ID (enabled state: $DEVICE_ENABLED)" # This requires sudo permission of course echo "1" | sudo tee /sys/bus/pci/devices/$PCI_ID/reset sleep 1 done fi sudo rm -fv /tmp/nvidia_driver set -e fi ) } post_install_nvidia_driver_common() { ( sudo modprobe nvidia || true echo "After installing NVIDIA driver" lspci lsmod modinfo nvidia || true ( set +e nvidia-smi # NB: Annoyingly, nvidia-smi command returns successfully with return code 0 even in # the case where the driver has already crashed as it still can get the driver version # and some basic information like the bus ID. However, the rest of the information # would be missing (ERR!), for example: # # +-----------------------------------------------------------------------------+ # | NVIDIA-SMI 525.89.02 Driver Version: 525.89.02 CUDA Version: 12.0 | # |-------------------------------+----------------------+----------------------+ # | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC | # | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. | # | | | MIG M. | # |===============================+======================+======================| # | 0 ERR! Off | 00000000:00:1E.0 Off | ERR! | # |ERR! ERR! ERR! ERR! / ERR! | 4184MiB / 23028MiB | ERR! Default | # | | | ERR! | # +-------------------------------+----------------------+----------------------+ # # +-----------------------------------------------------------------------------+ # | Processes: | # | GPU GI CI PID Type Process name GPU Memory | # | ID ID Usage | # |=============================================================================| # +-----------------------------------------------------------------------------+ # # This should be reported as a failure instead as it will guarantee to fail when # Docker tries to run with --gpus all # # So, the correct check here is to query one of the missing piece of info like # GPU name, so that the command can fail accordingly nvidia-smi --query-gpu=gpu_name --format=csv,noheader --id=0 NVIDIA_SMI_STATUS=$? # Allowable exit statuses for nvidia-smi, see: https://github.com/NVIDIA/gpu-operator/issues/285 if [ "$NVIDIA_SMI_STATUS" -eq 0 ] || [ "$NVIDIA_SMI_STATUS" -eq 14 ]; then echo "INFO: Ignoring allowed status ${NVIDIA_SMI_STATUS}" else echo "ERROR: nvidia-smi exited with unresolved status ${NVIDIA_SMI_STATUS}" exit ${NVIDIA_SMI_STATUS} fi set -e ) ) } install_nvidia_driver_amzn2() { ( set -x pre_install_nvidia_driver_amzn2 install_nvidia_driver_common post_install_nvidia_driver_common ) } install_nvidia_driver_ubuntu20() { ( set -x install_nvidia_driver_common post_install_nvidia_driver_common ) } echo "== Installing nvidia driver ${DRIVER_FN} ==" case "${DISTRIBUTION}" in amzn*) install_nvidia_driver_amzn2 ;; ubuntu20.04) install_nvidia_driver_ubuntu20 ;; *) echo "ERROR: Unknown distribution ${DISTRIBUTION}" exit 1 ;; esac # Install container toolkit based on distribution echo "== Installing nvidia container toolkit for ${DISTRIBUTION} ==" case "${DISTRIBUTION}" in amzn*) install_nvidia_docker2_amzn2 ;; ubuntu20.04) install_nvidia_docker2_ubuntu20 ;; *) echo "ERROR: Unknown distribution ${DISTRIBUTION}" exit 1 ;; esac echo "GPU_FLAG=--gpus all -e NVIDIA_DRIVER_CAPABILITIES=all" >> "${GITHUB_ENV}" # Fix https://github.com/NVIDIA/nvidia-docker/issues/1648 on runners with # more than one GPUs. This just needs to be run once. The command fails # on subsequent runs and complains that the mode is already on, but that's # ok sudo nvidia-persistenced || true # This should show persistence mode ON nvidia-smi # check if the container-toolkit is correctly installed and CUDA is available inside a container docker run --rm -t --gpus=all public.ecr.aws/docker/library/python:3.13 nvidia-smi 2025-09-09T14:03:23.2954219Z retry_wait_seconds: 10 2025-09-09T14:03:23.2954487Z polling_interval_seconds: 1 2025-09-09T14:03:23.2954786Z warning_on_retry: true 2025-09-09T14:03:23.2955067Z continue_on_error: false 2025-09-09T14:03:23.2955320Z env: 2025-09-09T14:03:23.2955571Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:23.2955914Z REPOSITORY: pytorch/ao 2025-09-09T14:03:23.2956165Z PR_NUMBER: 2963 2025-09-09T14:03:23.2957943Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:23.2959886Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:03:23.2960441Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:03:23.2960951Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:03:23.2961331Z DRIVER_VERSION: 580.65.06 2025-09-09T14:03:23.2961581Z ##[endgroup] 2025-09-09T14:03:23.3841932Z == Installing nvidia driver NVIDIA-Linux-x86_64-580.65.06.run == 2025-09-09T14:03:23.3842916Z + pre_install_nvidia_driver_amzn2 2025-09-09T14:03:23.3846562Z + sudo yum remove -y nvidia-driver-latest-dkms 2025-09-09T14:03:23.7031879Z No match for argument: nvidia-driver-latest-dkms 2025-09-09T14:03:23.7032771Z No packages marked for removal. 2025-09-09T14:03:23.7100666Z Dependencies resolved. 2025-09-09T14:03:23.7110332Z Nothing to do. 2025-09-09T14:03:23.7110826Z Complete! 2025-09-09T14:03:23.7455413Z + install_nvidia_driver_common 2025-09-09T14:03:23.7458835Z + echo 'Before installing NVIDIA driver' 2025-09-09T14:03:23.7459132Z + lspci 2025-09-09T14:03:23.7460566Z Before installing NVIDIA driver 2025-09-09T14:03:23.7593687Z 00:00.0 Host bridge: Intel Corporation 440FX - 82441FX PMC [Natoma] 2025-09-09T14:03:23.7594172Z 00:01.0 ISA bridge: Intel Corporation 82371SB PIIX3 ISA [Natoma/Triton II] 2025-09-09T14:03:23.7594919Z 00:01.3 Non-VGA unclassified device: Intel Corporation 82371AB/EB/MB PIIX4 ACPI (rev 08) 2025-09-09T14:03:23.7595946Z 00:03.0 VGA compatible controller: Amazon.com, Inc. Device 1111 2025-09-09T14:03:23.7596890Z 00:04.0 Non-Volatile memory controller: Amazon.com, Inc. NVMe EBS Controller 2025-09-09T14:03:23.7598198Z 00:05.0 Ethernet controller: Amazon.com, Inc. Elastic Network Adapter (ENA) 2025-09-09T14:03:23.7599131Z 00:1b.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:03:23.7599965Z 00:1c.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:03:23.7600782Z 00:1d.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:03:23.7601581Z 00:1e.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:03:23.7602492Z 00:1f.0 Non-Volatile memory controller: Amazon.com, Inc. NVMe SSD Controller 2025-09-09T14:03:23.7603256Z + lsmod 2025-09-09T14:03:23.7650782Z Module Size Used by 2025-09-09T14:03:23.7651063Z xt_conntrack 16384 1 2025-09-09T14:03:23.7651349Z nft_chain_nat 16384 3 2025-09-09T14:03:23.7651606Z xt_MASQUERADE 20480 1 2025-09-09T14:03:23.7651900Z nf_nat 57344 2 nft_chain_nat,xt_MASQUERADE 2025-09-09T14:03:23.7652230Z nf_conntrack_netlink 57344 0 2025-09-09T14:03:23.7652615Z nf_conntrack 184320 4 xt_conntrack,nf_nat,nf_conntrack_netlink,xt_MASQUERADE 2025-09-09T14:03:23.7653270Z nf_defrag_ipv6 24576 1 nf_conntrack 2025-09-09T14:03:23.7653579Z nf_defrag_ipv4 16384 1 nf_conntrack 2025-09-09T14:03:23.7653864Z xfrm_user 57344 1 2025-09-09T14:03:23.7654127Z xfrm_algo 16384 1 xfrm_user 2025-09-09T14:03:23.7654402Z xt_addrtype 16384 2 2025-09-09T14:03:23.7654662Z nft_compat 20480 4 2025-09-09T14:03:23.7654951Z nf_tables 311296 57 nft_compat,nft_chain_nat 2025-09-09T14:03:23.7655359Z nfnetlink 20480 4 nft_compat,nf_conntrack_netlink,nf_tables 2025-09-09T14:03:23.7655725Z br_netfilter 36864 0 2025-09-09T14:03:23.7656064Z bridge 323584 1 br_netfilter 2025-09-09T14:03:23.7656357Z stp 16384 1 bridge 2025-09-09T14:03:23.7656640Z llc 16384 2 bridge,stp 2025-09-09T14:03:23.7656921Z overlay 167936 0 2025-09-09T14:03:23.7657157Z tls 139264 0 2025-09-09T14:03:23.7657402Z nls_ascii 16384 1 2025-09-09T14:03:23.7657644Z nls_cp437 20480 1 2025-09-09T14:03:23.7657886Z vfat 24576 1 2025-09-09T14:03:23.7658124Z fat 86016 1 vfat 2025-09-09T14:03:23.7658381Z ena 180224 0 2025-09-09T14:03:23.7658610Z i8042 45056 0 2025-09-09T14:03:23.7658848Z sunrpc 700416 1 2025-09-09T14:03:23.7659097Z serio 28672 3 i8042 2025-09-09T14:03:23.7659357Z button 24576 0 2025-09-09T14:03:23.7659607Z ghash_clmulni_intel 16384 0 2025-09-09T14:03:23.7659865Z sch_fq_codel 20480 33 2025-09-09T14:03:23.7660116Z dm_mod 188416 0 2025-09-09T14:03:23.7660349Z fuse 184320 1 2025-09-09T14:03:23.7660731Z loop 36864 0 2025-09-09T14:03:23.7660969Z configfs 57344 1 2025-09-09T14:03:23.7661212Z dmi_sysfs 20480 0 2025-09-09T14:03:23.7661453Z crc32_pclmul 16384 0 2025-09-09T14:03:23.7661709Z crc32c_intel 24576 0 2025-09-09T14:03:23.7661951Z efivarfs 24576 1 2025-09-09T14:03:23.7662194Z + modinfo nvidia 2025-09-09T14:03:23.7674316Z filename: /lib/modules/6.1.141-155.222.amzn2023.x86_64/kernel/drivers/video/nvidia.ko 2025-09-09T14:03:23.7674773Z import_ns: DMA_BUF 2025-09-09T14:03:23.7675024Z alias: char-major-195-* 2025-09-09T14:03:23.7675281Z version: 570.133.07 2025-09-09T14:03:23.7675522Z supported: external 2025-09-09T14:03:23.7675764Z license: Dual MIT/GPL 2025-09-09T14:03:23.7676049Z firmware: nvidia/570.133.07/gsp_tu10x.bin 2025-09-09T14:03:23.7676379Z firmware: nvidia/570.133.07/gsp_ga10x.bin 2025-09-09T14:03:23.7676693Z srcversion: 49515739FD8F721A3F2F714 2025-09-09T14:03:23.7677013Z alias: pci:v000010DEd*sv*sd*bc06sc80i00* 2025-09-09T14:03:23.7677337Z alias: pci:v000010DEd*sv*sd*bc03sc02i00* 2025-09-09T14:03:23.7677662Z alias: pci:v000010DEd*sv*sd*bc03sc00i00* 2025-09-09T14:03:23.7677963Z depends: i2c-core,drm 2025-09-09T14:03:23.7678215Z retpoline: Y 2025-09-09T14:03:23.7678427Z name: nvidia 2025-09-09T14:03:23.7678779Z vermagic: 6.1.141-155.222.amzn2023.x86_64 SMP preempt mod_unload modversions 2025-09-09T14:03:23.7679240Z parm: NvSwitchRegDwords:NvSwitch regkey (charp) 2025-09-09T14:03:23.7679676Z parm: NvSwitchBlacklist:NvSwitchBlacklist=uuid[,uuid...] (charp) 2025-09-09T14:03:23.7680088Z parm: NVreg_ResmanDebugLevel:int 2025-09-09T14:03:23.7680383Z parm: NVreg_RmLogonRC:int 2025-09-09T14:03:23.7680678Z parm: NVreg_ModifyDeviceFiles:int 2025-09-09T14:03:23.7680977Z parm: NVreg_DeviceFileUID:int 2025-09-09T14:03:23.7681283Z parm: NVreg_DeviceFileGID:int 2025-09-09T14:03:23.7681573Z parm: NVreg_DeviceFileMode:int 2025-09-09T14:03:23.7681928Z parm: NVreg_InitializeSystemMemoryAllocations:int 2025-09-09T14:03:23.7682422Z parm: NVreg_UsePageAttributeTable:int 2025-09-09T14:03:23.7682743Z parm: NVreg_EnablePCIeGen3:int 2025-09-09T14:03:23.7683038Z parm: NVreg_EnableMSI:int 2025-09-09T14:03:23.7683331Z parm: NVreg_EnableStreamMemOPs:int 2025-09-09T14:03:23.7683684Z parm: NVreg_RestrictProfilingToAdminUsers:int 2025-09-09T14:03:23.7684318Z parm: NVreg_PreserveVideoMemoryAllocations:int 2025-09-09T14:03:23.7684774Z parm: NVreg_EnableS0ixPowerManagement:int 2025-09-09T14:03:23.7685333Z parm: NVreg_S0ixPowerManagementVideoMemoryThreshold:int 2025-09-09T14:03:23.7685773Z parm: NVreg_DynamicPowerManagement:int 2025-09-09T14:03:23.7686381Z parm: NVreg_DynamicPowerManagementVideoMemoryThreshold:int 2025-09-09T14:03:23.7686884Z parm: NVreg_EnableGpuFirmware:int 2025-09-09T14:03:23.7687254Z parm: NVreg_EnableGpuFirmwareLogs:int 2025-09-09T14:03:23.7700300Z parm: NVreg_OpenRmEnableUnsupportedGpus:int 2025-09-09T14:03:23.7700699Z parm: NVreg_EnableUserNUMAManagement:int 2025-09-09T14:03:23.7701037Z parm: NVreg_MemoryPoolSize:int 2025-09-09T14:03:23.7701348Z parm: NVreg_KMallocHeapMaxSize:int 2025-09-09T14:03:23.7701675Z parm: NVreg_VMallocHeapMaxSize:int 2025-09-09T14:03:23.7701987Z parm: NVreg_IgnoreMMIOCheck:int 2025-09-09T14:03:23.7702293Z parm: NVreg_NvLinkDisable:int 2025-09-09T14:03:23.7702631Z parm: NVreg_EnablePCIERelaxedOrderingMode:int 2025-09-09T14:03:23.7702993Z parm: NVreg_RegisterPCIDriver:int 2025-09-09T14:03:23.7703313Z parm: NVreg_EnableResizableBar:int 2025-09-09T14:03:23.7703646Z parm: NVreg_EnableDbgBreakpoint:int 2025-09-09T14:03:23.7703993Z parm: NVreg_EnableNonblockingOpen:int 2025-09-09T14:03:23.7704619Z parm: NVreg_RegistryDwords:charp 2025-09-09T14:03:23.7704960Z parm: NVreg_RegistryDwordsPerDevice:charp 2025-09-09T14:03:23.7705278Z parm: NVreg_RmMsg:charp 2025-09-09T14:03:23.7705568Z parm: NVreg_GpuBlacklist:charp 2025-09-09T14:03:23.7705879Z parm: NVreg_TemporaryFilePath:charp 2025-09-09T14:03:23.7706198Z parm: NVreg_ExcludedGpus:charp 2025-09-09T14:03:23.7706505Z parm: NVreg_DmaRemapPeerMmio:int 2025-09-09T14:03:23.7706829Z parm: NVreg_RmNvlinkBandwidth:charp 2025-09-09T14:03:23.7707178Z parm: NVreg_RmNvlinkBandwidthLinkCount:int 2025-09-09T14:03:23.7707518Z parm: NVreg_ImexChannelCount:int 2025-09-09T14:03:23.7707837Z parm: NVreg_CreateImexChannel0:int 2025-09-09T14:03:23.7708170Z parm: NVreg_GrdmaPciTopoCheckOverride:int 2025-09-09T14:03:23.7708507Z parm: rm_firmware_active:charp 2025-09-09T14:03:23.7708785Z + HAS_NVIDIA_DRIVER=0 2025-09-09T14:03:23.7709040Z ++ command -v nvidia-smi 2025-09-09T14:03:23.7709291Z + '[' -x /usr/bin/nvidia-smi ']' 2025-09-09T14:03:23.7709552Z + set +e 2025-09-09T14:03:23.7709852Z ++ nvidia-smi --query-gpu=driver_version --format=csv,noheader --id=0 2025-09-09T14:03:27.1789289Z + INSTALLED_DRIVER_VERSION=570.133.07 2025-09-09T14:03:27.1789620Z + NVIDIA_SMI_STATUS=0 2025-09-09T14:03:27.1789847Z + '[' 0 -ne 0 ']' 2025-09-09T14:03:27.1790061Z + '[' 570.133.07 '!=' 580.65.06 ']' 2025-09-09T14:03:27.1790531Z + echo 'NVIDIA driver (570.133.07) has been installed, but we expect to have 580.65.06 instead. Continuing' 2025-09-09T14:03:27.1791037Z + sudo killall nvidia-persistenced 2025-09-09T14:03:27.1791496Z NVIDIA driver (570.133.07) has been installed, but we expect to have 580.65.06 instead. Continuing 2025-09-09T14:03:27.2891793Z nvidia-persistenced: no process found 2025-09-09T14:03:27.2916772Z + true 2025-09-09T14:03:27.2917084Z + set -e 2025-09-09T14:03:27.2917296Z + '[' 0 -eq 0 ']' 2025-09-09T14:03:27.2917540Z + '[' amzn2023 '!=' ubuntu20.04 ']' 2025-09-09T14:03:27.2917858Z + sudo yum groupinstall -y 'Development Tools' 2025-09-09T14:03:27.8161797Z Last metadata expiration check: 0:06:18 ago on Tue Sep 9 13:57:09 2025. 2025-09-09T14:03:27.8499381Z No match for group package "system-rpm-config" 2025-09-09T14:03:27.8513062Z No match for group package "rcs" 2025-09-09T14:03:27.8530565Z No match for group package "pkgconfig" 2025-09-09T14:03:27.8982867Z Dependencies resolved. 2025-09-09T14:03:27.9201205Z ================================================================================ 2025-09-09T14:03:27.9201651Z Package Architecture Version Repository Size 2025-09-09T14:03:27.9202065Z ================================================================================ 2025-09-09T14:03:27.9202364Z Installing Groups: 2025-09-09T14:03:27.9202673Z Development Tools 2025-09-09T14:03:27.9202938Z 2025-09-09T14:03:27.9203035Z Transaction Summary 2025-09-09T14:03:27.9203285Z ================================================================================ 2025-09-09T14:03:27.9203496Z 2025-09-09T14:03:28.1246403Z ================================================================================ 2025-09-09T14:03:28.1246753Z WARNING: 2025-09-09T14:03:28.1247005Z A newer release of "Amazon Linux" is available. 2025-09-09T14:03:28.1247230Z 2025-09-09T14:03:28.1247322Z Available Versions: 2025-09-09T14:03:28.1247476Z 2025-09-09T14:03:28.1247563Z Version 2023.8.20250707: 2025-09-09T14:03:28.1247873Z Run the following command to upgrade to 2023.8.20250707: 2025-09-09T14:03:28.1248119Z 2025-09-09T14:03:28.1248236Z dnf upgrade --releasever=2023.8.20250707 2025-09-09T14:03:28.1248451Z 2025-09-09T14:03:28.1248536Z Release notes: 2025-09-09T14:03:28.1248940Z https://docs.aws.amazon.com/linux/al2023/release-notes/relnotes-2023.8.20250707.html 2025-09-09T14:03:28.1249315Z 2025-09-09T14:03:28.1249407Z Version 2023.8.20250715: 2025-09-09T14:03:28.1249931Z Run the following command to upgrade to 2023.8.20250715: 2025-09-09T14:03:28.1250183Z 2025-09-09T14:03:28.1250299Z dnf upgrade --releasever=2023.8.20250715 2025-09-09T14:03:28.1250504Z 2025-09-09T14:03:28.1250600Z Release notes: 2025-09-09T14:03:28.1250987Z https://docs.aws.amazon.com/linux/al2023/release-notes/relnotes-2023.8.20250715.html 2025-09-09T14:03:28.1251354Z 2025-09-09T14:03:28.1251445Z Version 2023.8.20250721: 2025-09-09T14:03:28.1251745Z Run the following command to upgrade to 2023.8.20250721: 2025-09-09T14:03:28.1251995Z 2025-09-09T14:03:28.1252122Z dnf upgrade --releasever=2023.8.20250721 2025-09-09T14:03:28.1252328Z 2025-09-09T14:03:28.1252425Z Release notes: 2025-09-09T14:03:28.1252808Z https://docs.aws.amazon.com/linux/al2023/release-notes/relnotes-2023.8.20250721.html 2025-09-09T14:03:28.1253175Z 2025-09-09T14:03:28.1253266Z Version 2023.8.20250808: 2025-09-09T14:03:28.1253569Z Run the following command to upgrade to 2023.8.20250808: 2025-09-09T14:03:28.1253828Z 2025-09-09T14:03:28.1253944Z dnf upgrade --releasever=2023.8.20250808 2025-09-09T14:03:28.1254150Z 2025-09-09T14:03:28.1254244Z Release notes: 2025-09-09T14:03:28.1254637Z https://docs.aws.amazon.com/linux/al2023/release-notes/relnotes-2023.8.20250808.html 2025-09-09T14:03:28.1255007Z 2025-09-09T14:03:28.1255097Z Version 2023.8.20250818: 2025-09-09T14:03:28.1255398Z Run the following command to upgrade to 2023.8.20250818: 2025-09-09T14:03:28.1255653Z 2025-09-09T14:03:28.1255768Z dnf upgrade --releasever=2023.8.20250818 2025-09-09T14:03:28.1256040Z 2025-09-09T14:03:28.1256134Z Release notes: 2025-09-09T14:03:28.1256521Z https://docs.aws.amazon.com/linux/al2023/release-notes/relnotes-2023.8.20250818.html 2025-09-09T14:03:28.1256878Z 2025-09-09T14:03:28.1256976Z Version 2023.8.20250908: 2025-09-09T14:03:28.1257286Z Run the following command to upgrade to 2023.8.20250908: 2025-09-09T14:03:28.1257547Z 2025-09-09T14:03:28.1257664Z dnf upgrade --releasever=2023.8.20250908 2025-09-09T14:03:28.1257870Z 2025-09-09T14:03:28.1257953Z Release notes: 2025-09-09T14:03:28.1258479Z https://docs.aws.amazon.com/linux/al2023/release-notes/relnotes-2023.8.20250908.html 2025-09-09T14:03:28.1258838Z 2025-09-09T14:03:28.1258956Z ================================================================================ 2025-09-09T14:03:28.1259254Z Complete! 2025-09-09T14:03:28.1685193Z ++ uname -r 2025-09-09T14:03:28.1701012Z + sudo yum install -y 'kernel-devel-uname-r == 6.1.141-155.222.amzn2023.x86_64' 2025-09-09T14:03:28.6219754Z Last metadata expiration check: 0:06:19 ago on Tue Sep 9 13:57:09 2025. 2025-09-09T14:03:28.6454078Z Using '==' operator in reldeps can result in an undefined behavior. It is deprecated and the support will be dropped in future versions. Use '=' operator instead. 2025-09-09T14:03:28.6567995Z Package kernel-devel-6.1.141-155.222.amzn2023.x86_64 is already installed. 2025-09-09T14:03:28.7033645Z Dependencies resolved. 2025-09-09T14:03:28.7259875Z Nothing to do. 2025-09-09T14:03:28.7260243Z Complete! 2025-09-09T14:03:28.7667009Z + sudo modprobe backlight 2025-09-09T14:03:28.9305475Z + sudo curl -fsL -o /tmp/nvidia_driver https://s3.amazonaws.com/ossci-linux/nvidia_driver/NVIDIA-Linux-x86_64-580.65.06.run 2025-09-09T14:03:33.0607180Z + set +e 2025-09-09T14:03:33.0607449Z + sudo /bin/bash /tmp/nvidia_driver -s --no-drm 2025-09-09T14:03:34.4792594Z Verifying archive integrity... OK 2025-09-09T14:03:37.3271252Z Uncompressing NVIDIA Accelerated Graphics Driver for Linux-x86_64 580.65.06.................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................... 2025-09-09T14:03:37.9590630Z 2025-09-09T14:03:37.9591457Z WARNING: The nvidia-drm module will not be installed. As a result, DRM-KMS will not function with this installation of the NVIDIA driver. 2025-09-09T14:03:37.9592022Z 2025-09-09T14:04:00.7297932Z 2025-09-09T14:04:00.7299242Z WARNING: nvidia-installer was forced to guess the X library path '/usr/lib64' and X module path '/usr/lib64/xorg/modules'; these paths were not queryable from the system. If X fails to find the NVIDIA X driver module, please install the `pkg-config` utility and the X.Org SDK/development package for your distribution and reinstall the driver. 2025-09-09T14:04:00.7300461Z 2025-09-09T14:04:00.7319149Z 2025-09-09T14:04:00.7324234Z WARNING: This NVIDIA driver package includes Vulkan components, but no Vulkan ICD loader was detected on this system. The NVIDIA Vulkan ICD will not function without the loader. Most distributions package the Vulkan loader; try installing the "vulkan-loader", "vulkan-icd-loader", or "libvulkan1" package. 2025-09-09T14:04:00.7325317Z 2025-09-09T14:04:13.3077385Z + NVIDIA_INSTALLATION_STATUS=0 2025-09-09T14:04:13.3077696Z + RESET_GPU=0 2025-09-09T14:04:13.3077919Z + '[' 0 -ne 0 ']' 2025-09-09T14:04:13.3080235Z ++ command -v nvidia-smi 2025-09-09T14:04:13.3083244Z + '[' -x /usr/bin/nvidia-smi ']' 2025-09-09T14:04:13.3088982Z ++ nvidia-smi --query-gpu=driver_version --format=csv,noheader --id=0 2025-09-09T14:04:17.1919249Z + INSTALLED_DRIVER_VERSION=580.65.06 2025-09-09T14:04:17.1920076Z + NVIDIA_SMI_STATUS=0 2025-09-09T14:04:17.1920319Z + '[' 0 -ne 0 ']' 2025-09-09T14:04:17.1920527Z + '[' 0 -eq 1 ']' 2025-09-09T14:04:17.1920760Z + sudo rm -fv /tmp/nvidia_driver 2025-09-09T14:04:17.3696101Z removed '/tmp/nvidia_driver' 2025-09-09T14:04:17.3719729Z + set -e 2025-09-09T14:04:17.3724204Z + post_install_nvidia_driver_common 2025-09-09T14:04:17.3727783Z + sudo modprobe nvidia 2025-09-09T14:04:17.5197745Z + echo 'After installing NVIDIA driver' 2025-09-09T14:04:17.5198323Z + lspci 2025-09-09T14:04:17.5198725Z After installing NVIDIA driver 2025-09-09T14:04:17.5332580Z 00:00.0 Host bridge: Intel Corporation 440FX - 82441FX PMC [Natoma] 2025-09-09T14:04:17.5333063Z 00:01.0 ISA bridge: Intel Corporation 82371SB PIIX3 ISA [Natoma/Triton II] 2025-09-09T14:04:17.5333607Z 00:01.3 Non-VGA unclassified device: Intel Corporation 82371AB/EB/MB PIIX4 ACPI (rev 08) 2025-09-09T14:04:17.5334110Z 00:03.0 VGA compatible controller: Amazon.com, Inc. Device 1111 2025-09-09T14:04:17.5334575Z 00:04.0 Non-Volatile memory controller: Amazon.com, Inc. NVMe EBS Controller 2025-09-09T14:04:17.5335085Z 00:05.0 Ethernet controller: Amazon.com, Inc. Elastic Network Adapter (ENA) 2025-09-09T14:04:17.5335558Z 00:1b.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:04:17.5336082Z 00:1c.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:04:17.5336490Z 00:1d.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:04:17.5336891Z 00:1e.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:04:17.5337351Z 00:1f.0 Non-Volatile memory controller: Amazon.com, Inc. NVMe SSD Controller 2025-09-09T14:04:17.5337737Z + lsmod 2025-09-09T14:04:17.5372878Z Module Size Used by 2025-09-09T14:04:17.5373432Z nvidia_uvm 1921024 0 2025-09-09T14:04:17.5373691Z nvidia 14274560 1 nvidia_uvm 2025-09-09T14:04:17.5373971Z drm 602112 1 nvidia 2025-09-09T14:04:17.5374262Z drm_panel_orientation_quirks 32768 1 drm 2025-09-09T14:04:17.5374574Z backlight 24576 1 drm 2025-09-09T14:04:17.5374849Z i2c_core 110592 2 nvidia,drm 2025-09-09T14:04:17.5375128Z xt_conntrack 16384 1 2025-09-09T14:04:17.5375380Z nft_chain_nat 16384 3 2025-09-09T14:04:17.5375626Z xt_MASQUERADE 20480 1 2025-09-09T14:04:17.5375975Z nf_nat 57344 2 nft_chain_nat,xt_MASQUERADE 2025-09-09T14:04:17.5376295Z nf_conntrack_netlink 57344 0 2025-09-09T14:04:17.5376685Z nf_conntrack 184320 4 xt_conntrack,nf_nat,nf_conntrack_netlink,xt_MASQUERADE 2025-09-09T14:04:17.5377104Z nf_defrag_ipv6 24576 1 nf_conntrack 2025-09-09T14:04:17.5377411Z nf_defrag_ipv4 16384 1 nf_conntrack 2025-09-09T14:04:17.5377694Z xfrm_user 57344 1 2025-09-09T14:04:17.5377952Z xfrm_algo 16384 1 xfrm_user 2025-09-09T14:04:17.5378234Z xt_addrtype 16384 2 2025-09-09T14:04:17.5378478Z nft_compat 20480 4 2025-09-09T14:04:17.5378782Z nf_tables 311296 57 nft_compat,nft_chain_nat 2025-09-09T14:04:17.5379177Z nfnetlink 20480 4 nft_compat,nf_conntrack_netlink,nf_tables 2025-09-09T14:04:17.5379542Z br_netfilter 36864 0 2025-09-09T14:04:17.5379806Z bridge 323584 1 br_netfilter 2025-09-09T14:04:17.5380093Z stp 16384 1 bridge 2025-09-09T14:04:17.5380363Z llc 16384 2 bridge,stp 2025-09-09T14:04:17.5380636Z overlay 167936 0 2025-09-09T14:04:17.5380884Z tls 139264 0 2025-09-09T14:04:17.5381117Z nls_ascii 16384 1 2025-09-09T14:04:17.5381364Z nls_cp437 20480 1 2025-09-09T14:04:17.5381597Z vfat 24576 1 2025-09-09T14:04:17.5381855Z fat 86016 1 vfat 2025-09-09T14:04:17.5382129Z ena 180224 0 2025-09-09T14:04:17.5382390Z i8042 45056 0 2025-09-09T14:04:17.5382621Z sunrpc 700416 1 2025-09-09T14:04:17.5383010Z serio 28672 3 i8042 2025-09-09T14:04:17.5383269Z button 24576 0 2025-09-09T14:04:17.5383516Z ghash_clmulni_intel 16384 0 2025-09-09T14:04:17.5383770Z sch_fq_codel 20480 33 2025-09-09T14:04:17.5384011Z dm_mod 188416 0 2025-09-09T14:04:17.5384247Z fuse 184320 1 2025-09-09T14:04:17.5384485Z loop 36864 0 2025-09-09T14:04:17.5384725Z configfs 57344 1 2025-09-09T14:04:17.5384962Z dmi_sysfs 20480 0 2025-09-09T14:04:17.5385203Z crc32_pclmul 16384 0 2025-09-09T14:04:17.5385443Z crc32c_intel 24576 0 2025-09-09T14:04:17.5385687Z efivarfs 24576 1 2025-09-09T14:04:17.5385921Z + modinfo nvidia 2025-09-09T14:04:17.5397213Z filename: /lib/modules/6.1.141-155.222.amzn2023.x86_64/kernel/drivers/video/nvidia.ko 2025-09-09T14:04:17.5397807Z import_ns: DMA_BUF 2025-09-09T14:04:17.5398049Z alias: char-major-195-* 2025-09-09T14:04:17.5398315Z version: 580.65.06 2025-09-09T14:04:17.5398555Z supported: external 2025-09-09T14:04:17.5398793Z license: Dual MIT/GPL 2025-09-09T14:04:17.5399073Z firmware: nvidia/580.65.06/gsp_tu10x.bin 2025-09-09T14:04:17.5399394Z firmware: nvidia/580.65.06/gsp_ga10x.bin 2025-09-09T14:04:17.5399713Z srcversion: A69EBF72FC9D60E11E9A05C 2025-09-09T14:04:17.5400027Z alias: of:N*T*Cnvidia,tegra264-displayC* 2025-09-09T14:04:17.5400368Z alias: of:N*T*Cnvidia,tegra264-display 2025-09-09T14:04:17.5400701Z alias: of:N*T*Cnvidia,tegra234-displayC* 2025-09-09T14:04:17.5401041Z alias: of:N*T*Cnvidia,tegra234-display 2025-09-09T14:04:17.5401372Z alias: pci:v000010DEd*sv*sd*bc06sc80i00* 2025-09-09T14:04:17.5401848Z alias: pci:v000010DEd*sv*sd*bc03sc02i00* 2025-09-09T14:04:17.5402177Z alias: pci:v000010DEd*sv*sd*bc03sc00i00* 2025-09-09T14:04:17.5402474Z depends: i2c-core,drm 2025-09-09T14:04:17.5402727Z retpoline: Y 2025-09-09T14:04:17.5402942Z name: nvidia 2025-09-09T14:04:17.5403295Z vermagic: 6.1.141-155.222.amzn2023.x86_64 SMP preempt mod_unload modversions 2025-09-09T14:04:17.5403748Z parm: NvSwitchRegDwords:NvSwitch regkey (charp) 2025-09-09T14:04:17.5404184Z parm: NvSwitchBlacklist:NvSwitchBlacklist=uuid[,uuid...] (charp) 2025-09-09T14:04:17.5404598Z parm: NVreg_ResmanDebugLevel:int 2025-09-09T14:04:17.5404897Z parm: NVreg_RmLogonRC:int 2025-09-09T14:04:17.5405194Z parm: NVreg_ModifyDeviceFiles:int 2025-09-09T14:04:17.5405496Z parm: NVreg_DeviceFileUID:int 2025-09-09T14:04:17.5405790Z parm: NVreg_DeviceFileGID:int 2025-09-09T14:04:17.5406082Z parm: NVreg_DeviceFileMode:int 2025-09-09T14:04:17.5406444Z parm: NVreg_InitializeSystemMemoryAllocations:int 2025-09-09T14:04:17.5406817Z parm: NVreg_UsePageAttributeTable:int 2025-09-09T14:04:17.5407143Z parm: NVreg_EnablePCIeGen3:int 2025-09-09T14:04:17.5407444Z parm: NVreg_EnableMSI:int 2025-09-09T14:04:17.5407735Z parm: NVreg_EnableStreamMemOPs:int 2025-09-09T14:04:17.5408095Z parm: NVreg_RestrictProfilingToAdminUsers:int 2025-09-09T14:04:17.5408483Z parm: NVreg_PreserveVideoMemoryAllocations:int 2025-09-09T14:04:17.5408858Z parm: NVreg_EnableS0ixPowerManagement:int 2025-09-09T14:04:17.5409258Z parm: NVreg_S0ixPowerManagementVideoMemoryThreshold:int 2025-09-09T14:04:17.5409662Z parm: NVreg_DynamicPowerManagement:int 2025-09-09T14:04:17.5410068Z parm: NVreg_DynamicPowerManagementVideoMemoryThreshold:int 2025-09-09T14:04:17.5410471Z parm: NVreg_EnableGpuFirmware:int 2025-09-09T14:04:17.5410799Z parm: NVreg_EnableGpuFirmwareLogs:int 2025-09-09T14:04:17.5411158Z parm: NVreg_OpenRmEnableUnsupportedGpus:int 2025-09-09T14:04:17.5411524Z parm: NVreg_EnableUserNUMAManagement:int 2025-09-09T14:04:17.5411847Z parm: NVreg_MemoryPoolSize:int 2025-09-09T14:04:17.5412287Z parm: NVreg_KMallocHeapMaxSize:int 2025-09-09T14:04:17.5412605Z parm: NVreg_VMallocHeapMaxSize:int 2025-09-09T14:04:17.5412924Z parm: NVreg_IgnoreMMIOCheck:int 2025-09-09T14:04:17.5413220Z parm: NVreg_NvLinkDisable:int 2025-09-09T14:04:17.5413557Z parm: NVreg_EnablePCIERelaxedOrderingMode:int 2025-09-09T14:04:17.5413907Z parm: NVreg_RegisterPCIDriver:int 2025-09-09T14:04:17.5414245Z parm: NVreg_RegisterPlatformDeviceDriver:int 2025-09-09T14:04:17.5414597Z parm: NVreg_EnableResizableBar:int 2025-09-09T14:04:17.5414916Z parm: NVreg_EnableDbgBreakpoint:int 2025-09-09T14:04:17.5415260Z parm: NVreg_EnableNonblockingOpen:int 2025-09-09T14:04:17.5415601Z parm: NVreg_CoherentGPUMemoryMode:charp 2025-09-09T14:04:17.5416006Z parm: NVreg_RegistryDwords:charp 2025-09-09T14:04:17.5416328Z parm: NVreg_RegistryDwordsPerDevice:charp 2025-09-09T14:04:17.5416666Z parm: NVreg_RmMsg:charp 2025-09-09T14:04:17.5416940Z parm: NVreg_GpuBlacklist:charp 2025-09-09T14:04:17.5417259Z parm: NVreg_TemporaryFilePath:charp 2025-09-09T14:04:17.5417574Z parm: NVreg_ExcludedGpus:charp 2025-09-09T14:04:17.5417876Z parm: NVreg_DmaRemapPeerMmio:int 2025-09-09T14:04:17.5418199Z parm: NVreg_RmNvlinkBandwidth:charp 2025-09-09T14:04:17.5418539Z parm: NVreg_RmNvlinkBandwidthLinkCount:int 2025-09-09T14:04:17.5418880Z parm: NVreg_ImexChannelCount:int 2025-09-09T14:04:17.5419192Z parm: NVreg_CreateImexChannel0:int 2025-09-09T14:04:17.5419533Z parm: NVreg_GrdmaPciTopoCheckOverride:int 2025-09-09T14:04:17.5419859Z parm: rm_firmware_active:charp 2025-09-09T14:04:17.5420262Z + set +e 2025-09-09T14:04:17.5420449Z + nvidia-smi 2025-09-09T14:04:19.9008134Z Tue Sep 9 14:04:19 2025 2025-09-09T14:04:19.9008537Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:04:19.9009074Z | NVIDIA-SMI 580.65.06 Driver Version: 580.65.06 CUDA Version: 13.0 | 2025-09-09T14:04:19.9009539Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:04:19.9010029Z | GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC | 2025-09-09T14:04:19.9010538Z | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | 2025-09-09T14:04:19.9010963Z | | | MIG M. | 2025-09-09T14:04:19.9011288Z |=========================================+========================+======================| 2025-09-09T14:04:19.9351615Z | 0 NVIDIA A10G Off | 00000000:00:1B.0 Off | 0 | 2025-09-09T14:04:19.9352511Z | 0% 28C P0 57W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:04:19.9353268Z | | | N/A | 2025-09-09T14:04:19.9353832Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:04:19.9354296Z | 1 NVIDIA A10G Off | 00000000:00:1C.0 Off | 0 | 2025-09-09T14:04:19.9354711Z | 0% 28C P0 58W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:04:19.9355076Z | | | N/A | 2025-09-09T14:04:19.9355445Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:04:19.9355874Z | 2 NVIDIA A10G Off | 00000000:00:1D.0 Off | 0 | 2025-09-09T14:04:19.9356291Z | 0% 28C P0 57W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:04:19.9356651Z | | | N/A | 2025-09-09T14:04:19.9357377Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:04:19.9357805Z | 3 NVIDIA A10G Off | 00000000:00:1E.0 Off | 0 | 2025-09-09T14:04:19.9358218Z | 0% 27C P0 59W / 300W | 0MiB / 23028MiB | 2% Default | 2025-09-09T14:04:19.9358580Z | | | N/A | 2025-09-09T14:04:19.9358951Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:04:19.9359248Z 2025-09-09T14:04:19.9359421Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:04:19.9359842Z | Processes: | 2025-09-09T14:04:19.9360274Z | GPU GI CI PID Type Process name GPU Memory | 2025-09-09T14:04:19.9360684Z | ID ID Usage | 2025-09-09T14:04:19.9361013Z |=========================================================================================| 2025-09-09T14:04:19.9379708Z | No running processes found | 2025-09-09T14:04:19.9380160Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:04:21.5973900Z + nvidia-smi --query-gpu=gpu_name --format=csv,noheader --id=0 2025-09-09T14:04:23.9432720Z NVIDIA A10G 2025-09-09T14:04:25.0346602Z + NVIDIA_SMI_STATUS=0 2025-09-09T14:04:25.0346873Z + '[' 0 -eq 0 ']' 2025-09-09T14:04:25.0347120Z + echo 'INFO: Ignoring allowed status 0' 2025-09-09T14:04:25.0347719Z + set -e 2025-09-09T14:04:25.0347938Z INFO: Ignoring allowed status 0 2025-09-09T14:04:25.0359057Z == Installing nvidia container toolkit for amzn2023 == 2025-09-09T14:04:25.0364541Z + sudo yum install -y yum-utils 2025-09-09T14:04:25.4806141Z Last metadata expiration check: 0:07:16 ago on Tue Sep 9 13:57:09 2025. 2025-09-09T14:04:25.5051773Z Package dnf-utils-4.3.0-13.amzn2023.0.5.noarch is already installed. 2025-09-09T14:04:25.5516864Z Dependencies resolved. 2025-09-09T14:04:25.5741455Z Nothing to do. 2025-09-09T14:04:25.5741918Z Complete! 2025-09-09T14:04:25.6132883Z + [[ amzn2023 == \a\m\z\n\2\0\2\3 ]] 2025-09-09T14:04:25.6134024Z + YUM_REPO_URL=https://nvidia.github.io/libnvidia-container/stable/rpm/nvidia-container-toolkit.repo 2025-09-09T14:04:25.6135104Z + sudo yum-config-manager --add-repo https://nvidia.github.io/libnvidia-container/stable/rpm/nvidia-container-toolkit.repo 2025-09-09T14:04:25.9285393Z Adding repo from: https://nvidia.github.io/libnvidia-container/stable/rpm/nvidia-container-toolkit.repo 2025-09-09T14:04:25.9762931Z + sudo yum install -y nvidia-container-toolkit-1.17.8 libnvidia-container-tools-1.17.8 libnvidia-container1-1.17.8 nvidia-container-toolkit-base-1.17.8 2025-09-09T14:04:26.4832320Z nvidia-container-toolkit 18 kB/s | 833 B 00:00 2025-09-09T14:04:26.5082621Z Package nvidia-container-toolkit-1.17.8-1.x86_64 is already installed. 2025-09-09T14:04:26.5088505Z Package libnvidia-container-tools-1.17.8-1.x86_64 is already installed. 2025-09-09T14:04:26.5092243Z Package libnvidia-container1-1.17.8-1.x86_64 is already installed. 2025-09-09T14:04:26.5099868Z Package nvidia-container-toolkit-base-1.17.8-1.x86_64 is already installed. 2025-09-09T14:04:26.5570466Z Dependencies resolved. 2025-09-09T14:04:26.5795036Z Nothing to do. 2025-09-09T14:04:26.5795253Z Complete! 2025-09-09T14:04:26.6195643Z + sudo systemctl restart docker 2025-09-09T14:04:45.3975443Z Tue Sep 9 14:04:45 2025 2025-09-09T14:04:45.3975827Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:04:45.3976468Z | NVIDIA-SMI 580.65.06 Driver Version: 580.65.06 CUDA Version: 13.0 | 2025-09-09T14:04:45.3977408Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:04:45.3977917Z | GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC | 2025-09-09T14:04:45.3978440Z | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | 2025-09-09T14:04:45.3978860Z | | | MIG M. | 2025-09-09T14:04:45.3979187Z |=========================================+========================+======================| 2025-09-09T14:04:45.4331526Z | 0 NVIDIA A10G On | 00000000:00:1B.0 Off | 0 | 2025-09-09T14:04:45.4331980Z | 0% 29C P0 57W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:04:45.4332379Z | | | N/A | 2025-09-09T14:04:45.4332773Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:04:45.4333236Z | 1 NVIDIA A10G On | 00000000:00:1C.0 Off | 0 | 2025-09-09T14:04:45.4333656Z | 0% 28C P0 58W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:04:45.4334025Z | | | N/A | 2025-09-09T14:04:45.4334404Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:04:45.4334843Z | 2 NVIDIA A10G On | 00000000:00:1D.0 Off | 0 | 2025-09-09T14:04:45.4335252Z | 0% 28C P0 57W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:04:45.4335989Z | | | N/A | 2025-09-09T14:04:45.4336430Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:04:45.4336989Z | 3 NVIDIA A10G On | 00000000:00:1E.0 Off | 0 | 2025-09-09T14:04:45.4337598Z | 0% 27C P0 55W / 300W | 0MiB / 23028MiB | 2% Default | 2025-09-09T14:04:45.4338074Z | | | N/A | 2025-09-09T14:04:45.4338543Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:04:45.4339495Z 2025-09-09T14:04:45.4339779Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:04:45.4350166Z | Processes: | 2025-09-09T14:04:45.4350831Z | GPU GI CI PID Type Process name GPU Memory | 2025-09-09T14:04:45.4351412Z | ID ID Usage | 2025-09-09T14:04:45.4351756Z |=========================================================================================| 2025-09-09T14:04:45.4360757Z | No running processes found | 2025-09-09T14:04:45.4361356Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:04:46.0676980Z Unable to find image 'public.ecr.aws/docker/library/python:3.13' locally 2025-09-09T14:04:46.3162615Z 3.13: Pulling from docker/library/python 2025-09-09T14:04:46.4249204Z 15b1d8a5ff03: Pulling fs layer 2025-09-09T14:04:46.4249754Z 22718812f617: Pulling fs layer 2025-09-09T14:04:46.4250273Z 401a98f7495b: Pulling fs layer 2025-09-09T14:04:46.4250773Z ad446e7df19a: Pulling fs layer 2025-09-09T14:04:46.4251280Z 5d32990caa16: Pulling fs layer 2025-09-09T14:04:46.4251626Z a79d633abf9a: Pulling fs layer 2025-09-09T14:04:46.4251892Z 249a56c8e466: Pulling fs layer 2025-09-09T14:04:46.4252137Z ad446e7df19a: Waiting 2025-09-09T14:04:46.4252358Z a79d633abf9a: Waiting 2025-09-09T14:04:46.4252580Z 249a56c8e466: Waiting 2025-09-09T14:04:46.5483964Z 22718812f617: Verifying Checksum 2025-09-09T14:04:46.5484261Z 22718812f617: Download complete 2025-09-09T14:04:46.6371699Z 15b1d8a5ff03: Verifying Checksum 2025-09-09T14:04:46.6371995Z 15b1d8a5ff03: Download complete 2025-09-09T14:04:46.7175111Z 5d32990caa16: Verifying Checksum 2025-09-09T14:04:46.7175504Z 5d32990caa16: Download complete 2025-09-09T14:04:46.7432191Z 401a98f7495b: Verifying Checksum 2025-09-09T14:04:46.7432574Z 401a98f7495b: Download complete 2025-09-09T14:04:46.8083674Z 249a56c8e466: Verifying Checksum 2025-09-09T14:04:46.8084057Z 249a56c8e466: Download complete 2025-09-09T14:04:46.8784071Z a79d633abf9a: Verifying Checksum 2025-09-09T14:04:46.8784464Z a79d633abf9a: Download complete 2025-09-09T14:04:47.2420610Z ad446e7df19a: Verifying Checksum 2025-09-09T14:04:47.2421274Z ad446e7df19a: Download complete 2025-09-09T14:04:48.3795833Z 15b1d8a5ff03: Pull complete 2025-09-09T14:04:49.0880601Z 22718812f617: Pull complete 2025-09-09T14:04:51.5570674Z 401a98f7495b: Pull complete 2025-09-09T14:04:58.7529456Z ad446e7df19a: Pull complete 2025-09-09T14:04:59.1498456Z 5d32990caa16: Pull complete 2025-09-09T14:04:59.9127877Z a79d633abf9a: Pull complete 2025-09-09T14:04:59.9362282Z 249a56c8e466: Pull complete 2025-09-09T14:04:59.9498335Z Digest: sha256:74503e0bff6cf811f029590a05e0218cc9ba3e099a4b7df0ab84a67df081e1bc 2025-09-09T14:04:59.9541985Z Status: Downloaded newer image for public.ecr.aws/docker/library/python:3.13 2025-09-09T14:05:05.9677370Z Tue Sep 9 14:05:05 2025 2025-09-09T14:05:05.9677902Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:05:05.9678480Z | NVIDIA-SMI 580.65.06 Driver Version: 580.65.06 CUDA Version: 13.0 | 2025-09-09T14:05:05.9681767Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:05:05.9682259Z | GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC | 2025-09-09T14:05:05.9682800Z | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | 2025-09-09T14:05:05.9683233Z | | | MIG M. | 2025-09-09T14:05:05.9683560Z |=========================================+========================+======================| 2025-09-09T14:05:06.0305229Z | 0 NVIDIA A10G On | 00000000:00:1B.0 Off | 0 | 2025-09-09T14:05:06.0305833Z | 0% 25C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:05:06.0306335Z | | | N/A | 2025-09-09T14:05:06.0306739Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:05:06.0307181Z | 1 NVIDIA A10G On | 00000000:00:1C.0 Off | 0 | 2025-09-09T14:05:06.0307602Z | 0% 25C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:05:06.0307970Z | | | N/A | 2025-09-09T14:05:06.0308367Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:05:06.0308811Z | 2 NVIDIA A10G On | 00000000:00:1D.0 Off | 0 | 2025-09-09T14:05:06.0309222Z | 0% 25C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:05:06.0309593Z | | | N/A | 2025-09-09T14:05:06.0309972Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:05:06.0310414Z | 3 NVIDIA A10G On | 00000000:00:1E.0 Off | 0 | 2025-09-09T14:05:06.0310824Z | 0% 25C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:05:06.0311453Z | | | N/A | 2025-09-09T14:05:06.0311854Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:05:06.0334544Z 2025-09-09T14:05:06.0335109Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:05:06.0335702Z | Processes: | 2025-09-09T14:05:06.0336351Z | GPU GI CI PID Type Process name GPU Memory | 2025-09-09T14:05:06.0336872Z | ID ID Usage | 2025-09-09T14:05:06.0337300Z |=========================================================================================| 2025-09-09T14:05:06.0372564Z | No running processes found | 2025-09-09T14:05:06.0373224Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:05:08.4406387Z Command completed after 1 attempt(s). 2025-09-09T14:05:08.4502499Z ##[group]Run # ignore expansion of "docker ps -q" since it could be empty 2025-09-09T14:05:08.4503066Z # ignore expansion of "docker ps -q" since it could be empty 2025-09-09T14:05:08.4503483Z # shellcheck disable=SC2046 2025-09-09T14:05:08.4503818Z docker stop $(docker ps -q) || true 2025-09-09T14:05:08.4504154Z # Prune all of the docker images 2025-09-09T14:05:08.4504465Z docker system prune -af 2025-09-09T14:05:08.4519779Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:05:08.4520135Z env: 2025-09-09T14:05:08.4520604Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:08.4520944Z REPOSITORY: pytorch/ao 2025-09-09T14:05:08.4521209Z PR_NUMBER: 2963 2025-09-09T14:05:08.4523006Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:05:08.4524981Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:05:08.4525523Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:05:08.4526035Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:05:08.4526468Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:05:08.4526805Z ##[endgroup] 2025-09-09T14:05:08.4866714Z "docker stop" requires at least 1 argument. 2025-09-09T14:05:08.4867115Z See 'docker stop --help'. 2025-09-09T14:05:08.4867291Z 2025-09-09T14:05:08.4867453Z Usage: docker stop [OPTIONS] CONTAINER [CONTAINER...] 2025-09-09T14:05:08.4867699Z 2025-09-09T14:05:08.4867804Z Stop one or more running containers 2025-09-09T14:05:09.7397693Z Deleted Images: 2025-09-09T14:05:09.7398107Z untagged: public.ecr.aws/docker/library/python:3.13 2025-09-09T14:05:09.7399007Z untagged: public.ecr.aws/docker/library/python@sha256:74503e0bff6cf811f029590a05e0218cc9ba3e099a4b7df0ab84a67df081e1bc 2025-09-09T14:05:09.7400009Z deleted: sha256:77f2b24be2b3987f6d59918787d226acb4e6612644bacb3dd37adc494e477d9e 2025-09-09T14:05:09.7400828Z deleted: sha256:1b9aa91044866f8707424c8fe367f924a48557eac69f7485fd6d2a3a116c74d5 2025-09-09T14:05:09.7401641Z deleted: sha256:b86402d18e73d4825a3bd2a09244a93487ba4687ca7c9dcba0f73e160840845c 2025-09-09T14:05:09.7402473Z deleted: sha256:5755f8963eb047a0086073c3a7dd0731296d6751a7445f3693a52b30020a5b65 2025-09-09T14:05:09.7403212Z deleted: sha256:7f33dbfa9475d25622f49ed51f4164c97de1303331c77dfdc738e084d100f50c 2025-09-09T14:05:09.7404037Z deleted: sha256:19daa38049795ba2c166dd898c81b17e31f4b5f98c1337846c6515fff97d8782 2025-09-09T14:05:09.7404626Z deleted: sha256:483bd23b5d7e66fc0f8a92dbfacc3d72fad97ef47dd4767889979a803bc1f5b8 2025-09-09T14:05:09.7405214Z deleted: sha256:185e04da9d947141fd703dbf36361bdc2ff77cc27cbf500fb9f4881cb5ddbe95 2025-09-09T14:05:09.7405556Z 2025-09-09T14:05:09.7612648Z Total reclaimed space: 1.109GB 2025-09-09T14:05:09.7694555Z ##[group]Run ./test-infra/.github/actions/setup-ssh 2025-09-09T14:05:09.7694894Z with: 2025-09-09T14:05:09.7695494Z github-secret: *** 2025-09-09T14:05:09.7696227Z instructions: All testing is done inside the container, to start an interactive session run: docker exec -it $(docker container ps --format '{{.ID}}') bash 2025-09-09T14:05:09.7696929Z activate-with-label: false 2025-09-09T14:05:09.7697190Z label: with-ssh 2025-09-09T14:05:09.7697667Z remove-existing-keys: true 2025-09-09T14:05:09.7697934Z fail-silently: true 2025-09-09T14:05:09.7698148Z env: 2025-09-09T14:05:09.7698398Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:09.7698736Z REPOSITORY: pytorch/ao 2025-09-09T14:05:09.7698979Z PR_NUMBER: 2963 2025-09-09T14:05:09.7700789Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:05:09.7702716Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:05:09.7703450Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:05:09.7703969Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:05:09.7704396Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:05:09.7704730Z ##[endgroup] 2025-09-09T14:05:09.8826049Z Please see https://github.com/pytorch/pytorch/wiki/Debugging-using-with-ssh-for-Github-Actions for more info. 2025-09-09T14:05:10.4640367Z Grabbing public ssh keys from https://github.com/andrewor14.keys 2025-09-09T14:05:10.5465695Z ~/.ssh/authorized_keys file found on node, removing ~/.ssh and starting fresh 2025-09-09T14:05:10.5479969Z Public keys pulled and installed to /home/ec2-user/.ssh/authorized_keys 2025-09-09T14:05:10.5518624Z Login using: ssh ec2-user@ec2-34-207-236-206.compute-1.amazonaws.com 2025-09-09T14:05:10.5519207Z All testing is done inside the container, to start an interactive session run: 2025-09-09T14:05:10.5519810Z docker exec -it $(docker container ps --format '{{.ID}}') bash 2025-09-09T14:05:10.5674861Z ##[group]Run actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 2025-09-09T14:05:10.5675271Z with: 2025-09-09T14:05:10.5675493Z repository: pytorch/ao 2025-09-09T14:05:10.5675744Z ref: refs/pull/2963/merge 2025-09-09T14:05:10.5675996Z path: pytorch/ao 2025-09-09T14:05:10.5676214Z fetch-depth: 1 2025-09-09T14:05:10.5676444Z submodules: recursive 2025-09-09T14:05:10.5676795Z token: *** 2025-09-09T14:05:10.5677007Z ssh-strict: true 2025-09-09T14:05:10.5677226Z ssh-user: git 2025-09-09T14:05:10.5677449Z persist-credentials: true 2025-09-09T14:05:10.5677703Z clean: true 2025-09-09T14:05:10.5677932Z sparse-checkout-cone-mode: true 2025-09-09T14:05:10.5678213Z fetch-tags: false 2025-09-09T14:05:10.5678432Z show-progress: true 2025-09-09T14:05:10.5678663Z lfs: false 2025-09-09T14:05:10.5678873Z set-safe-directory: true 2025-09-09T14:05:10.5679131Z env: 2025-09-09T14:05:10.5679397Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:10.5679773Z REPOSITORY: pytorch/ao 2025-09-09T14:05:10.5680012Z PR_NUMBER: 2963 2025-09-09T14:05:10.5681784Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:05:10.5683708Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:05:10.5684253Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:05:10.5684747Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:05:10.5685180Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:05:10.5685505Z ##[endgroup] 2025-09-09T14:05:10.6677247Z Syncing repository: pytorch/ao 2025-09-09T14:05:10.6684211Z ##[group]Getting Git version info 2025-09-09T14:05:10.6684627Z Working directory is '/home/ec2-user/actions-runner/_work/ao/ao/pytorch/ao' 2025-09-09T14:05:10.6711660Z [command]/usr/bin/git version 2025-09-09T14:05:10.6761654Z git version 2.47.1 2025-09-09T14:05:10.6787953Z ##[endgroup] 2025-09-09T14:05:10.6812077Z Temporarily overriding HOME='/home/ec2-user/actions-runner/_work/_temp/d64d79c5-f5d0-4103-9e85-d69d1eda0c9e' before making global git config changes 2025-09-09T14:05:10.6813101Z Adding repository directory to the temporary git global config as a safe directory 2025-09-09T14:05:10.6817725Z [command]/usr/bin/git config --global --add safe.directory /home/ec2-user/actions-runner/_work/ao/ao/pytorch/ao 2025-09-09T14:05:10.6856189Z ##[group]Initializing the repository 2025-09-09T14:05:10.6860910Z [command]/usr/bin/git init /home/ec2-user/actions-runner/_work/ao/ao/pytorch/ao 2025-09-09T14:05:10.6907491Z hint: Using 'master' as the name for the initial branch. This default branch name 2025-09-09T14:05:10.6908069Z hint: is subject to change. To configure the initial branch name to use in all 2025-09-09T14:05:10.6908583Z hint: of your new repositories, which will suppress this warning, call: 2025-09-09T14:05:10.6908963Z hint: 2025-09-09T14:05:10.6909222Z hint: git config --global init.defaultBranch 2025-09-09T14:05:10.6909537Z hint: 2025-09-09T14:05:10.6909837Z hint: Names commonly chosen instead of 'master' are 'main', 'trunk' and 2025-09-09T14:05:10.6910352Z hint: 'development'. The just-created branch can be renamed via this command: 2025-09-09T14:05:10.6910747Z hint: 2025-09-09T14:05:10.6910948Z hint: git branch -m 2025-09-09T14:05:10.6911411Z Initialized empty Git repository in /home/ec2-user/actions-runner/_work/ao/ao/pytorch/ao/.git/ 2025-09-09T14:05:10.6919974Z [command]/usr/bin/git remote add origin https://github.com/pytorch/ao 2025-09-09T14:05:10.6968415Z ##[endgroup] 2025-09-09T14:05:10.6969120Z ##[group]Disabling automatic garbage collection 2025-09-09T14:05:10.6972745Z [command]/usr/bin/git config --local gc.auto 0 2025-09-09T14:05:10.7008634Z ##[endgroup] 2025-09-09T14:05:10.7009190Z ##[group]Setting up auth 2025-09-09T14:05:10.7013653Z [command]/usr/bin/git config --local --name-only --get-regexp core\.sshCommand 2025-09-09T14:05:10.7048858Z [command]/usr/bin/git submodule foreach --recursive sh -c "git config --local --name-only --get-regexp 'core\.sshCommand' && git config --local --unset-all 'core.sshCommand' || :" 2025-09-09T14:05:10.7468143Z [command]/usr/bin/git config --local --name-only --get-regexp http\.https\:\/\/github\.com\/\.extraheader 2025-09-09T14:05:10.7503462Z [command]/usr/bin/git submodule foreach --recursive sh -c "git config --local --name-only --get-regexp 'http\.https\:\/\/github\.com\/\.extraheader' && git config --local --unset-all 'http.https://github.com/.extraheader' || :" 2025-09-09T14:05:10.7923276Z [command]/usr/bin/git config --local http.https://github.com/.extraheader AUTHORIZATION: basic *** 2025-09-09T14:05:10.7972822Z ##[endgroup] 2025-09-09T14:05:10.7973247Z ##[group]Fetching the repository 2025-09-09T14:05:10.7980217Z [command]/usr/bin/git -c protocol.version=2 fetch --no-tags --prune --no-recurse-submodules --depth=1 origin +refs/pull/2963/merge:refs/remotes/pull/2963/merge 2025-09-09T14:05:11.5375121Z From https://github.com/pytorch/ao 2025-09-09T14:05:11.5375967Z * [new ref] refs/pull/2963/merge -> pull/2963/merge 2025-09-09T14:05:11.5405701Z ##[endgroup] 2025-09-09T14:05:11.5406104Z ##[group]Determining the checkout info 2025-09-09T14:05:11.5407813Z ##[endgroup] 2025-09-09T14:05:11.5412428Z [command]/usr/bin/git sparse-checkout disable 2025-09-09T14:05:11.5456653Z [command]/usr/bin/git config --local --unset-all extensions.worktreeConfig 2025-09-09T14:05:11.5494021Z ##[group]Checking out the ref 2025-09-09T14:05:11.5496463Z [command]/usr/bin/git checkout --progress --force refs/remotes/pull/2963/merge 2025-09-09T14:05:11.6861776Z Note: switching to 'refs/remotes/pull/2963/merge'. 2025-09-09T14:05:11.6862147Z 2025-09-09T14:05:11.6862421Z You are in 'detached HEAD' state. You can look around, make experimental 2025-09-09T14:05:11.6863084Z changes and commit them, and you can discard any commits you make in this 2025-09-09T14:05:11.6863747Z state without impacting any branches by switching back to a branch. 2025-09-09T14:05:11.6864140Z 2025-09-09T14:05:11.6864400Z If you want to create a new branch to retain commits you create, you may 2025-09-09T14:05:11.6865008Z do so (now or later) by using -c with the switch command. Example: 2025-09-09T14:05:11.6865347Z 2025-09-09T14:05:11.6865487Z git switch -c 2025-09-09T14:05:11.6865734Z 2025-09-09T14:05:11.6865868Z Or undo this operation with: 2025-09-09T14:05:11.6866084Z 2025-09-09T14:05:11.6866202Z git switch - 2025-09-09T14:05:11.6866360Z 2025-09-09T14:05:11.6867006Z Turn off this advice by setting config variable advice.detachedHead to false 2025-09-09T14:05:11.6867445Z 2025-09-09T14:05:11.6867944Z HEAD is now at 7c05f81 Merge c21284c127b039bc49cc7ffda0e692894ed3b094 into 8b72284fd363b5c096de93fb7ac9cc960a6a601e 2025-09-09T14:05:11.6880019Z ##[endgroup] 2025-09-09T14:05:11.6880575Z ##[group]Setting up auth for fetching submodules 2025-09-09T14:05:11.6886002Z [command]/usr/bin/git config --global http.https://github.com/.extraheader AUTHORIZATION: basic *** 2025-09-09T14:05:11.6936234Z [command]/usr/bin/git config --global --unset-all url.https://github.com/.insteadOf 2025-09-09T14:05:11.6970456Z [command]/usr/bin/git config --global --add url.https://github.com/.insteadOf git@github.com: 2025-09-09T14:05:11.7007821Z [command]/usr/bin/git config --global --add url.https://github.com/.insteadOf org-21003710@github.com: 2025-09-09T14:05:11.7038809Z ##[endgroup] 2025-09-09T14:05:11.7039191Z ##[group]Fetching submodules 2025-09-09T14:05:11.7041943Z [command]/usr/bin/git submodule sync --recursive 2025-09-09T14:05:11.7455701Z [command]/usr/bin/git -c protocol.version=2 submodule update --init --force --depth=1 --recursive 2025-09-09T14:05:11.7847867Z Submodule 'third_party/cutlass' (https://github.com/NVIDIA/cutlass) registered for path 'third_party/cutlass' 2025-09-09T14:05:11.7888185Z Cloning into '/home/ec2-user/actions-runner/_work/ao/ao/pytorch/ao/third_party/cutlass'... 2025-09-09T14:05:13.5953853Z From https://github.com/NVIDIA/cutlass 2025-09-09T14:05:13.5954331Z * branch e51efbfe18fe4f4cbb66ab814c55bf4aa0185491 -> FETCH_HEAD 2025-09-09T14:05:14.3721497Z Submodule path 'third_party/cutlass': checked out 'e51efbfe18fe4f4cbb66ab814c55bf4aa0185491' 2025-09-09T14:05:14.3775606Z [command]/usr/bin/git submodule foreach --recursive git config --local gc.auto 0 2025-09-09T14:05:14.4172849Z Entering 'third_party/cutlass' 2025-09-09T14:05:14.4258060Z ##[endgroup] 2025-09-09T14:05:14.4258470Z ##[group]Persisting credentials for submodules 2025-09-09T14:05:14.4264875Z [command]/usr/bin/git submodule foreach --recursive sh -c "git config --local --name-only --get-regexp 'url\.https\:\/\/github\.com\/\.insteadOf' && git config --local --unset-all 'url.https://github.com/.insteadOf' || :" 2025-09-09T14:05:14.4654517Z Entering 'third_party/cutlass' 2025-09-09T14:05:14.4766037Z [command]/usr/bin/git submodule foreach --recursive sh -c "git config --local 'http.https://github.com/.extraheader' 'AUTHORIZATION: basic ***' && git config --local --show-origin --name-only --get-regexp remote.origin.url" 2025-09-09T14:05:14.5162695Z Entering 'third_party/cutlass' 2025-09-09T14:05:14.5241526Z file:/home/ec2-user/actions-runner/_work/ao/ao/pytorch/ao/.git/modules/third_party/cutlass/config remote.origin.url 2025-09-09T14:05:14.5305662Z [command]/usr/bin/git submodule foreach --recursive git config --local --add 'url.https://github.com/.insteadOf' 'git@github.com:' 2025-09-09T14:05:14.5692593Z Entering 'third_party/cutlass' 2025-09-09T14:05:14.5785156Z [command]/usr/bin/git submodule foreach --recursive git config --local --add 'url.https://github.com/.insteadOf' 'org-21003710@github.com:' 2025-09-09T14:05:14.6172544Z Entering 'third_party/cutlass' 2025-09-09T14:05:14.6257969Z ##[endgroup] 2025-09-09T14:05:14.6304243Z [command]/usr/bin/git log -1 --format=%H 2025-09-09T14:05:14.6334161Z 7c05f811b89289f7be3e0e3546626827f2cc1ca4 2025-09-09T14:05:14.6552610Z Prepare all required actions 2025-09-09T14:05:14.6553051Z Getting action download info 2025-09-09T14:05:14.7861274Z Download action repository 'nick-fields/retry@v3.0.0' (SHA:7152eba30c6575329ac0576536151aca5a72780e) 2025-09-09T14:05:14.9936634Z ##[group]Run ./test-infra/.github/actions/calculate-docker-image 2025-09-09T14:05:14.9937009Z with: 2025-09-09T14:05:14.9937242Z use-custom-docker-registry: true 2025-09-09T14:05:14.9937613Z docker-image-name: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:14.9937976Z docker-build-dir: .ci/docker 2025-09-09T14:05:14.9938266Z working-directory: pytorch/ao 2025-09-09T14:05:14.9938734Z docker-build-script: ./build.sh 2025-09-09T14:05:14.9939101Z docker-registry: 308535385114.dkr.ecr.us-east-1.amazonaws.com 2025-09-09T14:05:14.9939467Z force-push: false 2025-09-09T14:05:14.9939691Z env: 2025-09-09T14:05:14.9939935Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:14.9940263Z REPOSITORY: pytorch/ao 2025-09-09T14:05:14.9940536Z PR_NUMBER: 2963 2025-09-09T14:05:14.9942312Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:05:14.9944249Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:05:14.9944808Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:05:14.9945333Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:05:14.9945762Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:05:14.9946100Z ##[endgroup] 2025-09-09T14:05:14.9969583Z ##[group]Run set -ex 2025-09-09T14:05:14.9969870Z set -ex 2025-09-09T14:05:14.9970075Z  2025-09-09T14:05:14.9970444Z # If the docker build directory or the build script doesn't exist, the action will 2025-09-09T14:05:14.9971046Z # gracefully return the docker image name as it is. Pulling docker image in Linux 2025-09-09T14:05:14.9971577Z # job could then download the pre-built image as usual 2025-09-09T14:05:14.9972202Z if [[ -d "${DOCKER_BUILD_DIR}" ]] && [[ -f "${DOCKER_BUILD_DIR}/${DOCKER_BUILD_SCRIPT}" ]] && [[ "${USE_CUSTOM_DOCKER_REGISTRY}" == "true" ]]; then 2025-09-09T14:05:14.9972783Z  echo "skip=false" >> "${GITHUB_OUTPUT}" 2025-09-09T14:05:14.9973093Z else 2025-09-09T14:05:14.9973339Z  echo "skip=true" >> "${GITHUB_OUTPUT}" 2025-09-09T14:05:14.9973757Z  echo "docker-image=${DOCKER_IMAGE_NAME}" >> "${GITHUB_OUTPUT}" 2025-09-09T14:05:14.9974133Z  2025-09-09T14:05:14.9974639Z  echo "Not using custom ECR registry. Either it was not requested or there is no Docker build script in the ${REPO_NAME} repo..." 2025-09-09T14:05:14.9975216Z  exit 0 2025-09-09T14:05:14.9975419Z fi 2025-09-09T14:05:14.9975620Z  2025-09-09T14:05:14.9976031Z if [[ "${DOCKER_IMAGE_NAME}" == *"${DOCKER_REGISTRY}/${REPO_NAME}"* ]]; then 2025-09-09T14:05:14.9976575Z  # The docker image name already includes the ECR prefix and tag, so we can just 2025-09-09T14:05:14.9977062Z  # use it as it is, but first let's extract the tag 2025-09-09T14:05:14.9977499Z  DOCKER_TAG=$(echo "${DOCKER_IMAGE_NAME}" | awk -F '[:,]' '{print $2}') 2025-09-09T14:05:14.9977966Z  echo "docker-tag=${DOCKER_TAG}" >> "${GITHUB_OUTPUT}" 2025-09-09T14:05:14.9978465Z  echo "docker-image=${DOCKER_IMAGE_NAME}" >> "${GITHUB_OUTPUT}" 2025-09-09T14:05:14.9978839Z else 2025-09-09T14:05:14.9979294Z  if [[ "${DOCKER_IMAGE_NAME}" == *:* ]]; then 2025-09-09T14:05:14.9979659Z  CUSTOM_TAG_PREFIX=${DOCKER_IMAGE_NAME#*:} 2025-09-09T14:05:14.9980023Z  DOCKER_IMAGE_NAME=${DOCKER_IMAGE_NAME%%:*} 2025-09-09T14:05:14.9980371Z  fi 2025-09-09T14:05:14.9980808Z  DOCKER_TAG=${CUSTOM_TAG_PREFIX:+${CUSTOM_TAG_PREFIX}-}$(git rev-parse HEAD:"${DOCKER_BUILD_DIR}") 2025-09-09T14:05:14.9981364Z  echo "docker-tag=${DOCKER_TAG}" >> "${GITHUB_OUTPUT}" 2025-09-09T14:05:14.9981947Z  echo "docker-image=${DOCKER_REGISTRY}/${REPO_NAME}/${DOCKER_IMAGE_NAME}:${DOCKER_TAG}" >> "${GITHUB_OUTPUT}" 2025-09-09T14:05:14.9982676Z  echo "custom-tag-prefix=${CUSTOM_TAG_PREFIX}" >> "${GITHUB_OUTPUT}" 2025-09-09T14:05:14.9983061Z fi 2025-09-09T14:05:14.9991966Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:05:14.9992321Z env: 2025-09-09T14:05:14.9992589Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:14.9992940Z REPOSITORY: pytorch/ao 2025-09-09T14:05:14.9993195Z PR_NUMBER: 2963 2025-09-09T14:05:14.9994961Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:05:14.9996885Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:05:14.9997692Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:05:14.9998261Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:05:14.9998693Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:05:14.9999063Z REPO_NAME: ao 2025-09-09T14:05:14.9999357Z DOCKER_IMAGE_NAME: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:14.9999726Z DOCKER_BUILD_DIR: .ci/docker 2025-09-09T14:05:15.0000014Z DOCKER_BUILD_SCRIPT: ./build.sh 2025-09-09T14:05:15.0000394Z DOCKER_REGISTRY: 308535385114.dkr.ecr.us-east-1.amazonaws.com 2025-09-09T14:05:15.0000795Z USE_CUSTOM_DOCKER_REGISTRY: true 2025-09-09T14:05:15.0001084Z CUSTOM_TAG_PREFIX: 2025-09-09T14:05:15.0001338Z ##[endgroup] 2025-09-09T14:05:15.0034502Z + [[ -d .ci/docker ]] 2025-09-09T14:05:15.0034945Z + echo skip=true 2025-09-09T14:05:15.0035265Z + echo docker-image=pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:15.0035894Z + echo 'Not using custom ECR registry. Either it was not requested or there is no Docker build script in the ao repo...' 2025-09-09T14:05:15.0036433Z + exit 0 2025-09-09T14:05:15.0036880Z Not using custom ECR registry. Either it was not requested or there is no Docker build script in the ao repo... 2025-09-09T14:05:15.0075998Z ##[group]Run set -eux 2025-09-09T14:05:15.0076296Z set -eux 2025-09-09T14:05:15.0076715Z # It's ok if this steps fails, it would then be an anonymous user like what we used to have 2025-09-09T14:05:15.0077764Z aws secretsmanager get-secret-value --secret-id docker_hub_readonly_token | jq --raw-output '.SecretString' | jq -r .docker_hub_readonly_token | docker login --username pytorchbot --password-stdin || true 2025-09-09T14:05:15.0087585Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:05:15.0087951Z env: 2025-09-09T14:05:15.0088211Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:15.0088593Z REPOSITORY: pytorch/ao 2025-09-09T14:05:15.0088854Z PR_NUMBER: 2963 2025-09-09T14:05:15.0090879Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:05:15.0092849Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:05:15.0093421Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:05:15.0094007Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:05:15.0094523Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:05:15.0095308Z ##[endgroup] 2025-09-09T14:05:15.0144300Z + aws secretsmanager get-secret-value --secret-id docker_hub_readonly_token 2025-09-09T14:05:15.0145332Z + jq --raw-output .SecretString 2025-09-09T14:05:15.0147696Z + jq -r .docker_hub_readonly_token 2025-09-09T14:05:15.0148052Z + docker login --username pytorchbot --password-stdin 2025-09-09T14:05:15.6107233Z WARNING! Your password will be stored unencrypted in /home/ec2-user/.docker/config.json. 2025-09-09T14:05:15.6107788Z Configure a credential helper to remove this warning. See 2025-09-09T14:05:15.6108320Z https://docs.docker.com/engine/reference/commandline/login/#credentials-store 2025-09-09T14:05:15.6108676Z 2025-09-09T14:05:15.6109334Z Login Succeeded 2025-09-09T14:05:15.6197629Z Prepare all required actions 2025-09-09T14:05:15.6235227Z ##[group]Run ./test-infra/.github/actions/pull-docker-image 2025-09-09T14:05:15.6235567Z with: 2025-09-09T14:05:15.6235830Z docker-image: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:15.6236255Z docker-registry: 308535385114.dkr.ecr.us-east-1.amazonaws.com 2025-09-09T14:05:15.6236614Z env: 2025-09-09T14:05:15.6236861Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:15.6237195Z REPOSITORY: pytorch/ao 2025-09-09T14:05:15.6237454Z PR_NUMBER: 2963 2025-09-09T14:05:15.6239256Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:05:15.6241197Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:05:15.6241734Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:05:15.6242253Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:05:15.6242685Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:05:15.6243013Z ##[endgroup] 2025-09-09T14:05:15.6262933Z ##[group]Run set -x 2025-09-09T14:05:15.6263230Z set -x 2025-09-09T14:05:15.6263456Z set +e 2025-09-09T14:05:15.6263692Z  2025-09-09T14:05:15.6263898Z login() { 2025-09-09T14:05:15.6264361Z  aws ecr get-login-password --region us-east-1 | docker login -u AWS --password-stdin "$1" 2025-09-09T14:05:15.6264841Z } 2025-09-09T14:05:15.6265047Z  2025-09-09T14:05:15.6265264Z retry () { 2025-09-09T14:05:15.6265525Z  $* || (sleep 1 && $*) || (sleep 2 && $*) 2025-09-09T14:05:15.6265830Z } 2025-09-09T14:05:15.6266032Z  2025-09-09T14:05:15.6266272Z retry login "${DOCKER_REGISTRY}" 2025-09-09T14:05:15.6266562Z  2025-09-09T14:05:15.6267036Z IMAGE_SIZE=$(docker manifest inspect "${DOCKER_IMAGE}" | jq '[.layers[].size, .config.size] | add / 1024 / 1024') 2025-09-09T14:05:15.6267649Z echo "Compressed size of image in MB: ${IMAGE_SIZE}" 2025-09-09T14:05:15.6267999Z  2025-09-09T14:05:15.6268209Z set -e 2025-09-09T14:05:15.6268528Z # ignore output since only exit code is used for conditional 2025-09-09T14:05:15.6268988Z # only pull docker image if it's not available locally 2025-09-09T14:05:15.6269490Z if ! docker inspect --type=image "${DOCKER_IMAGE}" >/dev/null 2>/dev/null; then 2025-09-09T14:05:15.6269953Z  retry docker pull "${DOCKER_IMAGE}" 2025-09-09T14:05:15.6270271Z fi 2025-09-09T14:05:15.6280527Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:05:15.6280877Z env: 2025-09-09T14:05:15.6281146Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:15.6281495Z REPOSITORY: pytorch/ao 2025-09-09T14:05:15.6282000Z PR_NUMBER: 2963 2025-09-09T14:05:15.6283774Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:05:15.6285692Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:05:15.6286385Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:05:15.6286898Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:05:15.6287317Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:05:15.6287738Z DOCKER_REGISTRY: 308535385114.dkr.ecr.us-east-1.amazonaws.com 2025-09-09T14:05:15.6288106Z ##[endgroup] 2025-09-09T14:05:15.6320094Z + set +e 2025-09-09T14:05:15.6320584Z + retry login 308535385114.dkr.ecr.us-east-1.amazonaws.com 2025-09-09T14:05:15.6321000Z + login 308535385114.dkr.ecr.us-east-1.amazonaws.com 2025-09-09T14:05:15.6325413Z + aws ecr get-login-password --region us-east-1 2025-09-09T14:05:15.6325933Z + docker login -u AWS --password-stdin 308535385114.dkr.ecr.us-east-1.amazonaws.com 2025-09-09T14:05:16.1813676Z WARNING! Your password will be stored unencrypted in /home/ec2-user/.docker/config.json. 2025-09-09T14:05:16.1814245Z Configure a credential helper to remove this warning. See 2025-09-09T14:05:16.1814813Z https://docs.docker.com/engine/reference/commandline/login/#credentials-store 2025-09-09T14:05:16.1815176Z 2025-09-09T14:05:16.1815535Z Login Succeeded 2025-09-09T14:05:16.1848468Z ++ docker manifest inspect pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:16.1848912Z ++ jq '[.layers[].size, .config.size] | add / 1024 / 1024' 2025-09-09T14:05:16.3377452Z + IMAGE_SIZE=7943.772059440613 2025-09-09T14:05:16.3377779Z Compressed size of image in MB: 7943.772059440613 2025-09-09T14:05:16.3378171Z + echo 'Compressed size of image in MB: 7943.772059440613' 2025-09-09T14:05:16.3378498Z + set -e 2025-09-09T14:05:16.3378810Z + docker inspect --type=image pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:16.3544236Z + retry docker pull pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:16.3544641Z + docker pull pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:16.5173028Z cuda12.6: Pulling from pytorch/almalinux-builder 2025-09-09T14:05:16.5173755Z 19877a9af8e3: Pulling fs layer 2025-09-09T14:05:16.5174330Z 3b95f7accc18: Pulling fs layer 2025-09-09T14:05:16.5174856Z 09fcdf4cf4fd: Pulling fs layer 2025-09-09T14:05:16.5175381Z 17af5086235f: Pulling fs layer 2025-09-09T14:05:16.5175965Z c3175a707c2d: Pulling fs layer 2025-09-09T14:05:16.5176484Z 550b3c83242f: Pulling fs layer 2025-09-09T14:05:16.5177004Z 018f40a634ae: Pulling fs layer 2025-09-09T14:05:16.5177521Z 4f4fb700ef54: Pulling fs layer 2025-09-09T14:05:16.5178041Z cabce7a916a3: Pulling fs layer 2025-09-09T14:05:16.5178564Z 0b3a66ab554e: Pulling fs layer 2025-09-09T14:05:16.5179077Z 72728e4acc07: Pulling fs layer 2025-09-09T14:05:16.5179581Z 2ca30f8660e0: Pulling fs layer 2025-09-09T14:05:16.5180261Z 45f90a05dbb6: Pulling fs layer 2025-09-09T14:05:16.5180767Z 16125e2dbaa8: Pulling fs layer 2025-09-09T14:05:16.5181038Z 8e08c86db3a1: Pulling fs layer 2025-09-09T14:05:16.5181302Z 550d67135f81: Pulling fs layer 2025-09-09T14:05:16.5181571Z cac5e14b36bd: Pulling fs layer 2025-09-09T14:05:16.5181837Z d9fc50fb0d36: Pulling fs layer 2025-09-09T14:05:16.5182108Z 8b2ffa49399c: Pulling fs layer 2025-09-09T14:05:16.5182371Z 05e4f4570fa0: Pulling fs layer 2025-09-09T14:05:16.5182641Z e16b313a64bb: Pulling fs layer 2025-09-09T14:05:16.5182903Z 87b47e27ca53: Pulling fs layer 2025-09-09T14:05:16.5183161Z 282dc51a39ad: Pulling fs layer 2025-09-09T14:05:16.5183417Z cabce7a916a3: Waiting 2025-09-09T14:05:16.5183916Z 0b3a66ab554e: Waiting 2025-09-09T14:05:16.5184155Z 2ca30f8660e0: Waiting 2025-09-09T14:05:16.5184370Z 72728e4acc07: Waiting 2025-09-09T14:05:16.5184594Z 8b2ffa49399c: Waiting 2025-09-09T14:05:16.5184881Z 45f90a05dbb6: Waiting 2025-09-09T14:05:16.5185102Z 05e4f4570fa0: Waiting 2025-09-09T14:05:16.5185322Z e16b313a64bb: Waiting 2025-09-09T14:05:16.5185538Z 16125e2dbaa8: Waiting 2025-09-09T14:05:16.5185760Z 282dc51a39ad: Waiting 2025-09-09T14:05:16.5185973Z 8e08c86db3a1: Waiting 2025-09-09T14:05:16.5186242Z 87b47e27ca53: Waiting 2025-09-09T14:05:16.5186455Z 550d67135f81: Waiting 2025-09-09T14:05:16.5186673Z cac5e14b36bd: Waiting 2025-09-09T14:05:16.5186887Z 550b3c83242f: Waiting 2025-09-09T14:05:16.5187283Z c3175a707c2d: Waiting 2025-09-09T14:05:16.5187499Z 018f40a634ae: Waiting 2025-09-09T14:05:16.5187722Z d9fc50fb0d36: Waiting 2025-09-09T14:05:16.5187947Z 17af5086235f: Waiting 2025-09-09T14:05:16.5188166Z 4f4fb700ef54: Waiting 2025-09-09T14:05:16.5909018Z 09fcdf4cf4fd: Verifying Checksum 2025-09-09T14:05:16.5909319Z 09fcdf4cf4fd: Download complete 2025-09-09T14:05:16.9634013Z 17af5086235f: Verifying Checksum 2025-09-09T14:05:16.9634300Z 17af5086235f: Download complete 2025-09-09T14:05:17.2534470Z 19877a9af8e3: Download complete 2025-09-09T14:05:17.2869964Z 550b3c83242f: Verifying Checksum 2025-09-09T14:05:17.2870306Z 550b3c83242f: Download complete 2025-09-09T14:05:17.7614866Z 018f40a634ae: Download complete 2025-09-09T14:05:17.7812379Z 3b95f7accc18: Verifying Checksum 2025-09-09T14:05:17.7812684Z 3b95f7accc18: Download complete 2025-09-09T14:05:17.8020031Z 4f4fb700ef54: Verifying Checksum 2025-09-09T14:05:17.8020319Z 4f4fb700ef54: Download complete 2025-09-09T14:05:17.8444494Z cabce7a916a3: Download complete 2025-09-09T14:05:17.8468197Z 0b3a66ab554e: Download complete 2025-09-09T14:05:17.8922226Z 2ca30f8660e0: Verifying Checksum 2025-09-09T14:05:17.8922527Z 2ca30f8660e0: Download complete 2025-09-09T14:05:17.8982653Z 72728e4acc07: Verifying Checksum 2025-09-09T14:05:17.8982962Z 72728e4acc07: Download complete 2025-09-09T14:05:18.0069519Z 16125e2dbaa8: Download complete 2025-09-09T14:05:18.0531786Z 8e08c86db3a1: Verifying Checksum 2025-09-09T14:05:18.0532096Z 8e08c86db3a1: Download complete 2025-09-09T14:05:18.8646820Z c3175a707c2d: Verifying Checksum 2025-09-09T14:05:18.8647242Z c3175a707c2d: Download complete 2025-09-09T14:05:18.8999151Z cac5e14b36bd: Verifying Checksum 2025-09-09T14:05:18.8999462Z cac5e14b36bd: Download complete 2025-09-09T14:05:18.9479113Z d9fc50fb0d36: Download complete 2025-09-09T14:05:18.9829096Z 8b2ffa49399c: Verifying Checksum 2025-09-09T14:05:18.9829535Z 8b2ffa49399c: Download complete 2025-09-09T14:05:19.1120116Z 05e4f4570fa0: Verifying Checksum 2025-09-09T14:05:19.1120472Z 05e4f4570fa0: Download complete 2025-09-09T14:05:19.1799203Z e16b313a64bb: Verifying Checksum 2025-09-09T14:05:19.1799493Z e16b313a64bb: Download complete 2025-09-09T14:05:19.2215170Z 87b47e27ca53: Download complete 2025-09-09T14:05:19.8724813Z 19877a9af8e3: Pull complete 2025-09-09T14:05:22.3351922Z 3b95f7accc18: Pull complete 2025-09-09T14:05:22.4147183Z 550d67135f81: Verifying Checksum 2025-09-09T14:05:22.4147492Z 550d67135f81: Download complete 2025-09-09T14:05:22.5080235Z 09fcdf4cf4fd: Pull complete 2025-09-09T14:05:22.7810568Z 17af5086235f: Pull complete 2025-09-09T14:05:25.0673519Z 282dc51a39ad: Verifying Checksum 2025-09-09T14:05:25.0673854Z 282dc51a39ad: Download complete 2025-09-09T14:05:28.6117946Z c3175a707c2d: Pull complete 2025-09-09T14:05:28.6217732Z 550b3c83242f: Pull complete 2025-09-09T14:05:29.8888655Z 018f40a634ae: Pull complete 2025-09-09T14:05:29.8995341Z 4f4fb700ef54: Pull complete 2025-09-09T14:06:16.3460493Z cabce7a916a3: Pull complete 2025-09-09T14:06:16.3686839Z 0b3a66ab554e: Pull complete 2025-09-09T14:06:16.3914344Z 72728e4acc07: Pull complete 2025-09-09T14:06:16.4135528Z 2ca30f8660e0: Pull complete 2025-09-09T14:06:26.1249556Z 45f90a05dbb6: Verifying Checksum 2025-09-09T14:06:26.1249989Z 45f90a05dbb6: Download complete 2025-09-09T14:07:26.5105609Z 45f90a05dbb6: Pull complete 2025-09-09T14:07:27.1346973Z 16125e2dbaa8: Pull complete 2025-09-09T14:07:27.5985566Z 8e08c86db3a1: Pull complete 2025-09-09T14:07:43.3838546Z 550d67135f81: Pull complete 2025-09-09T14:07:43.7446622Z cac5e14b36bd: Pull complete 2025-09-09T14:07:44.1035180Z d9fc50fb0d36: Pull complete 2025-09-09T14:07:44.4413219Z 8b2ffa49399c: Pull complete 2025-09-09T14:07:45.0198670Z 05e4f4570fa0: Pull complete 2025-09-09T14:07:45.4720715Z e16b313a64bb: Pull complete 2025-09-09T14:07:45.9266691Z 87b47e27ca53: Pull complete 2025-09-09T14:08:05.5489712Z 282dc51a39ad: Pull complete 2025-09-09T14:08:05.7541714Z Digest: sha256:be7f2a4c6f467933b154ac0b3ded894ad1bf06ce95f8f8d908dba108e68806f3 2025-09-09T14:08:05.8614825Z Status: Downloaded newer image for pytorch/almalinux-builder:cuda12.6 2025-09-09T14:08:05.8940820Z docker.io/pytorch/almalinux-builder:cuda12.6 2025-09-09T14:08:05.9005895Z ##[group]Run echo "IN_CONTAINER_RUNNER=$(if [ -f /.inarc ] || [ -f /.incontainer ]; then echo true ; else echo false; fi)" >> "$GITHUB_OUTPUT" 2025-09-09T14:08:05.9006779Z echo "IN_CONTAINER_RUNNER=$(if [ -f /.inarc ] || [ -f /.incontainer ]; then echo true ; else echo false; fi)" >> "$GITHUB_OUTPUT" 2025-09-09T14:08:05.9017860Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:08:05.9018205Z env: 2025-09-09T14:08:05.9018460Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:08:05.9018796Z REPOSITORY: pytorch/ao 2025-09-09T14:08:05.9019034Z PR_NUMBER: 2963 2025-09-09T14:08:05.9020818Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:08:05.9022780Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:08:05.9023310Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:08:05.9023816Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:08:05.9024226Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:08:05.9024564Z ##[endgroup] 2025-09-09T14:08:05.9166677Z Prepare all required actions 2025-09-09T14:08:05.9167029Z Getting action download info 2025-09-09T14:08:06.1121617Z ##[group]Run ./test-infra/.github/actions/setup-nvidia 2025-09-09T14:08:06.1121946Z with: 2025-09-09T14:08:06.1122151Z driver-version: 580.65.06 2025-09-09T14:08:06.1122383Z env: 2025-09-09T14:08:06.1122628Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:08:06.1122954Z REPOSITORY: pytorch/ao 2025-09-09T14:08:06.1123194Z PR_NUMBER: 2963 2025-09-09T14:08:06.1125008Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:08:06.1127014Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:08:06.1127558Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:08:06.1128063Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:08:06.1128497Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:08:06.1128819Z ##[endgroup] 2025-09-09T14:08:06.1275043Z ##[group]Run nick-fields/retry@3e91a01664abd3c5cd539100d10d33b9c5b68482 2025-09-09T14:08:06.1275692Z with: 2025-09-09T14:08:06.1275898Z timeout_minutes: 10 2025-09-09T14:08:06.1276127Z max_attempts: 3 2025-09-09T14:08:06.1300657Z command: # Is it disgusting to have a full shell script here in this github action? Sure # But is it the best way to make it so that this action relies on nothing else? Absolutely set -eou pipefail DISTRIBUTION=$(. /etc/os-release;echo $ID$VERSION_ID) DRIVER_FN="NVIDIA-Linux-x86_64-${DRIVER_VERSION}.run" install_nvidia_docker2_amzn2() { ( set -x # Needed for yum-config-manager sudo yum install -y yum-utils if [[ "${DISTRIBUTION}" == "amzn2023" ]] ; then YUM_REPO_URL="https://nvidia.github.io/libnvidia-container/stable/rpm/nvidia-container-toolkit.repo" else # Amazon Linux 2 YUM_REPO_URL="https://nvidia.github.io/nvidia-docker/${DISTRIBUTION}/nvidia-docker.repo" fi sudo yum-config-manager --add-repo "${YUM_REPO_URL}" sudo yum install -y \ nvidia-container-toolkit-1.17.8 \ libnvidia-container-tools-1.17.8 \ libnvidia-container1-1.17.8 \ nvidia-container-toolkit-base-1.17.8 sudo systemctl restart docker ) } install_nvidia_docker2_ubuntu20() { ( set -x # Install nvidia-driver package if not installed status="$(dpkg-query -W --showformat='${db:Status-Status}' nvidia-docker2 2>&1)" if [ ! $? = 0 ] || [ ! "$status" = installed ]; then sudo apt-get install -y nvidia-container-toolkit-1.17.8 sudo systemctl restart docker fi ) } pre_install_nvidia_driver_amzn2() { ( # Purge any nvidia driver installed from RHEL repo sudo yum remove -y nvidia-driver-latest-dkms ) } install_nvidia_driver_common() { ( # Try to gather more information about the runner and its existing NVIDIA driver if any echo "Before installing NVIDIA driver" lspci lsmod modinfo nvidia || true HAS_NVIDIA_DRIVER=0 # Check if NVIDIA driver has already been installed if [ -x "$(command -v nvidia-smi)" ]; then set +e # The driver exists, check its version next. Also check only the first GPU if there are more than one of them # so that the same driver version is not print over multiple lines INSTALLED_DRIVER_VERSION=$(nvidia-smi --query-gpu=driver_version --format=csv,noheader --id=0) NVIDIA_SMI_STATUS=$? if [ "$NVIDIA_SMI_STATUS" -ne 0 ] && [ "$NVIDIA_SMI_STATUS" -ne 14 ]; then echo "Failed to get NVIDIA driver version ($INSTALLED_DRIVER_VERSION). Continuing" elif [ "$INSTALLED_DRIVER_VERSION" != "$DRIVER_VERSION" ]; then echo "NVIDIA driver ($INSTALLED_DRIVER_VERSION) has been installed, but we expect to have $DRIVER_VERSION instead. Continuing" # Turn off persistent mode so that the installation script can unload the kernel module sudo killall nvidia-persistenced || true else HAS_NVIDIA_DRIVER=1 echo "NVIDIA driver ($INSTALLED_DRIVER_VERSION) has already been installed. Skipping NVIDIA driver installation" fi set -e fi if [ "$HAS_NVIDIA_DRIVER" -eq 0 ]; then # CAUTION: this may need to be updated in future if [ "${DISTRIBUTION}" != ubuntu20.04 ]; then sudo yum groupinstall -y "Development Tools" # ensure our kernel install is the same as our underlying kernel, # groupinstall "Development Tools" has a habit of mismatching kernel headers sudo yum install -y "kernel-devel-uname-r == $(uname -r)" sudo modprobe backlight fi sudo curl -fsL -o /tmp/nvidia_driver "https://s3.amazonaws.com/ossci-linux/nvidia_driver/$DRIVER_FN" set +e sudo /bin/bash /tmp/nvidia_driver -s --no-drm NVIDIA_INSTALLATION_STATUS=$? RESET_GPU=0 if [ "$NVIDIA_INSTALLATION_STATUS" -ne 0 ]; then sudo cat /var/log/nvidia-installer.log # Fail to install NVIDIA driver, try to reset the GPU RESET_GPU=1 elif [ -x "$(command -v nvidia-smi)" ]; then # Check again if nvidia-smi works even if the driver installation completes successfully INSTALLED_DRIVER_VERSION=$(nvidia-smi --query-gpu=driver_version --format=csv,noheader --id=0) NVIDIA_SMI_STATUS=$? if [ "$NVIDIA_SMI_STATUS" -ne 0 ] && [ "$NVIDIA_SMI_STATUS" -ne 14 ]; then RESET_GPU=1 fi fi if [ "$RESET_GPU" -eq 1 ]; then NVIDIA_DEVICES=$(lspci -D | grep -i NVIDIA | cut -d' ' -f1) # The GPU can get stuck in a failure state if somehow the test crashs the GPU microcode. When this # happens, we'll try to reset all NVIDIA devices https://github.com/pytorch/pytorch/issues/88388 for PCI_ID in $NVIDIA_DEVICES; do DEVICE_ENABLED=$(cat /sys/bus/pci/devices/$PCI_ID/enable) echo "Reseting $PCI_ID (enabled state: $DEVICE_ENABLED)" # This requires sudo permission of course echo "1" | sudo tee /sys/bus/pci/devices/$PCI_ID/reset sleep 1 done fi sudo rm -fv /tmp/nvidia_driver set -e fi ) } post_install_nvidia_driver_common() { ( sudo modprobe nvidia || true echo "After installing NVIDIA driver" lspci lsmod modinfo nvidia || true ( set +e nvidia-smi # NB: Annoyingly, nvidia-smi command returns successfully with return code 0 even in # the case where the driver has already crashed as it still can get the driver version # and some basic information like the bus ID. However, the rest of the information # would be missing (ERR!), for example: # # +-----------------------------------------------------------------------------+ # | NVIDIA-SMI 525.89.02 Driver Version: 525.89.02 CUDA Version: 12.0 | # |-------------------------------+----------------------+----------------------+ # | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC | # | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. | # | | | MIG M. | # |===============================+======================+======================| # | 0 ERR! Off | 00000000:00:1E.0 Off | ERR! | # |ERR! ERR! ERR! ERR! / ERR! | 4184MiB / 23028MiB | ERR! Default | # | | | ERR! | # +-------------------------------+----------------------+----------------------+ # # +-----------------------------------------------------------------------------+ # | Processes: | # | GPU GI CI PID Type Process name GPU Memory | # | ID ID Usage | # |=============================================================================| # +-----------------------------------------------------------------------------+ # # This should be reported as a failure instead as it will guarantee to fail when # Docker tries to run with --gpus all # # So, the correct check here is to query one of the missing piece of info like # GPU name, so that the command can fail accordingly nvidia-smi --query-gpu=gpu_name --format=csv,noheader --id=0 NVIDIA_SMI_STATUS=$? # Allowable exit statuses for nvidia-smi, see: https://github.com/NVIDIA/gpu-operator/issues/285 if [ "$NVIDIA_SMI_STATUS" -eq 0 ] || [ "$NVIDIA_SMI_STATUS" -eq 14 ]; then echo "INFO: Ignoring allowed status ${NVIDIA_SMI_STATUS}" else echo "ERROR: nvidia-smi exited with unresolved status ${NVIDIA_SMI_STATUS}" exit ${NVIDIA_SMI_STATUS} fi set -e ) ) } install_nvidia_driver_amzn2() { ( set -x pre_install_nvidia_driver_amzn2 install_nvidia_driver_common post_install_nvidia_driver_common ) } install_nvidia_driver_ubuntu20() { ( set -x install_nvidia_driver_common post_install_nvidia_driver_common ) } echo "== Installing nvidia driver ${DRIVER_FN} ==" case "${DISTRIBUTION}" in amzn*) install_nvidia_driver_amzn2 ;; ubuntu20.04) install_nvidia_driver_ubuntu20 ;; *) echo "ERROR: Unknown distribution ${DISTRIBUTION}" exit 1 ;; esac # Install container toolkit based on distribution echo "== Installing nvidia container toolkit for ${DISTRIBUTION} ==" case "${DISTRIBUTION}" in amzn*) install_nvidia_docker2_amzn2 ;; ubuntu20.04) install_nvidia_docker2_ubuntu20 ;; *) echo "ERROR: Unknown distribution ${DISTRIBUTION}" exit 1 ;; esac echo "GPU_FLAG=--gpus all -e NVIDIA_DRIVER_CAPABILITIES=all" >> "${GITHUB_ENV}" # Fix https://github.com/NVIDIA/nvidia-docker/issues/1648 on runners with # more than one GPUs. This just needs to be run once. The command fails # on subsequent runs and complains that the mode is already on, but that's # ok sudo nvidia-persistenced || true # This should show persistence mode ON nvidia-smi # check if the container-toolkit is correctly installed and CUDA is available inside a container docker run --rm -t --gpus=all public.ecr.aws/docker/library/python:3.13 nvidia-smi 2025-09-09T14:08:06.1324957Z retry_wait_seconds: 10 2025-09-09T14:08:06.1325261Z polling_interval_seconds: 1 2025-09-09T14:08:06.1325538Z warning_on_retry: true 2025-09-09T14:08:06.1325786Z continue_on_error: false 2025-09-09T14:08:06.1326040Z env: 2025-09-09T14:08:06.1326289Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:08:06.1326731Z REPOSITORY: pytorch/ao 2025-09-09T14:08:06.1326982Z PR_NUMBER: 2963 2025-09-09T14:08:06.1328740Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:08:06.1330685Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:08:06.1331241Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:08:06.1331745Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:08:06.1332172Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:08:06.1332519Z DRIVER_VERSION: 580.65.06 2025-09-09T14:08:06.1332770Z ##[endgroup] 2025-09-09T14:08:06.2537684Z == Installing nvidia driver NVIDIA-Linux-x86_64-580.65.06.run == 2025-09-09T14:08:06.2539221Z + pre_install_nvidia_driver_amzn2 2025-09-09T14:08:06.2542266Z + sudo yum remove -y nvidia-driver-latest-dkms 2025-09-09T14:08:06.5912368Z No match for argument: nvidia-driver-latest-dkms 2025-09-09T14:08:06.5912758Z No packages marked for removal. 2025-09-09T14:08:06.5963556Z Dependencies resolved. 2025-09-09T14:08:06.5973913Z Nothing to do. 2025-09-09T14:08:06.5974347Z Complete! 2025-09-09T14:08:06.7232489Z + install_nvidia_driver_common 2025-09-09T14:08:06.7237762Z + echo 'Before installing NVIDIA driver' 2025-09-09T14:08:06.7238054Z + lspci 2025-09-09T14:08:06.7239776Z Before installing NVIDIA driver 2025-09-09T14:08:06.7374826Z 00:00.0 Host bridge: Intel Corporation 440FX - 82441FX PMC [Natoma] 2025-09-09T14:08:06.7375311Z 00:01.0 ISA bridge: Intel Corporation 82371SB PIIX3 ISA [Natoma/Triton II] 2025-09-09T14:08:06.7375986Z 00:01.3 Non-VGA unclassified device: Intel Corporation 82371AB/EB/MB PIIX4 ACPI (rev 08) 2025-09-09T14:08:06.7376500Z 00:03.0 VGA compatible controller: Amazon.com, Inc. Device 1111 2025-09-09T14:08:06.7376970Z 00:04.0 Non-Volatile memory controller: Amazon.com, Inc. NVMe EBS Controller 2025-09-09T14:08:06.7377507Z 00:05.0 Ethernet controller: Amazon.com, Inc. Elastic Network Adapter (ENA) 2025-09-09T14:08:06.7377978Z 00:1b.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:08:06.7378403Z 00:1c.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:08:06.7378825Z 00:1d.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:08:06.7379243Z 00:1e.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:08:06.7379717Z 00:1f.0 Non-Volatile memory controller: Amazon.com, Inc. NVMe SSD Controller 2025-09-09T14:08:06.7380111Z + lsmod 2025-09-09T14:08:06.7434881Z Module Size Used by 2025-09-09T14:08:06.7435168Z veth 36864 0 2025-09-09T14:08:06.7435457Z nvidia_modeset 1740800 0 2025-09-09T14:08:06.7435754Z video 65536 1 nvidia_modeset 2025-09-09T14:08:06.7436047Z wmi 36864 1 video 2025-09-09T14:08:06.7436310Z nvidia_uvm 1921024 0 2025-09-09T14:08:06.7436605Z nvidia 14274560 19 nvidia_uvm,nvidia_modeset 2025-09-09T14:08:06.7436958Z drm 602112 1 nvidia 2025-09-09T14:08:06.7437251Z drm_panel_orientation_quirks 32768 1 drm 2025-09-09T14:08:06.7437619Z backlight 24576 3 video,drm,nvidia_modeset 2025-09-09T14:08:06.7437962Z i2c_core 110592 2 nvidia,drm 2025-09-09T14:08:06.7438250Z xt_conntrack 16384 1 2025-09-09T14:08:06.7438507Z nft_chain_nat 16384 3 2025-09-09T14:08:06.7438755Z xt_MASQUERADE 20480 1 2025-09-09T14:08:06.7439049Z nf_nat 57344 2 nft_chain_nat,xt_MASQUERADE 2025-09-09T14:08:06.7439373Z nf_conntrack_netlink 57344 0 2025-09-09T14:08:06.7439999Z nf_conntrack 184320 4 xt_conntrack,nf_nat,nf_conntrack_netlink,xt_MASQUERADE 2025-09-09T14:08:06.7440432Z nf_defrag_ipv6 24576 1 nf_conntrack 2025-09-09T14:08:06.7440742Z nf_defrag_ipv4 16384 1 nf_conntrack 2025-09-09T14:08:06.7441024Z xfrm_user 57344 1 2025-09-09T14:08:06.7441284Z xfrm_algo 16384 1 xfrm_user 2025-09-09T14:08:06.7441566Z xt_addrtype 16384 2 2025-09-09T14:08:06.7441814Z nft_compat 20480 4 2025-09-09T14:08:06.7442113Z nf_tables 311296 57 nft_compat,nft_chain_nat 2025-09-09T14:08:06.7442517Z nfnetlink 20480 4 nft_compat,nf_conntrack_netlink,nf_tables 2025-09-09T14:08:06.7442892Z br_netfilter 36864 0 2025-09-09T14:08:06.7443161Z bridge 323584 1 br_netfilter 2025-09-09T14:08:06.7443449Z stp 16384 1 bridge 2025-09-09T14:08:06.7443725Z llc 16384 2 bridge,stp 2025-09-09T14:08:06.7444014Z overlay 167936 0 2025-09-09T14:08:06.7444260Z tls 139264 0 2025-09-09T14:08:06.7444503Z nls_ascii 16384 1 2025-09-09T14:08:06.7444753Z nls_cp437 20480 1 2025-09-09T14:08:06.7444994Z vfat 24576 1 2025-09-09T14:08:06.7445240Z fat 86016 1 vfat 2025-09-09T14:08:06.7445500Z ena 180224 0 2025-09-09T14:08:06.7445783Z i8042 45056 0 2025-09-09T14:08:06.7446023Z sunrpc 700416 1 2025-09-09T14:08:06.7446277Z serio 28672 3 i8042 2025-09-09T14:08:06.7446672Z button 24576 0 2025-09-09T14:08:06.7446926Z ghash_clmulni_intel 16384 0 2025-09-09T14:08:06.7447190Z sch_fq_codel 20480 33 2025-09-09T14:08:06.7447438Z dm_mod 188416 0 2025-09-09T14:08:06.7447687Z fuse 184320 1 2025-09-09T14:08:06.7447924Z loop 36864 0 2025-09-09T14:08:06.7448175Z configfs 57344 1 2025-09-09T14:08:06.7448426Z dmi_sysfs 20480 0 2025-09-09T14:08:06.7448676Z crc32_pclmul 16384 0 2025-09-09T14:08:06.7448922Z crc32c_intel 24576 0 2025-09-09T14:08:06.7449183Z efivarfs 24576 1 2025-09-09T14:08:06.7449423Z + modinfo nvidia 2025-09-09T14:08:06.7460724Z filename: /lib/modules/6.1.141-155.222.amzn2023.x86_64/kernel/drivers/video/nvidia.ko 2025-09-09T14:08:06.7461182Z import_ns: DMA_BUF 2025-09-09T14:08:06.7461423Z alias: char-major-195-* 2025-09-09T14:08:06.7461692Z version: 580.65.06 2025-09-09T14:08:06.7461937Z supported: external 2025-09-09T14:08:06.7462188Z license: Dual MIT/GPL 2025-09-09T14:08:06.7462469Z firmware: nvidia/580.65.06/gsp_tu10x.bin 2025-09-09T14:08:06.7462803Z firmware: nvidia/580.65.06/gsp_ga10x.bin 2025-09-09T14:08:06.7463124Z srcversion: A69EBF72FC9D60E11E9A05C 2025-09-09T14:08:06.7463448Z alias: of:N*T*Cnvidia,tegra264-displayC* 2025-09-09T14:08:06.7463801Z alias: of:N*T*Cnvidia,tegra264-display 2025-09-09T14:08:06.7464140Z alias: of:N*T*Cnvidia,tegra234-displayC* 2025-09-09T14:08:06.7464486Z alias: of:N*T*Cnvidia,tegra234-display 2025-09-09T14:08:06.7464818Z alias: pci:v000010DEd*sv*sd*bc06sc80i00* 2025-09-09T14:08:06.7465172Z alias: pci:v000010DEd*sv*sd*bc03sc02i00* 2025-09-09T14:08:06.7465521Z alias: pci:v000010DEd*sv*sd*bc03sc00i00* 2025-09-09T14:08:06.7465826Z depends: i2c-core,drm 2025-09-09T14:08:06.7466077Z retpoline: Y 2025-09-09T14:08:06.7466288Z name: nvidia 2025-09-09T14:08:06.7466644Z vermagic: 6.1.141-155.222.amzn2023.x86_64 SMP preempt mod_unload modversions 2025-09-09T14:08:06.7467108Z parm: NvSwitchRegDwords:NvSwitch regkey (charp) 2025-09-09T14:08:06.7467553Z parm: NvSwitchBlacklist:NvSwitchBlacklist=uuid[,uuid...] (charp) 2025-09-09T14:08:06.7467963Z parm: NVreg_ResmanDebugLevel:int 2025-09-09T14:08:06.7468271Z parm: NVreg_RmLogonRC:int 2025-09-09T14:08:06.7468669Z parm: NVreg_ModifyDeviceFiles:int 2025-09-09T14:08:06.7468982Z parm: NVreg_DeviceFileUID:int 2025-09-09T14:08:06.7469283Z parm: NVreg_DeviceFileGID:int 2025-09-09T14:08:06.7469583Z parm: NVreg_DeviceFileMode:int 2025-09-09T14:08:06.7469940Z parm: NVreg_InitializeSystemMemoryAllocations:int 2025-09-09T14:08:06.7470316Z parm: NVreg_UsePageAttributeTable:int 2025-09-09T14:08:06.7470643Z parm: NVreg_EnablePCIeGen3:int 2025-09-09T14:08:06.7470933Z parm: NVreg_EnableMSI:int 2025-09-09T14:08:06.7471238Z parm: NVreg_EnableStreamMemOPs:int 2025-09-09T14:08:06.7471591Z parm: NVreg_RestrictProfilingToAdminUsers:int 2025-09-09T14:08:06.7471984Z parm: NVreg_PreserveVideoMemoryAllocations:int 2025-09-09T14:08:06.7472360Z parm: NVreg_EnableS0ixPowerManagement:int 2025-09-09T14:08:06.7472762Z parm: NVreg_S0ixPowerManagementVideoMemoryThreshold:int 2025-09-09T14:08:06.7473176Z parm: NVreg_DynamicPowerManagement:int 2025-09-09T14:08:06.7473584Z parm: NVreg_DynamicPowerManagementVideoMemoryThreshold:int 2025-09-09T14:08:06.7473988Z parm: NVreg_EnableGpuFirmware:int 2025-09-09T14:08:06.7474316Z parm: NVreg_EnableGpuFirmwareLogs:int 2025-09-09T14:08:06.7474680Z parm: NVreg_OpenRmEnableUnsupportedGpus:int 2025-09-09T14:08:06.7475050Z parm: NVreg_EnableUserNUMAManagement:int 2025-09-09T14:08:06.7475399Z parm: NVreg_MemoryPoolSize:int 2025-09-09T14:08:06.7475745Z parm: NVreg_KMallocHeapMaxSize:int 2025-09-09T14:08:06.7476172Z parm: NVreg_VMallocHeapMaxSize:int 2025-09-09T14:08:06.7476500Z parm: NVreg_IgnoreMMIOCheck:int 2025-09-09T14:08:06.7476808Z parm: NVreg_NvLinkDisable:int 2025-09-09T14:08:06.7477148Z parm: NVreg_EnablePCIERelaxedOrderingMode:int 2025-09-09T14:08:06.7477510Z parm: NVreg_RegisterPCIDriver:int 2025-09-09T14:08:06.7477859Z parm: NVreg_RegisterPlatformDeviceDriver:int 2025-09-09T14:08:06.7478217Z parm: NVreg_EnableResizableBar:int 2025-09-09T14:08:06.7478547Z parm: NVreg_EnableDbgBreakpoint:int 2025-09-09T14:08:06.7478892Z parm: NVreg_EnableNonblockingOpen:int 2025-09-09T14:08:06.7479240Z parm: NVreg_CoherentGPUMemoryMode:charp 2025-09-09T14:08:06.7479583Z parm: NVreg_RegistryDwords:charp 2025-09-09T14:08:06.7479917Z parm: NVreg_RegistryDwordsPerDevice:charp 2025-09-09T14:08:06.7480255Z parm: NVreg_RmMsg:charp 2025-09-09T14:08:06.7480550Z parm: NVreg_GpuBlacklist:charp 2025-09-09T14:08:06.7480872Z parm: NVreg_TemporaryFilePath:charp 2025-09-09T14:08:06.7481196Z parm: NVreg_ExcludedGpus:charp 2025-09-09T14:08:06.7481504Z parm: NVreg_DmaRemapPeerMmio:int 2025-09-09T14:08:06.7481829Z parm: NVreg_RmNvlinkBandwidth:charp 2025-09-09T14:08:06.7482183Z parm: NVreg_RmNvlinkBandwidthLinkCount:int 2025-09-09T14:08:06.7482540Z parm: NVreg_ImexChannelCount:int 2025-09-09T14:08:06.7482858Z parm: NVreg_CreateImexChannel0:int 2025-09-09T14:08:06.7483203Z parm: NVreg_GrdmaPciTopoCheckOverride:int 2025-09-09T14:08:06.7483542Z parm: rm_firmware_active:charp 2025-09-09T14:08:06.7483821Z + HAS_NVIDIA_DRIVER=0 2025-09-09T14:08:06.7484059Z ++ command -v nvidia-smi 2025-09-09T14:08:06.7484313Z + '[' -x /usr/bin/nvidia-smi ']' 2025-09-09T14:08:06.7484573Z + set +e 2025-09-09T14:08:06.7484879Z ++ nvidia-smi --query-gpu=driver_version --format=csv,noheader --id=0 2025-09-09T14:08:06.8217264Z + INSTALLED_DRIVER_VERSION=580.65.06 2025-09-09T14:08:06.8217828Z + NVIDIA_SMI_STATUS=0 2025-09-09T14:08:06.8218268Z + '[' 0 -ne 0 ']' 2025-09-09T14:08:06.8218693Z + '[' 580.65.06 '!=' 580.65.06 ']' 2025-09-09T14:08:06.8219187Z + HAS_NVIDIA_DRIVER=1 2025-09-09T14:08:06.8220009Z + echo 'NVIDIA driver (580.65.06) has already been installed. Skipping NVIDIA driver installation' 2025-09-09T14:08:06.8220895Z + set -e 2025-09-09T14:08:06.8221550Z + '[' 1 -eq 0 ']' 2025-09-09T14:08:06.8222282Z NVIDIA driver (580.65.06) has already been installed. Skipping NVIDIA driver installation 2025-09-09T14:08:06.8223178Z + post_install_nvidia_driver_common 2025-09-09T14:08:06.8225166Z + sudo modprobe nvidia 2025-09-09T14:08:06.9398137Z + echo 'After installing NVIDIA driver' 2025-09-09T14:08:06.9398709Z + lspci 2025-09-09T14:08:06.9399121Z After installing NVIDIA driver 2025-09-09T14:08:06.9530542Z 00:00.0 Host bridge: Intel Corporation 440FX - 82441FX PMC [Natoma] 2025-09-09T14:08:06.9531021Z 00:01.0 ISA bridge: Intel Corporation 82371SB PIIX3 ISA [Natoma/Triton II] 2025-09-09T14:08:06.9531578Z 00:01.3 Non-VGA unclassified device: Intel Corporation 82371AB/EB/MB PIIX4 ACPI (rev 08) 2025-09-09T14:08:06.9532098Z 00:03.0 VGA compatible controller: Amazon.com, Inc. Device 1111 2025-09-09T14:08:06.9532570Z 00:04.0 Non-Volatile memory controller: Amazon.com, Inc. NVMe EBS Controller 2025-09-09T14:08:06.9533103Z 00:05.0 Ethernet controller: Amazon.com, Inc. Elastic Network Adapter (ENA) 2025-09-09T14:08:06.9533571Z 00:1b.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:08:06.9533995Z 00:1c.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:08:06.9534404Z 00:1d.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:08:06.9534830Z 00:1e.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:08:06.9535302Z 00:1f.0 Non-Volatile memory controller: Amazon.com, Inc. NVMe SSD Controller 2025-09-09T14:08:06.9535736Z + lsmod 2025-09-09T14:08:06.9575267Z Module Size Used by 2025-09-09T14:08:06.9575611Z veth 36864 0 2025-09-09T14:08:06.9575969Z nvidia_modeset 1740800 0 2025-09-09T14:08:06.9576242Z video 65536 1 nvidia_modeset 2025-09-09T14:08:06.9576538Z wmi 36864 1 video 2025-09-09T14:08:06.9576808Z nvidia_uvm 1921024 0 2025-09-09T14:08:06.9577103Z nvidia 14274560 19 nvidia_uvm,nvidia_modeset 2025-09-09T14:08:06.9577426Z drm 602112 1 nvidia 2025-09-09T14:08:06.9577724Z drm_panel_orientation_quirks 32768 1 drm 2025-09-09T14:08:06.9578089Z backlight 24576 3 video,drm,nvidia_modeset 2025-09-09T14:08:06.9578434Z i2c_core 110592 2 nvidia,drm 2025-09-09T14:08:06.9578720Z xt_conntrack 16384 1 2025-09-09T14:08:06.9578971Z nft_chain_nat 16384 3 2025-09-09T14:08:06.9579227Z xt_MASQUERADE 20480 1 2025-09-09T14:08:06.9579519Z nf_nat 57344 2 nft_chain_nat,xt_MASQUERADE 2025-09-09T14:08:06.9579845Z nf_conntrack_netlink 57344 0 2025-09-09T14:08:06.9580236Z nf_conntrack 184320 4 xt_conntrack,nf_nat,nf_conntrack_netlink,xt_MASQUERADE 2025-09-09T14:08:06.9580666Z nf_defrag_ipv6 24576 1 nf_conntrack 2025-09-09T14:08:06.9580993Z nf_defrag_ipv4 16384 1 nf_conntrack 2025-09-09T14:08:06.9581284Z xfrm_user 57344 1 2025-09-09T14:08:06.9581544Z xfrm_algo 16384 1 xfrm_user 2025-09-09T14:08:06.9581827Z xt_addrtype 16384 2 2025-09-09T14:08:06.9582079Z nft_compat 20480 4 2025-09-09T14:08:06.9582377Z nf_tables 311296 57 nft_compat,nft_chain_nat 2025-09-09T14:08:06.9582777Z nfnetlink 20480 4 nft_compat,nf_conntrack_netlink,nf_tables 2025-09-09T14:08:06.9583148Z br_netfilter 36864 0 2025-09-09T14:08:06.9583415Z bridge 323584 1 br_netfilter 2025-09-09T14:08:06.9583709Z stp 16384 1 bridge 2025-09-09T14:08:06.9583986Z llc 16384 2 bridge,stp 2025-09-09T14:08:06.9584272Z overlay 167936 0 2025-09-09T14:08:06.9584519Z tls 139264 0 2025-09-09T14:08:06.9584757Z nls_ascii 16384 1 2025-09-09T14:08:06.9585007Z nls_cp437 20480 1 2025-09-09T14:08:06.9585243Z vfat 24576 1 2025-09-09T14:08:06.9585492Z fat 86016 1 vfat 2025-09-09T14:08:06.9585895Z ena 180224 0 2025-09-09T14:08:06.9586142Z i8042 45056 0 2025-09-09T14:08:06.9586377Z sunrpc 700416 1 2025-09-09T14:08:06.9586629Z serio 28672 3 i8042 2025-09-09T14:08:06.9586905Z button 24576 0 2025-09-09T14:08:06.9587153Z ghash_clmulni_intel 16384 0 2025-09-09T14:08:06.9587409Z sch_fq_codel 20480 33 2025-09-09T14:08:06.9587663Z dm_mod 188416 0 2025-09-09T14:08:06.9587908Z fuse 184320 1 2025-09-09T14:08:06.9588144Z loop 36864 0 2025-09-09T14:08:06.9588398Z configfs 57344 1 2025-09-09T14:08:06.9588639Z dmi_sysfs 20480 0 2025-09-09T14:08:06.9588895Z crc32_pclmul 16384 0 2025-09-09T14:08:06.9589143Z crc32c_intel 24576 0 2025-09-09T14:08:06.9589392Z efivarfs 24576 1 2025-09-09T14:08:06.9589637Z + modinfo nvidia 2025-09-09T14:08:06.9597013Z filename: /lib/modules/6.1.141-155.222.amzn2023.x86_64/kernel/drivers/video/nvidia.ko 2025-09-09T14:08:06.9611211Z import_ns: DMA_BUF 2025-09-09T14:08:06.9611502Z alias: char-major-195-* 2025-09-09T14:08:06.9611779Z version: 580.65.06 2025-09-09T14:08:06.9612039Z supported: external 2025-09-09T14:08:06.9612284Z license: Dual MIT/GPL 2025-09-09T14:08:06.9612594Z firmware: nvidia/580.65.06/gsp_tu10x.bin 2025-09-09T14:08:06.9612937Z firmware: nvidia/580.65.06/gsp_ga10x.bin 2025-09-09T14:08:06.9613286Z srcversion: A69EBF72FC9D60E11E9A05C 2025-09-09T14:08:06.9613624Z alias: of:N*T*Cnvidia,tegra264-displayC* 2025-09-09T14:08:06.9614194Z alias: of:N*T*Cnvidia,tegra264-display 2025-09-09T14:08:06.9614575Z alias: of:N*T*Cnvidia,tegra234-displayC* 2025-09-09T14:08:06.9614930Z alias: of:N*T*Cnvidia,tegra234-display 2025-09-09T14:08:06.9615285Z alias: pci:v000010DEd*sv*sd*bc06sc80i00* 2025-09-09T14:08:06.9615619Z alias: pci:v000010DEd*sv*sd*bc03sc02i00* 2025-09-09T14:08:06.9616034Z alias: pci:v000010DEd*sv*sd*bc03sc00i00* 2025-09-09T14:08:06.9616342Z depends: i2c-core,drm 2025-09-09T14:08:06.9616596Z retpoline: Y 2025-09-09T14:08:06.9616814Z name: nvidia 2025-09-09T14:08:06.9617172Z vermagic: 6.1.141-155.222.amzn2023.x86_64 SMP preempt mod_unload modversions 2025-09-09T14:08:06.9617644Z parm: NvSwitchRegDwords:NvSwitch regkey (charp) 2025-09-09T14:08:06.9618083Z parm: NvSwitchBlacklist:NvSwitchBlacklist=uuid[,uuid...] (charp) 2025-09-09T14:08:06.9618502Z parm: NVreg_ResmanDebugLevel:int 2025-09-09T14:08:06.9618812Z parm: NVreg_RmLogonRC:int 2025-09-09T14:08:06.9619115Z parm: NVreg_ModifyDeviceFiles:int 2025-09-09T14:08:06.9619424Z parm: NVreg_DeviceFileUID:int 2025-09-09T14:08:06.9619730Z parm: NVreg_DeviceFileGID:int 2025-09-09T14:08:06.9620030Z parm: NVreg_DeviceFileMode:int 2025-09-09T14:08:06.9620409Z parm: NVreg_InitializeSystemMemoryAllocations:int 2025-09-09T14:08:06.9620811Z parm: NVreg_UsePageAttributeTable:int 2025-09-09T14:08:06.9621136Z parm: NVreg_EnablePCIeGen3:int 2025-09-09T14:08:06.9621438Z parm: NVreg_EnableMSI:int 2025-09-09T14:08:06.9621735Z parm: NVreg_EnableStreamMemOPs:int 2025-09-09T14:08:06.9622102Z parm: NVreg_RestrictProfilingToAdminUsers:int 2025-09-09T14:08:06.9622488Z parm: NVreg_PreserveVideoMemoryAllocations:int 2025-09-09T14:08:06.9622867Z parm: NVreg_EnableS0ixPowerManagement:int 2025-09-09T14:08:06.9623269Z parm: NVreg_S0ixPowerManagementVideoMemoryThreshold:int 2025-09-09T14:08:06.9623678Z parm: NVreg_DynamicPowerManagement:int 2025-09-09T14:08:06.9624089Z parm: NVreg_DynamicPowerManagementVideoMemoryThreshold:int 2025-09-09T14:08:06.9624486Z parm: NVreg_EnableGpuFirmware:int 2025-09-09T14:08:06.9624851Z parm: NVreg_EnableGpuFirmwareLogs:int 2025-09-09T14:08:06.9625380Z parm: NVreg_OpenRmEnableUnsupportedGpus:int 2025-09-09T14:08:06.9625779Z parm: NVreg_EnableUserNUMAManagement:int 2025-09-09T14:08:06.9626113Z parm: NVreg_MemoryPoolSize:int 2025-09-09T14:08:06.9626430Z parm: NVreg_KMallocHeapMaxSize:int 2025-09-09T14:08:06.9626758Z parm: NVreg_VMallocHeapMaxSize:int 2025-09-09T14:08:06.9627069Z parm: NVreg_IgnoreMMIOCheck:int 2025-09-09T14:08:06.9627383Z parm: NVreg_NvLinkDisable:int 2025-09-09T14:08:06.9627722Z parm: NVreg_EnablePCIERelaxedOrderingMode:int 2025-09-09T14:08:06.9628095Z parm: NVreg_RegisterPCIDriver:int 2025-09-09T14:08:06.9628448Z parm: NVreg_RegisterPlatformDeviceDriver:int 2025-09-09T14:08:06.9628813Z parm: NVreg_EnableResizableBar:int 2025-09-09T14:08:06.9629142Z parm: NVreg_EnableDbgBreakpoint:int 2025-09-09T14:08:06.9629483Z parm: NVreg_EnableNonblockingOpen:int 2025-09-09T14:08:06.9629834Z parm: NVreg_CoherentGPUMemoryMode:charp 2025-09-09T14:08:06.9630174Z parm: NVreg_RegistryDwords:charp 2025-09-09T14:08:06.9630512Z parm: NVreg_RegistryDwordsPerDevice:charp 2025-09-09T14:08:06.9630839Z parm: NVreg_RmMsg:charp 2025-09-09T14:08:06.9631132Z parm: NVreg_GpuBlacklist:charp 2025-09-09T14:08:06.9631451Z parm: NVreg_TemporaryFilePath:charp 2025-09-09T14:08:06.9631783Z parm: NVreg_ExcludedGpus:charp 2025-09-09T14:08:06.9632092Z parm: NVreg_DmaRemapPeerMmio:int 2025-09-09T14:08:06.9632423Z parm: NVreg_RmNvlinkBandwidth:charp 2025-09-09T14:08:06.9632879Z parm: NVreg_RmNvlinkBandwidthLinkCount:int 2025-09-09T14:08:06.9633227Z parm: NVreg_ImexChannelCount:int 2025-09-09T14:08:06.9633552Z parm: NVreg_CreateImexChannel0:int 2025-09-09T14:08:06.9633897Z parm: NVreg_GrdmaPciTopoCheckOverride:int 2025-09-09T14:08:06.9634238Z parm: rm_firmware_active:charp 2025-09-09T14:08:06.9634511Z + set +e 2025-09-09T14:08:06.9634710Z + nvidia-smi 2025-09-09T14:08:07.0042285Z Tue Sep 9 14:08:06 2025 2025-09-09T14:08:07.0042638Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:08:07.0043148Z | NVIDIA-SMI 580.65.06 Driver Version: 580.65.06 CUDA Version: 13.0 | 2025-09-09T14:08:07.0043627Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:08:07.0044123Z | GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC | 2025-09-09T14:08:07.0044639Z | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | 2025-09-09T14:08:07.0045080Z | | | MIG M. | 2025-09-09T14:08:07.0045421Z |=========================================+========================+======================| 2025-09-09T14:08:07.0649718Z | 0 NVIDIA A10G On | 00000000:00:1B.0 Off | 0 | 2025-09-09T14:08:07.0650200Z | 0% 23C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:08:07.0650573Z | | | N/A | 2025-09-09T14:08:07.0650971Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:08:07.0651407Z | 1 NVIDIA A10G On | 00000000:00:1C.0 Off | 0 | 2025-09-09T14:08:07.0651816Z | 0% 22C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:08:07.0652195Z | | | N/A | 2025-09-09T14:08:07.0652572Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:08:07.0653005Z | 2 NVIDIA A10G On | 00000000:00:1D.0 Off | 0 | 2025-09-09T14:08:07.0653813Z | 0% 22C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:08:07.0654226Z | | | N/A | 2025-09-09T14:08:07.0654669Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:08:07.0655176Z | 3 NVIDIA A10G On | 00000000:00:1E.0 Off | 0 | 2025-09-09T14:08:07.0655692Z | 0% 21C P8 11W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:08:07.0656164Z | | | N/A | 2025-09-09T14:08:07.0656541Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:08:07.0676454Z 2025-09-09T14:08:07.0676899Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:08:07.0677326Z | Processes: | 2025-09-09T14:08:07.0677771Z | GPU GI CI PID Type Process name GPU Memory | 2025-09-09T14:08:07.0678184Z | ID ID Usage | 2025-09-09T14:08:07.0678515Z |=========================================================================================| 2025-09-09T14:08:07.0716079Z | No running processes found | 2025-09-09T14:08:07.0717008Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:08:08.2167722Z + nvidia-smi --query-gpu=gpu_name --format=csv,noheader --id=0 2025-09-09T14:08:08.2349912Z NVIDIA A10G 2025-09-09T14:08:08.2620658Z + NVIDIA_SMI_STATUS=0 2025-09-09T14:08:08.2621180Z + '[' 0 -eq 0 ']' 2025-09-09T14:08:08.2621658Z + echo 'INFO: Ignoring allowed status 0' 2025-09-09T14:08:08.2622227Z + set -e 2025-09-09T14:08:08.2622669Z INFO: Ignoring allowed status 0 2025-09-09T14:08:08.2631822Z == Installing nvidia container toolkit for amzn2023 == 2025-09-09T14:08:08.2636329Z + sudo yum install -y yum-utils 2025-09-09T14:08:08.7161399Z Last metadata expiration check: 0:03:42 ago on Tue Sep 9 14:04:26 2025. 2025-09-09T14:08:08.7403597Z Package dnf-utils-4.3.0-13.amzn2023.0.5.noarch is already installed. 2025-09-09T14:08:08.7861277Z Dependencies resolved. 2025-09-09T14:08:08.8086928Z Nothing to do. 2025-09-09T14:08:08.8087147Z Complete! 2025-09-09T14:08:08.9521799Z + [[ amzn2023 == \a\m\z\n\2\0\2\3 ]] 2025-09-09T14:08:08.9522371Z + YUM_REPO_URL=https://nvidia.github.io/libnvidia-container/stable/rpm/nvidia-container-toolkit.repo 2025-09-09T14:08:08.9523240Z + sudo yum-config-manager --add-repo https://nvidia.github.io/libnvidia-container/stable/rpm/nvidia-container-toolkit.repo 2025-09-09T14:08:09.2533898Z Adding repo from: https://nvidia.github.io/libnvidia-container/stable/rpm/nvidia-container-toolkit.repo 2025-09-09T14:08:09.3049866Z + sudo yum install -y nvidia-container-toolkit-1.17.8 libnvidia-container-tools-1.17.8 libnvidia-container1-1.17.8 nvidia-container-toolkit-base-1.17.8 2025-09-09T14:08:09.8197588Z nvidia-container-toolkit 18 kB/s | 833 B 00:00 2025-09-09T14:08:09.8449183Z Package nvidia-container-toolkit-1.17.8-1.x86_64 is already installed. 2025-09-09T14:08:09.8455969Z Package libnvidia-container-tools-1.17.8-1.x86_64 is already installed. 2025-09-09T14:08:09.8459393Z Package libnvidia-container1-1.17.8-1.x86_64 is already installed. 2025-09-09T14:08:09.8466338Z Package nvidia-container-toolkit-base-1.17.8-1.x86_64 is already installed. 2025-09-09T14:08:09.8942972Z Dependencies resolved. 2025-09-09T14:08:09.9168144Z Nothing to do. 2025-09-09T14:08:09.9168699Z Complete! 2025-09-09T14:08:10.0189439Z + sudo systemctl restart docker 2025-09-09T14:08:48.5277922Z nvidia-persistenced failed to initialize. Check syslog for more details. 2025-09-09T14:08:48.5739105Z Tue Sep 9 14:08:48 2025 2025-09-09T14:08:48.5739822Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:08:48.5740333Z | NVIDIA-SMI 580.65.06 Driver Version: 580.65.06 CUDA Version: 13.0 | 2025-09-09T14:08:48.5740813Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:08:48.5741297Z | GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC | 2025-09-09T14:08:48.5741820Z | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | 2025-09-09T14:08:48.5742255Z | | | MIG M. | 2025-09-09T14:08:48.5742581Z |=========================================+========================+======================| 2025-09-09T14:08:48.6340264Z | 0 NVIDIA A10G On | 00000000:00:1B.0 Off | 0 | 2025-09-09T14:08:48.6340721Z | 0% 23C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:08:48.6341093Z | | | N/A | 2025-09-09T14:08:48.6341480Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:08:48.6341914Z | 1 NVIDIA A10G On | 00000000:00:1C.0 Off | 0 | 2025-09-09T14:08:48.6342320Z | 0% 22C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:08:48.6342684Z | | | N/A | 2025-09-09T14:08:48.6343262Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:08:48.6343684Z | 2 NVIDIA A10G On | 00000000:00:1D.0 Off | 0 | 2025-09-09T14:08:48.6344096Z | 0% 22C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:08:48.6344456Z | | | N/A | 2025-09-09T14:08:48.6344842Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:08:48.6345270Z | 3 NVIDIA A10G On | 00000000:00:1E.0 Off | 0 | 2025-09-09T14:08:48.6345677Z | 0% 21C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:08:48.6346044Z | | | N/A | 2025-09-09T14:08:48.6346424Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:08:48.6367168Z 2025-09-09T14:08:48.6367391Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:08:48.6367824Z | Processes: | 2025-09-09T14:08:48.6368263Z | GPU GI CI PID Type Process name GPU Memory | 2025-09-09T14:08:48.6368669Z | ID ID Usage | 2025-09-09T14:08:48.6369001Z |=========================================================================================| 2025-09-09T14:08:48.6406419Z | No running processes found | 2025-09-09T14:08:48.6407306Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:08:49.6922986Z Unable to find image 'public.ecr.aws/docker/library/python:3.13' locally 2025-09-09T14:08:49.9401628Z 3.13: Pulling from docker/library/python 2025-09-09T14:08:50.0429290Z 15b1d8a5ff03: Pulling fs layer 2025-09-09T14:08:50.0429587Z 22718812f617: Pulling fs layer 2025-09-09T14:08:50.0429859Z 401a98f7495b: Pulling fs layer 2025-09-09T14:08:50.0430121Z ad446e7df19a: Pulling fs layer 2025-09-09T14:08:50.0430595Z 5d32990caa16: Pulling fs layer 2025-09-09T14:08:50.0430863Z a79d633abf9a: Pulling fs layer 2025-09-09T14:08:50.0431134Z 249a56c8e466: Pulling fs layer 2025-09-09T14:08:50.0431373Z 5d32990caa16: Waiting 2025-09-09T14:08:50.0431594Z ad446e7df19a: Waiting 2025-09-09T14:08:50.0431812Z a79d633abf9a: Waiting 2025-09-09T14:08:50.0432029Z 249a56c8e466: Waiting 2025-09-09T14:08:50.1736732Z 22718812f617: Verifying Checksum 2025-09-09T14:08:50.1737092Z 22718812f617: Download complete 2025-09-09T14:08:50.2059419Z 15b1d8a5ff03: Download complete 2025-09-09T14:08:50.2819586Z 5d32990caa16: Verifying Checksum 2025-09-09T14:08:50.2819949Z 5d32990caa16: Download complete 2025-09-09T14:08:50.3471060Z 401a98f7495b: Verifying Checksum 2025-09-09T14:08:50.3471344Z 401a98f7495b: Download complete 2025-09-09T14:08:50.4104738Z a79d633abf9a: Verifying Checksum 2025-09-09T14:08:50.4105014Z a79d633abf9a: Download complete 2025-09-09T14:08:50.4189611Z 249a56c8e466: Verifying Checksum 2025-09-09T14:08:50.4189923Z 249a56c8e466: Download complete 2025-09-09T14:08:51.0642259Z ad446e7df19a: Verifying Checksum 2025-09-09T14:08:51.0642866Z ad446e7df19a: Download complete 2025-09-09T14:08:51.9727688Z 15b1d8a5ff03: Pull complete 2025-09-09T14:08:52.7107922Z 22718812f617: Pull complete 2025-09-09T14:08:55.1737989Z 401a98f7495b: Pull complete 2025-09-09T14:09:01.8158510Z ad446e7df19a: Pull complete 2025-09-09T14:09:02.1066746Z 5d32990caa16: Pull complete 2025-09-09T14:09:02.8733799Z a79d633abf9a: Pull complete 2025-09-09T14:09:02.8954647Z 249a56c8e466: Pull complete 2025-09-09T14:09:02.9090938Z Digest: sha256:74503e0bff6cf811f029590a05e0218cc9ba3e099a4b7df0ab84a67df081e1bc 2025-09-09T14:09:02.9133959Z Status: Downloaded newer image for public.ecr.aws/docker/library/python:3.13 2025-09-09T14:09:09.9313408Z Tue Sep 9 14:09:09 2025 2025-09-09T14:09:09.9313814Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:09:09.9314340Z | NVIDIA-SMI 580.65.06 Driver Version: 580.65.06 CUDA Version: 13.0 | 2025-09-09T14:09:09.9314806Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:09:09.9315290Z | GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC | 2025-09-09T14:09:09.9315818Z | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | 2025-09-09T14:09:09.9316234Z | | | MIG M. | 2025-09-09T14:09:09.9316562Z |=========================================+========================+======================| 2025-09-09T14:09:09.9918478Z | 0 NVIDIA A10G On | 00000000:00:1B.0 Off | 0 | 2025-09-09T14:09:09.9918918Z | 0% 23C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:09:09.9919308Z | | | N/A | 2025-09-09T14:09:09.9919700Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:09:09.9920137Z | 1 NVIDIA A10G On | 00000000:00:1C.0 Off | 0 | 2025-09-09T14:09:09.9920544Z | 0% 22C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:09:09.9920907Z | | | N/A | 2025-09-09T14:09:09.9921293Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:09:09.9921722Z | 2 NVIDIA A10G On | 00000000:00:1D.0 Off | 0 | 2025-09-09T14:09:09.9922134Z | 0% 22C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:09:09.9922494Z | | | N/A | 2025-09-09T14:09:09.9923264Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:09:09.9923710Z | 3 NVIDIA A10G On | 00000000:00:1E.0 Off | 0 | 2025-09-09T14:09:09.9924113Z | 0% 22C P8 11W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:09:09.9924486Z | | | N/A | 2025-09-09T14:09:09.9924861Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:09:09.9945717Z 2025-09-09T14:09:09.9946187Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:09:09.9946656Z | Processes: | 2025-09-09T14:09:09.9947091Z | GPU GI CI PID Type Process name GPU Memory | 2025-09-09T14:09:09.9947493Z | ID ID Usage | 2025-09-09T14:09:09.9947830Z |=========================================================================================| 2025-09-09T14:09:09.9981918Z | No running processes found | 2025-09-09T14:09:09.9982389Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:09:12.2398696Z Command completed after 1 attempt(s). 2025-09-09T14:09:12.2513289Z ##[group]Run set -ex 2025-09-09T14:09:12.2513581Z set -ex 2025-09-09T14:09:12.2513788Z { 2025-09-09T14:09:12.2514218Z  echo "#!/usr/bin/env bash"; 2025-09-09T14:09:12.2514528Z  echo "set -eou pipefail"; 2025-09-09T14:09:12.2514825Z  # shellcheck disable=SC2016 2025-09-09T14:09:12.2515147Z  echo 'eval "$(conda shell.bash hook)"'; 2025-09-09T14:09:12.2515462Z  echo "set -x"; 2025-09-09T14:09:12.2515726Z  echo "${SCRIPT}"; 2025-09-09T14:09:12.2515999Z } > "${RUNNER_TEMP}/exec_script" 2025-09-09T14:09:12.2516323Z chmod +x "${RUNNER_TEMP}/exec_script" 2025-09-09T14:09:12.2516921Z python3 "/home/ec2-user/actions-runner/_work/ao/ao/test-infra/.github/scripts/run_with_env_secrets.py" "" 2025-09-09T14:09:12.2531425Z shell: /usr/bin/bash -e {0} 2025-09-09T14:09:12.2531671Z env: 2025-09-09T14:09:12.2531921Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:09:12.2532285Z REPOSITORY: pytorch/ao 2025-09-09T14:09:12.2532520Z PR_NUMBER: 2963 2025-09-09T14:09:12.2534271Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:09:12.2536256Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:09:12.2536784Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:09:12.2537277Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:09:12.2537692Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:09:12.2538242Z ALL_SECRETS: { "github_token": "***" } 2025-09-09T14:09:12.2538552Z ##[endgroup] 2025-09-09T14:09:12.2592089Z + echo '#!/usr/bin/env bash' 2025-09-09T14:09:12.2592393Z + echo 'set -eou pipefail' 2025-09-09T14:09:12.2592667Z + echo 'eval "$(conda shell.bash hook)"' 2025-09-09T14:09:12.2592946Z + echo 'set -x' 2025-09-09T14:09:12.2593182Z + echo 'conda create -n venv python=3.9 -y 2025-09-09T14:09:12.2593473Z conda activate venv 2025-09-09T14:09:12.2593838Z echo "::group::Install newer objcopy that supports --set-section-alignment" 2025-09-09T14:09:12.2594264Z dnf install -y gcc-toolset-10-binutils 2025-09-09T14:09:12.2594606Z export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH 2025-09-09T14:09:12.2594969Z python -m pip install --upgrade pip 2025-09-09T14:09:12.2595246Z pip install torch==2.7.0 2025-09-09T14:09:12.2595501Z sed -i '\'''\'' dev-requirements.txt 2025-09-09T14:09:12.2595793Z pip install -r dev-requirements.txt 2025-09-09T14:09:12.2596055Z pip install . 2025-09-09T14:09:12.2596312Z export CONDA=$(dirname $(dirname $(which conda))) 2025-09-09T14:09:12.2596671Z export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH 2025-09-09T14:09:12.2597000Z pytest test --verbose -s 2025-09-09T14:09:12.2597225Z ' 2025-09-09T14:09:12.2597742Z + chmod +x /home/ec2-user/actions-runner/_work/_temp/exec_script 2025-09-09T14:09:12.2610085Z + python3 /home/ec2-user/actions-runner/_work/ao/ao/test-infra/.github/scripts/run_with_env_secrets.py '' 2025-09-09T14:09:19.4509961Z Running command: 2025-09-09T14:09:19.4514837Z docker run -e PR_NUMBER -e RUNNER_ARTIFACT_DIR=/artifacts -e RUNNER_DOCS_DIR=/docs -e RUNNER_TEST_RESULTS_DIR=/test-results --env-file="/home/ec2-user/actions-runner/_work/_temp/github_env_17585175130" `# It is unknown why the container sees a different value for this.` -e GITHUB_STEP_SUMMARY -e SECRET_GITHUB_TOKEN --cap-add=SYS_PTRACE --detach --ipc=host --security-opt seccomp=unconfined --shm-size=2g --tty --ulimit stack=10485760:83886080 --ulimit core=0 --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all -v "/home/ec2-user/actions-runner/_work/ao/ao/pytorch/ao:/pytorch/ao" -v "/home/ec2-user/actions-runner/_work/ao/ao/test-infra:/test-infra" -v "/home/ec2-user/actions-runner/_work/_temp/artifacts:/artifacts" -v "/home/ec2-user/actions-runner/_work/_temp/docs:/docs" -v "/home/ec2-user/actions-runner/_work/_temp/test-results:/test-results" -v "/home/ec2-user/actions-runner/_work/_temp/exec_script:/exec" -v "/home/ec2-user/actions-runner/_work/_temp/_runner_file_commands/step_summary_87f8cbe1-652b-4a3a-a987-4c6637b6825f":"/home/ec2-user/actions-runner/_work/_temp/_runner_file_commands/step_summary_87f8cbe1-652b-4a3a-a987-4c6637b6825f" -w /pytorch/ao "pytorch/almalinux-builder:cuda12.6" 2025-09-09T14:09:19.4519589Z 2025-09-09T14:09:19.4519897Z c429987d197112cb1c1dca45efc7b01c8ed17990b5489cf4c359012ede81fdb6 2025-09-09T14:09:19.4520518Z Running command: docker exec -t c429987d197112cb1c1dca45efc7b01c8ed17990b5489cf4c359012ede81fdb6 /exec 2025-09-09T14:09:19.4521049Z + conda create -n venv python=3.9 -y 2025-09-09T14:09:19.4521318Z + local cmd=create 2025-09-09T14:09:19.4521530Z + case "$cmd" in 2025-09-09T14:09:19.4521768Z + __conda_exe create -n venv python=3.9 -y 2025-09-09T14:09:19.4522105Z + /opt/conda/bin/conda create -n venv python=3.9 -y 2025-09-09T14:09:19.4523016Z Collecting package metadata (current_repodata.json): - \ | / - \ | / - \ | / - \ | / - \ | / - \ | / - \ | / - \ | / - \ | / - \ | done 2025-09-09T14:09:19.4523604Z Solving environment: - done 2025-09-09T14:09:19.4523783Z 2025-09-09T14:09:19.4523788Z 2025-09-09T14:09:19.4523920Z ==> WARNING: A newer version of conda exists. <== 2025-09-09T14:09:19.4524229Z current version: 23.5.2 2025-09-09T14:09:19.4524476Z latest version: 25.7.0 2025-09-09T14:09:19.4524628Z 2025-09-09T14:09:19.4524745Z Please update conda by running 2025-09-09T14:09:19.4524916Z 2025-09-09T14:09:19.4525032Z $ conda update -n base -c defaults conda 2025-09-09T14:09:19.4525235Z 2025-09-09T14:09:19.4525433Z Or to minimize the number of packages updated during conda update use 2025-09-09T14:09:19.4525722Z 2025-09-09T14:09:19.4525820Z conda install conda=25.7.0 2025-09-09T14:09:19.4525999Z 2025-09-09T14:09:19.4526003Z 2025-09-09T14:09:19.4526007Z 2025-09-09T14:09:19.4526093Z ## Package Plan ## 2025-09-09T14:09:19.4526223Z 2025-09-09T14:09:19.4526350Z environment location: /opt/conda/envs/venv 2025-09-09T14:09:19.4526561Z 2025-09-09T14:09:19.4526653Z added / updated specs: 2025-09-09T14:09:19.4526896Z - python=3.9 2025-09-09T14:09:19.4527028Z 2025-09-09T14:09:19.4527031Z 2025-09-09T14:09:19.4527159Z The following packages will be downloaded: 2025-09-09T14:09:19.4527383Z 2025-09-09T14:09:19.4527494Z package | build 2025-09-09T14:09:19.4527812Z ---------------------------|----------------- 2025-09-09T14:09:19.4528152Z bzip2-1.0.8 | h5eee18b_6 262 KB 2025-09-09T14:09:19.4528563Z ld_impl_linux-64-2.40 | h12ee557_0 710 KB 2025-09-09T14:09:19.4528963Z libffi-3.4.4 | h6a678d5_1 141 KB 2025-09-09T14:09:19.4529331Z libxcb-1.17.0 | h9b100fa_0 430 KB 2025-09-09T14:09:19.4529699Z ncurses-6.5 | h7934f7d_0 1.1 MB 2025-09-09T14:09:19.4530058Z pip-25.2 | pyhc872135_0 1.2 MB 2025-09-09T14:09:19.4530430Z pthread-stubs-0.3 | h0ce48e5_1 5 KB 2025-09-09T14:09:19.4530816Z python-3.9.23 | he99959a_0 24.7 MB 2025-09-09T14:09:19.4531322Z readline-8.3 | hc2a1206_0 471 KB 2025-09-09T14:09:19.4531710Z setuptools-78.1.1 | py39h06a4308_0 1.7 MB 2025-09-09T14:09:19.4532102Z sqlite-3.50.2 | hb25bd0a_1 1.1 MB 2025-09-09T14:09:19.4532524Z tk-8.6.15 | h54e0aa7_0 3.4 MB 2025-09-09T14:09:19.4532885Z tzdata-2025b | h04d1e81_0 116 KB 2025-09-09T14:09:19.4533249Z wheel-0.45.1 | py39h06a4308_0 114 KB 2025-09-09T14:09:19.4533630Z xorg-libx11-1.8.12 | h9b100fa_1 895 KB 2025-09-09T14:09:19.4534026Z xorg-libxau-1.0.12 | h9b100fa_0 13 KB 2025-09-09T14:09:19.4534421Z xorg-libxdmcp-1.1.5 | h9b100fa_0 19 KB 2025-09-09T14:09:19.4534837Z xorg-xorgproto-2024.1 | h5eee18b_1 580 KB 2025-09-09T14:09:19.4535212Z xz-5.6.4 | h5eee18b_1 567 KB 2025-09-09T14:09:19.4535564Z zlib-1.2.13 | h5eee18b_1 111 KB 2025-09-09T14:09:19.4535977Z ------------------------------------------------------------ 2025-09-09T14:09:19.4536359Z Total: 37.6 MB 2025-09-09T14:09:19.4536565Z 2025-09-09T14:09:19.4536692Z The following NEW packages will be INSTALLED: 2025-09-09T14:09:19.4536908Z 2025-09-09T14:09:19.4537104Z _libgcc_mutex pkgs/main/linux-64::_libgcc_mutex-0.1-main 2025-09-09T14:09:19.4537536Z _openmp_mutex pkgs/main/linux-64::_openmp_mutex-5.1-1_gnu 2025-09-09T14:09:19.4537954Z bzip2 pkgs/main/linux-64::bzip2-1.0.8-h5eee18b_6 2025-09-09T14:09:19.4538434Z ca-certificates pkgs/main/linux-64::ca-certificates-2025.7.15-h06a4308_0 2025-09-09T14:09:19.4538905Z expat pkgs/main/linux-64::expat-2.7.1-h6a678d5_0 2025-09-09T14:09:19.4539351Z ld_impl_linux-64 pkgs/main/linux-64::ld_impl_linux-64-2.40-h12ee557_0 2025-09-09T14:09:19.4539804Z libffi pkgs/main/linux-64::libffi-3.4.4-h6a678d5_1 2025-09-09T14:09:19.4540228Z libgcc-ng pkgs/main/linux-64::libgcc-ng-11.2.0-h1234567_1 2025-09-09T14:09:19.4540660Z libgomp pkgs/main/linux-64::libgomp-11.2.0-h1234567_1 2025-09-09T14:09:19.4541115Z libstdcxx-ng pkgs/main/linux-64::libstdcxx-ng-11.2.0-h1234567_1 2025-09-09T14:09:19.4541564Z libxcb pkgs/main/linux-64::libxcb-1.17.0-h9b100fa_0 2025-09-09T14:09:19.4541967Z ncurses pkgs/main/linux-64::ncurses-6.5-h7934f7d_0 2025-09-09T14:09:19.4542389Z openssl pkgs/main/linux-64::openssl-3.0.17-h5eee18b_0 2025-09-09T14:09:19.4542787Z pip pkgs/main/noarch::pip-25.2-pyhc872135_0 2025-09-09T14:09:19.4543226Z pthread-stubs pkgs/main/linux-64::pthread-stubs-0.3-h0ce48e5_1 2025-09-09T14:09:19.4543681Z python pkgs/main/linux-64::python-3.9.23-he99959a_0 2025-09-09T14:09:19.4544096Z readline pkgs/main/linux-64::readline-8.3-hc2a1206_0 2025-09-09T14:09:19.4544568Z setuptools pkgs/main/linux-64::setuptools-78.1.1-py39h06a4308_0 2025-09-09T14:09:19.4545020Z sqlite pkgs/main/linux-64::sqlite-3.50.2-hb25bd0a_1 2025-09-09T14:09:19.4545412Z tk pkgs/main/linux-64::tk-8.6.15-h54e0aa7_0 2025-09-09T14:09:19.4545793Z tzdata pkgs/main/noarch::tzdata-2025b-h04d1e81_0 2025-09-09T14:09:19.4546194Z wheel pkgs/main/linux-64::wheel-0.45.1-py39h06a4308_0 2025-09-09T14:09:19.4546635Z xorg-libx11 pkgs/main/linux-64::xorg-libx11-1.8.12-h9b100fa_1 2025-09-09T14:09:19.4547097Z xorg-libxau pkgs/main/linux-64::xorg-libxau-1.0.12-h9b100fa_0 2025-09-09T14:09:19.4547589Z xorg-libxdmcp pkgs/main/linux-64::xorg-libxdmcp-1.1.5-h9b100fa_0 2025-09-09T14:09:19.4548101Z xorg-xorgproto pkgs/main/linux-64::xorg-xorgproto-2024.1-h5eee18b_1 2025-09-09T14:09:19.4548538Z xz pkgs/main/linux-64::xz-5.6.4-h5eee18b_1 2025-09-09T14:09:19.4549002Z zlib pkgs/main/linux-64::zlib-1.2.13-h5eee18b_1 2025-09-09T14:09:19.4549241Z 2025-09-09T14:09:19.4549245Z 2025-09-09T14:09:19.4549249Z 2025-09-09T14:09:19.4549359Z Downloading and Extracting Packages 2025-09-09T14:09:19.4549633Z 2025-09-09T14:09:19.4549773Z libxcb-1.17.0 | 430 KB | : 0% 0/1 [00:00=4.10.0 (from torch==2.7.0) 2025-09-09T14:09:33.0901901Z Downloading typing_extensions-4.15.0-py3-none-any.whl.metadata (3.3 kB) 2025-09-09T14:09:33.0902343Z Collecting sympy>=1.13.3 (from torch==2.7.0) 2025-09-09T14:09:33.0902808Z Downloading sympy-1.14.0-py3-none-any.whl.metadata (12 kB) 2025-09-09T14:09:33.0903275Z Collecting networkx (from torch==2.7.0) 2025-09-09T14:09:33.0903759Z Downloading networkx-3.2.1-py3-none-any.whl.metadata (5.2 kB) 2025-09-09T14:09:33.0904253Z Collecting jinja2 (from torch==2.7.0) 2025-09-09T14:09:33.0904661Z Downloading jinja2-3.1.6-py3-none-any.whl.metadata (2.9 kB) 2025-09-09T14:09:33.0905051Z Collecting fsspec (from torch==2.7.0) 2025-09-09T14:09:33.0905423Z Downloading fsspec-2025.9.0-py3-none-any.whl.metadata (10 kB) 2025-09-09T14:09:33.0905883Z Collecting nvidia-cuda-nvrtc-cu12==12.6.77 (from torch==2.7.0) 2025-09-09T14:09:33.0906428Z Downloading nvidia_cuda_nvrtc_cu12-12.6.77-py3-none-manylinux2014_x86_64.whl.metadata (1.5 kB) 2025-09-09T14:09:33.0907004Z Collecting nvidia-cuda-runtime-cu12==12.6.77 (from torch==2.7.0) 2025-09-09T14:09:33.0907649Z Downloading nvidia_cuda_runtime_cu12-12.6.77-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (1.5 kB) 2025-09-09T14:09:33.0908300Z Collecting nvidia-cuda-cupti-cu12==12.6.80 (from torch==2.7.0) 2025-09-09T14:09:33.0908935Z Downloading nvidia_cuda_cupti_cu12-12.6.80-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (1.6 kB) 2025-09-09T14:09:33.0909682Z Collecting nvidia-cudnn-cu12==9.5.1.17 (from torch==2.7.0) 2025-09-09T14:09:33.0910208Z Downloading nvidia_cudnn_cu12-9.5.1.17-py3-none-manylinux_2_28_x86_64.whl.metadata (1.6 kB) 2025-09-09T14:09:33.0910846Z Collecting nvidia-cublas-cu12==12.6.4.1 (from torch==2.7.0) 2025-09-09T14:09:33.0911459Z Downloading nvidia_cublas_cu12-12.6.4.1-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (1.5 kB) 2025-09-09T14:09:33.0912069Z Collecting nvidia-cufft-cu12==11.3.0.4 (from torch==2.7.0) 2025-09-09T14:09:33.0912662Z Downloading nvidia_cufft_cu12-11.3.0.4-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (1.5 kB) 2025-09-09T14:09:33.0913321Z Collecting nvidia-curand-cu12==10.3.7.77 (from torch==2.7.0) 2025-09-09T14:09:33.0913929Z Downloading nvidia_curand_cu12-10.3.7.77-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (1.5 kB) 2025-09-09T14:09:33.0914559Z Collecting nvidia-cusolver-cu12==11.7.1.2 (from torch==2.7.0) 2025-09-09T14:09:33.0915181Z Downloading nvidia_cusolver_cu12-11.7.1.2-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (1.6 kB) 2025-09-09T14:09:33.0915827Z Collecting nvidia-cusparse-cu12==12.5.4.2 (from torch==2.7.0) 2025-09-09T14:09:33.0916457Z Downloading nvidia_cusparse_cu12-12.5.4.2-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (1.6 kB) 2025-09-09T14:09:33.0917078Z Collecting nvidia-cusparselt-cu12==0.6.3 (from torch==2.7.0) 2025-09-09T14:09:33.0917623Z Downloading nvidia_cusparselt_cu12-0.6.3-py3-none-manylinux2014_x86_64.whl.metadata (6.8 kB) 2025-09-09T14:09:33.0918156Z Collecting nvidia-nccl-cu12==2.26.2 (from torch==2.7.0) 2025-09-09T14:09:33.0918731Z Downloading nvidia_nccl_cu12-2.26.2-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (2.0 kB) 2025-09-09T14:09:33.0919319Z Collecting nvidia-nvtx-cu12==12.6.77 (from torch==2.7.0) 2025-09-09T14:09:33.0919897Z Downloading nvidia_nvtx_cu12-12.6.77-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (1.6 kB) 2025-09-09T14:09:33.0920518Z Collecting nvidia-nvjitlink-cu12==12.6.85 (from torch==2.7.0) 2025-09-09T14:09:33.0921142Z Downloading nvidia_nvjitlink_cu12-12.6.85-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl.metadata (1.5 kB) 2025-09-09T14:09:33.0921771Z Collecting nvidia-cufile-cu12==1.11.1.6 (from torch==2.7.0) 2025-09-09T14:09:33.0922381Z Downloading nvidia_cufile_cu12-1.11.1.6-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (1.5 kB) 2025-09-09T14:09:33.0923033Z Collecting triton==3.3.0 (from torch==2.7.0) 2025-09-09T14:09:33.0923674Z Downloading triton-3.3.0-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl.metadata (1.5 kB) 2025-09-09T14:09:33.0924761Z Requirement already satisfied: setuptools>=40.8.0 in /opt/conda/envs/venv/lib/python3.9/site-packages (from triton==3.3.0->torch==2.7.0) (78.1.1) 2025-09-09T14:09:33.0925647Z Collecting mpmath<1.4,>=1.1.0 (from sympy>=1.13.3->torch==2.7.0) 2025-09-09T14:09:33.0926115Z Downloading mpmath-1.3.0-py3-none-any.whl.metadata (8.6 kB) 2025-09-09T14:09:33.0926539Z Collecting MarkupSafe>=2.0 (from jinja2->torch==2.7.0) 2025-09-09T14:09:33.0927109Z Downloading MarkupSafe-3.0.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (4.0 kB) 2025-09-09T14:09:33.0927742Z Downloading torch-2.7.0-cp39-cp39-manylinux_2_28_x86_64.whl (865.2 MB) 2025-09-09T14:09:33.0928791Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/865.2 MB ? eta -:--:-- 2025-09-09T14:09:33.0929501Z  ━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 61.6/865.2 MB 311.5 MB/s eta 0:00:03 2025-09-09T14:09:40.7427689Z  ━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 118.0/865.2 MB 295.6 MB/s eta 0:00:03 2025-09-09T14:09:40.7429426Z  ━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 180.1/865.2 MB 298.4 MB/s eta 0:00:03 2025-09-09T14:09:40.7431307Z  ━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 242.7/865.2 MB 301.7 MB/s eta 0:00:03 2025-09-09T14:09:40.7432720Z  ━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━ 301.7/865.2 MB 296.3 MB/s eta 0:00:02 2025-09-09T14:09:40.7434440Z  ━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━ 360.7/865.2 MB 302.7 MB/s eta 0:00:02 2025-09-09T14:09:40.7437888Z  ━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━ 414.4/865.2 MB 294.0 MB/s eta 0:00:02 2025-09-09T14:09:40.7438715Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━ 480.0/865.2 MB 295.7 MB/s eta 0:00:02 2025-09-09T14:09:40.7439543Z  ━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━ 544.2/865.2 MB 303.0 MB/s eta 0:00:02 2025-09-09T14:09:40.7440416Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━ 607.1/865.2 MB 306.5 MB/s eta 0:00:01 2025-09-09T14:09:40.7441244Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━ 661.1/865.2 MB 305.6 MB/s eta 0:00:01 2025-09-09T14:09:40.7442096Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 716.2/865.2 MB 295.2 MB/s eta 0:00:01 2025-09-09T14:09:40.7442924Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━ 776.5/865.2 MB 291.4 MB/s eta 0:00:01 2025-09-09T14:09:40.7443743Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 839.9/865.2 MB 289.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7444496Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7445175Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7445862Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7446573Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7447225Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7447882Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7448539Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7449309Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7449977Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7450721Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7451381Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7452047Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7452704Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7453366Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7454045Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7454703Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7455387Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7456150Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7456816Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7457482Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7458154Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7458816Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7459488Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7460156Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:40.7460815Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:46.7668396Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:46.7669119Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:46.7670090Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:46.7670777Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:46.7671614Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:46.7672274Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:46.7672937Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 865.1/865.2 MB 292.3 MB/s eta 0:00:01 2025-09-09T14:09:46.7673574Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 865.2/865.2 MB 35.3 MB/s 0:00:09 2025-09-09T14:09:46.7674342Z [?25hDownloading nvidia_cublas_cu12-12.6.4.1-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (393.1 MB) 2025-09-09T14:09:46.7675096Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/393.1 MB ? eta -:--:-- 2025-09-09T14:09:46.7675789Z  ━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 49.3/393.1 MB 260.5 MB/s eta 0:00:02 2025-09-09T14:09:46.7676504Z  ━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 91.2/393.1 MB 227.5 MB/s eta 0:00:02 2025-09-09T14:09:46.7677215Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━ 134.2/393.1 MB 223.8 MB/s eta 0:00:02 2025-09-09T14:09:46.7677931Z  ━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━ 179.6/393.1 MB 223.1 MB/s eta 0:00:01 2025-09-09T14:09:46.7678647Z  ━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━ 222.6/393.1 MB 221.2 MB/s eta 0:00:01 2025-09-09T14:09:46.7679461Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━ 265.6/393.1 MB 219.1 MB/s eta 0:00:01 2025-09-09T14:09:46.7680332Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━ 312.5/393.1 MB 218.7 MB/s eta 0:00:01 2025-09-09T14:09:46.7681041Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━ 347.3/393.1 MB 211.9 MB/s eta 0:00:01 2025-09-09T14:09:46.7681745Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 379.3/393.1 MB 203.2 MB/s eta 0:00:01 2025-09-09T14:09:46.7682434Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 393.0/393.1 MB 205.1 MB/s eta 0:00:01 2025-09-09T14:09:46.7683107Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 393.0/393.1 MB 205.1 MB/s eta 0:00:01 2025-09-09T14:09:46.7683794Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 393.0/393.1 MB 205.1 MB/s eta 0:00:01 2025-09-09T14:09:46.7684465Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 393.0/393.1 MB 205.1 MB/s eta 0:00:01 2025-09-09T14:09:46.7685151Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 393.0/393.1 MB 205.1 MB/s eta 0:00:01 2025-09-09T14:09:46.7685825Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 393.0/393.1 MB 205.1 MB/s eta 0:00:01 2025-09-09T14:09:46.7686487Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 393.0/393.1 MB 205.1 MB/s eta 0:00:01 2025-09-09T14:09:46.7687149Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 393.0/393.1 MB 205.1 MB/s eta 0:00:01 2025-09-09T14:09:46.7687859Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 393.0/393.1 MB 205.1 MB/s eta 0:00:01 2025-09-09T14:09:46.7688541Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 393.0/393.1 MB 205.1 MB/s eta 0:00:01 2025-09-09T14:09:46.7689226Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 393.0/393.1 MB 205.1 MB/s eta 0:00:01 2025-09-09T14:09:46.7689864Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 393.1/393.1 MB 72.2 MB/s 0:00:04 2025-09-09T14:09:46.7690599Z [?25hDownloading nvidia_cuda_cupti_cu12-12.6.80-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (8.9 MB) 2025-09-09T14:09:46.7691343Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/8.9 MB ? eta -:--:-- 2025-09-09T14:09:46.7691939Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 8.9/8.9 MB 157.1 MB/s 0:00:00 2025-09-09T14:09:46.7692723Z [?25hDownloading nvidia_cuda_nvrtc_cu12-12.6.77-py3-none-manylinux2014_x86_64.whl (23.7 MB) 2025-09-09T14:09:46.7693410Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/23.7 MB ? eta -:--:-- 2025-09-09T14:09:46.7694107Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 23.7/23.7 MB 148.7 MB/s 0:00:00 2025-09-09T14:09:46.7694858Z [?25hDownloading nvidia_cuda_runtime_cu12-12.6.77-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (897 kB) 2025-09-09T14:09:46.7695624Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/897.7 kB ? eta -:--:-- 2025-09-09T14:09:46.7696326Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 897.7/897.7 kB 88.7 MB/s 0:00:00 2025-09-09T14:09:46.7696957Z [?25hDownloading nvidia_cudnn_cu12-9.5.1.17-py3-none-manylinux_2_28_x86_64.whl (571.0 MB) 2025-09-09T14:09:46.7697872Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/571.0 MB ? eta -:--:-- 2025-09-09T14:09:54.2261029Z  ━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 29.9/571.0 MB 149.3 MB/s eta 0:00:04 2025-09-09T14:09:54.2261854Z  ━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 71.6/571.0 MB 178.5 MB/s eta 0:00:03 2025-09-09T14:09:54.2262562Z  ━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 112.7/571.0 MB 187.2 MB/s eta 0:00:03 2025-09-09T14:09:54.2263274Z  ━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 146.0/571.0 MB 181.4 MB/s eta 0:00:03 2025-09-09T14:09:54.2263982Z  ━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━ 186.6/571.0 MB 185.7 MB/s eta 0:00:03 2025-09-09T14:09:54.2264681Z  ━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━ 228.6/571.0 MB 189.5 MB/s eta 0:00:02 2025-09-09T14:09:54.2265409Z  ━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━ 272.6/571.0 MB 199.4 MB/s eta 0:00:02 2025-09-09T14:09:54.2266125Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━ 318.5/571.0 MB 206.0 MB/s eta 0:00:02 2025-09-09T14:09:54.2266829Z  ━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━ 363.6/571.0 MB 202.6 MB/s eta 0:00:02 2025-09-09T14:09:54.2267539Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━ 413.4/571.0 MB 221.1 MB/s eta 0:00:01 2025-09-09T14:09:54.2268239Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━ 456.1/571.0 MB 224.8 MB/s eta 0:00:01 2025-09-09T14:09:54.2268943Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━ 500.2/571.0 MB 226.2 MB/s eta 0:00:01 2025-09-09T14:09:54.2269947Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 548.1/571.0 MB 226.8 MB/s eta 0:00:01 2025-09-09T14:09:54.2270640Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 570.9/571.0 MB 224.6 MB/s eta 0:00:01 2025-09-09T14:09:54.2271477Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 570.9/571.0 MB 224.6 MB/s eta 0:00:01 2025-09-09T14:09:54.2272142Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 570.9/571.0 MB 224.6 MB/s eta 0:00:01 2025-09-09T14:09:54.2272808Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 570.9/571.0 MB 224.6 MB/s eta 0:00:01 2025-09-09T14:09:54.2273474Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 570.9/571.0 MB 224.6 MB/s eta 0:00:01 2025-09-09T14:09:54.2274146Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 570.9/571.0 MB 224.6 MB/s eta 0:00:01 2025-09-09T14:09:54.2274809Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 570.9/571.0 MB 224.6 MB/s eta 0:00:01 2025-09-09T14:09:54.2275491Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 570.9/571.0 MB 224.6 MB/s eta 0:00:01 2025-09-09T14:09:54.2276145Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 570.9/571.0 MB 224.6 MB/s eta 0:00:01 2025-09-09T14:09:54.2276805Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 570.9/571.0 MB 224.6 MB/s eta 0:00:01 2025-09-09T14:09:54.2277457Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 570.9/571.0 MB 224.6 MB/s eta 0:00:01 2025-09-09T14:09:54.2278118Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 570.9/571.0 MB 224.6 MB/s eta 0:00:01 2025-09-09T14:09:54.2278801Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 570.9/571.0 MB 224.6 MB/s eta 0:00:01 2025-09-09T14:09:54.2279471Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 570.9/571.0 MB 224.6 MB/s eta 0:00:01 2025-09-09T14:09:54.2280179Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 570.9/571.0 MB 224.6 MB/s eta 0:00:01 2025-09-09T14:09:54.2280833Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 570.9/571.0 MB 224.6 MB/s eta 0:00:01 2025-09-09T14:09:54.2281491Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 570.9/571.0 MB 224.6 MB/s eta 0:00:01 2025-09-09T14:09:54.2282153Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 570.9/571.0 MB 224.6 MB/s eta 0:00:01 2025-09-09T14:09:54.2282913Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 570.9/571.0 MB 224.6 MB/s eta 0:00:01 2025-09-09T14:09:54.2283553Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 571.0/571.0 MB 51.6 MB/s 0:00:06 2025-09-09T14:09:54.2284399Z [?25hDownloading nvidia_cufft_cu12-11.3.0.4-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (200.2 MB) 2025-09-09T14:09:54.2285148Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/200.2 MB ? eta -:--:-- 2025-09-09T14:09:54.2285819Z  ━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 46.7/200.2 MB 233.4 MB/s eta 0:00:01 2025-09-09T14:09:54.2286531Z  ━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━ 95.4/200.2 MB 238.1 MB/s eta 0:00:01 2025-09-09T14:09:54.2287242Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━ 143.4/200.2 MB 237.6 MB/s eta 0:00:01 2025-09-09T14:09:54.2287968Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━ 183.5/200.2 MB 228.5 MB/s eta 0:00:01 2025-09-09T14:09:59.9337677Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 200.0/200.2 MB 228.7 MB/s eta 0:00:01 2025-09-09T14:09:59.9338432Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 200.0/200.2 MB 228.7 MB/s eta 0:00:01 2025-09-09T14:09:59.9339371Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 200.0/200.2 MB 228.7 MB/s eta 0:00:01 2025-09-09T14:09:59.9340217Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 200.0/200.2 MB 228.7 MB/s eta 0:00:01 2025-09-09T14:09:59.9340875Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 200.0/200.2 MB 228.7 MB/s eta 0:00:01 2025-09-09T14:09:59.9341804Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 200.0/200.2 MB 228.7 MB/s eta 0:00:01 2025-09-09T14:09:59.9342525Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 200.0/200.2 MB 228.7 MB/s eta 0:00:01 2025-09-09T14:09:59.9343340Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 200.0/200.2 MB 228.7 MB/s eta 0:00:01 2025-09-09T14:09:59.9343982Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 200.2/200.2 MB 76.5 MB/s 0:00:02 2025-09-09T14:09:59.9344714Z [?25hDownloading nvidia_cufile_cu12-1.11.1.6-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (1.1 MB) 2025-09-09T14:09:59.9345455Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/1.1 MB ? eta -:--:-- 2025-09-09T14:09:59.9346053Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.1/1.1 MB 15.5 MB/s 0:00:00 2025-09-09T14:09:59.9346788Z [?25hDownloading nvidia_curand_cu12-10.3.7.77-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (56.3 MB) 2025-09-09T14:09:59.9347520Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/56.3 MB ? eta -:--:-- 2025-09-09T14:09:59.9348202Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━ 43.5/56.3 MB 221.4 MB/s eta 0:00:01 2025-09-09T14:09:59.9348880Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 56.1/56.3 MB 220.9 MB/s eta 0:00:01 2025-09-09T14:09:59.9349535Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 56.1/56.3 MB 220.9 MB/s eta 0:00:01 2025-09-09T14:09:59.9350185Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 56.1/56.3 MB 220.9 MB/s eta 0:00:01 2025-09-09T14:09:59.9350809Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 56.3/56.3 MB 65.1 MB/s 0:00:00 2025-09-09T14:09:59.9351568Z [?25hDownloading nvidia_cusolver_cu12-11.7.1.2-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (158.2 MB) 2025-09-09T14:09:59.9352308Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/158.2 MB ? eta -:--:-- 2025-09-09T14:09:59.9352989Z  ━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 37.2/158.2 MB 187.3 MB/s eta 0:00:01 2025-09-09T14:09:59.9353694Z  ━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━ 66.1/158.2 MB 165.3 MB/s eta 0:00:01 2025-09-09T14:09:59.9354405Z  ━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━ 104.6/158.2 MB 173.4 MB/s eta 0:00:01 2025-09-09T14:09:59.9355114Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━ 135.5/158.2 MB 168.5 MB/s eta 0:00:01 2025-09-09T14:09:59.9355908Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 158.1/158.2 MB 169.0 MB/s eta 0:00:01 2025-09-09T14:09:59.9356583Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 158.1/158.2 MB 169.0 MB/s eta 0:00:01 2025-09-09T14:09:59.9357314Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 158.2/158.2 MB 122.7 MB/s 0:00:01 2025-09-09T14:09:59.9358050Z [?25hDownloading nvidia_cusparse_cu12-12.5.4.2-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (216.6 MB) 2025-09-09T14:09:59.9358798Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/216.6 MB ? eta -:--:-- 2025-09-09T14:09:59.9359456Z  ━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 49.3/216.6 MB 246.2 MB/s eta 0:00:01 2025-09-09T14:09:59.9360163Z  ━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━ 93.3/216.6 MB 233.1 MB/s eta 0:00:01 2025-09-09T14:09:59.9360894Z  ━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━ 134.2/216.6 MB 222.9 MB/s eta 0:00:01 2025-09-09T14:09:59.9361641Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━ 173.0/216.6 MB 215.4 MB/s eta 0:00:01 2025-09-09T14:09:59.9362347Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 216.5/216.6 MB 215.9 MB/s eta 0:00:01 2025-09-09T14:09:59.9363005Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 216.5/216.6 MB 215.9 MB/s eta 0:00:01 2025-09-09T14:09:59.9363670Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 216.5/216.6 MB 215.9 MB/s eta 0:00:01 2025-09-09T14:09:59.9364336Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 216.5/216.6 MB 215.9 MB/s eta 0:00:01 2025-09-09T14:09:59.9365009Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 216.5/216.6 MB 215.9 MB/s eta 0:00:01 2025-09-09T14:09:59.9365653Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 216.6/216.6 MB 116.9 MB/s 0:00:01 2025-09-09T14:09:59.9366326Z [?25hDownloading nvidia_cusparselt_cu12-0.6.3-py3-none-manylinux2014_x86_64.whl (156.8 MB) 2025-09-09T14:10:07.0424141Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/156.8 MB ? eta -:--:-- 2025-09-09T14:10:07.0424888Z  ━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━ 59.0/156.8 MB 295.1 MB/s eta 0:00:01 2025-09-09T14:10:07.0425598Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━ 107.0/156.8 MB 267.3 MB/s eta 0:00:01 2025-09-09T14:10:07.0426292Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺ 153.9/156.8 MB 255.1 MB/s eta 0:00:01 2025-09-09T14:10:07.0427264Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 156.8/156.8 MB 247.6 MB/s eta 0:00:01 2025-09-09T14:10:07.0427928Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 156.8/156.8 MB 247.6 MB/s eta 0:00:01 2025-09-09T14:10:07.0428736Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 156.8/156.8 MB 133.3 MB/s 0:00:01 2025-09-09T14:10:07.0429474Z [?25hDownloading nvidia_nccl_cu12-2.26.2-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (201.3 MB) 2025-09-09T14:10:07.0430205Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/201.3 MB ? eta -:--:-- 2025-09-09T14:10:07.0430875Z  ━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━ 65.3/201.3 MB 327.1 MB/s eta 0:00:01 2025-09-09T14:10:07.0431603Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━ 108.0/201.3 MB 269.7 MB/s eta 0:00:01 2025-09-09T14:10:07.0432311Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━ 148.9/201.3 MB 247.6 MB/s eta 0:00:01 2025-09-09T14:10:07.0433033Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━ 181.9/201.3 MB 226.2 MB/s eta 0:00:01 2025-09-09T14:10:07.0433710Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 201.1/201.3 MB 226.3 MB/s eta 0:00:01 2025-09-09T14:10:07.0434374Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 201.1/201.3 MB 226.3 MB/s eta 0:00:01 2025-09-09T14:10:07.0435036Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 201.1/201.3 MB 226.3 MB/s eta 0:00:01 2025-09-09T14:10:07.0435690Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 201.1/201.3 MB 226.3 MB/s eta 0:00:01 2025-09-09T14:10:07.0436371Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 201.1/201.3 MB 226.3 MB/s eta 0:00:01 2025-09-09T14:10:07.0437020Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 201.3/201.3 MB 108.3 MB/s 0:00:01 2025-09-09T14:10:07.0437777Z [?25hDownloading nvidia_nvjitlink_cu12-12.6.85-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl (19.7 MB) 2025-09-09T14:10:07.0438519Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/19.7 MB ? eta -:--:-- 2025-09-09T14:10:07.0439115Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 19.7/19.7 MB 123.9 MB/s 0:00:00 2025-09-09T14:10:07.0439817Z [?25hDownloading nvidia_nvtx_cu12-12.6.77-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (89 kB) 2025-09-09T14:10:07.0440503Z Downloading triton-3.3.0-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (156.4 MB) 2025-09-09T14:10:07.0441289Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/156.4 MB ? eta -:--:-- 2025-09-09T14:10:07.0441955Z  ━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━ 57.7/156.4 MB 289.2 MB/s eta 0:00:01 2025-09-09T14:10:07.0442752Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 121.9/156.4 MB 303.4 MB/s eta 0:00:01 2025-09-09T14:10:07.0443437Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 156.2/156.4 MB 304.5 MB/s eta 0:00:01 2025-09-09T14:10:07.0444098Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 156.2/156.4 MB 304.5 MB/s eta 0:00:01 2025-09-09T14:10:07.0444752Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 156.2/156.4 MB 304.5 MB/s eta 0:00:01 2025-09-09T14:10:07.0445448Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 156.2/156.4 MB 304.5 MB/s eta 0:00:01 2025-09-09T14:10:07.0446102Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 156.2/156.4 MB 304.5 MB/s eta 0:00:01 2025-09-09T14:10:07.0446754Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 156.4/156.4 MB 105.4 MB/s 0:00:01 2025-09-09T14:10:07.0447272Z [?25hDownloading sympy-1.14.0-py3-none-any.whl (6.3 MB) 2025-09-09T14:10:07.0447815Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/6.3 MB ? eta -:--:-- 2025-09-09T14:10:07.0448406Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 6.3/6.3 MB 146.2 MB/s 0:00:00 2025-09-09T14:10:07.0448904Z [?25hDownloading mpmath-1.3.0-py3-none-any.whl (536 kB) 2025-09-09T14:10:07.0449449Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/536.2 kB ? eta -:--:-- 2025-09-09T14:10:07.0450157Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 536.2/536.2 kB 19.7 MB/s 0:00:00 2025-09-09T14:10:07.0450716Z [?25hDownloading typing_extensions-4.15.0-py3-none-any.whl (44 kB) 2025-09-09T14:10:07.0451249Z Downloading filelock-3.19.1-py3-none-any.whl (15 kB) 2025-09-09T14:10:07.0451646Z Downloading fsspec-2025.9.0-py3-none-any.whl (199 kB) 2025-09-09T14:10:07.0452024Z Downloading jinja2-3.1.6-py3-none-any.whl (134 kB) 2025-09-09T14:10:07.0452545Z Downloading MarkupSafe-3.0.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (20 kB) 2025-09-09T14:10:07.0453082Z Downloading networkx-3.2.1-py3-none-any.whl (1.6 MB) 2025-09-09T14:10:07.0453630Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/1.6 MB ? eta -:--:-- 2025-09-09T14:10:07.0454218Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.6/1.6 MB 120.5 MB/s 0:00:00 2025-09-09T14:10:14.6665120Z [?25hInstalling collected packages: nvidia-cusparselt-cu12, mpmath, typing-extensions, triton, sympy, nvidia-nvtx-cu12, nvidia-nvjitlink-cu12, nvidia-nccl-cu12, nvidia-curand-cu12, nvidia-cufile-cu12, nvidia-cuda-runtime-cu12, nvidia-cuda-nvrtc-cu12, nvidia-cuda-cupti-cu12, nvidia-cublas-cu12, networkx, MarkupSafe, fsspec, filelock, nvidia-cusparse-cu12, nvidia-cufft-cu12, nvidia-cudnn-cu12, jinja2, nvidia-cusolver-cu12, torch 2025-09-09T14:10:14.6666959Z [?25l 2025-09-09T14:10:14.6667366Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  0/24 [nvidia-cusparselt-cu12] 2025-09-09T14:10:14.6667916Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  0/24 [nvidia-cusparselt-cu12] 2025-09-09T14:10:14.6668451Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  0/24 [nvidia-cusparselt-cu12] 2025-09-09T14:10:14.6668985Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  0/24 [nvidia-cusparselt-cu12] 2025-09-09T14:10:14.6669518Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  0/24 [nvidia-cusparselt-cu12] 2025-09-09T14:10:14.6670059Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  0/24 [nvidia-cusparselt-cu12] 2025-09-09T14:10:14.6670595Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  0/24 [nvidia-cusparselt-cu12] 2025-09-09T14:10:14.6671144Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  0/24 [nvidia-cusparselt-cu12] 2025-09-09T14:10:14.6671679Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  0/24 [nvidia-cusparselt-cu12] 2025-09-09T14:10:14.6672234Z  ━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  1/24 [mpmath] 2025-09-09T14:10:14.6672771Z  ━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  1/24 [mpmath] 2025-09-09T14:10:14.6673342Z  ━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  2/24 [typing-extensions] 2025-09-09T14:10:14.6673903Z  ━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  3/24 [triton] 2025-09-09T14:10:14.6674674Z  ━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  3/24 [triton] 2025-09-09T14:10:14.6675227Z  ━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  3/24 [triton] 2025-09-09T14:10:14.6675940Z  ━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  3/24 [triton] 2025-09-09T14:10:14.6676487Z  ━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  3/24 [triton] 2025-09-09T14:10:14.6677021Z  ━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  3/24 [triton] 2025-09-09T14:10:14.6677562Z  ━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  3/24 [triton] 2025-09-09T14:10:14.6678101Z  ━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  3/24 [triton] 2025-09-09T14:10:14.6678645Z  ━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  3/24 [triton] 2025-09-09T14:10:14.6679205Z  ━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  3/24 [triton] 2025-09-09T14:10:14.6679740Z  ━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  3/24 [triton] 2025-09-09T14:10:14.6680300Z  ━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  3/24 [triton] 2025-09-09T14:10:14.6680836Z  ━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  3/24 [triton] 2025-09-09T14:10:14.6681378Z  ━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  3/24 [triton] 2025-09-09T14:10:14.6681925Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:14.6682464Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:14.6683005Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:14.6683555Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:14.6684096Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:14.6684652Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:14.6685186Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:14.6685723Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:14.6686254Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:14.6686789Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:14.6687330Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:14.6687958Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:14.6688503Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:14.6689126Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:14.6698526Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:14.6699271Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:14.6699825Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:21.8782829Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:21.8783436Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:21.8784020Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:21.8784557Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:21.8785130Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:21.8785653Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:21.8786186Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:21.8786714Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/24 [sympy] 2025-09-09T14:10:21.8787282Z  ━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  5/24 [nvidia-nvtx-cu12] 2025-09-09T14:10:21.8787890Z  ━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  6/24 [nvidia-nvjitlink-cu12] 2025-09-09T14:10:21.8788825Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  7/24 [nvidia-nccl-cu12] 2025-09-09T14:10:21.8789423Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  7/24 [nvidia-nccl-cu12] 2025-09-09T14:10:21.8790179Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  7/24 [nvidia-nccl-cu12] 2025-09-09T14:10:21.8790764Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  7/24 [nvidia-nccl-cu12] 2025-09-09T14:10:21.8791360Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  7/24 [nvidia-nccl-cu12] 2025-09-09T14:10:21.8791937Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  7/24 [nvidia-nccl-cu12] 2025-09-09T14:10:21.8792523Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  7/24 [nvidia-nccl-cu12] 2025-09-09T14:10:21.8793117Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  7/24 [nvidia-nccl-cu12] 2025-09-09T14:10:21.8793700Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  7/24 [nvidia-nccl-cu12] 2025-09-09T14:10:21.8794302Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  7/24 [nvidia-nccl-cu12] 2025-09-09T14:10:21.8794877Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  7/24 [nvidia-nccl-cu12] 2025-09-09T14:10:21.8795468Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━  8/24 [nvidia-curand-cu12] 2025-09-09T14:10:21.8796074Z  ━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━  9/24 [nvidia-cufile-cu12] 2025-09-09T14:10:21.8796687Z  ━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━ 11/24 [nvidia-cuda-nvrtc-cu12] 2025-09-09T14:10:21.8797502Z  ━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━ 11/24 [nvidia-cuda-nvrtc-cu12] 2025-09-09T14:10:21.8798111Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━ 13/24 [nvidia-cublas-cu12] 2025-09-09T14:10:21.8798740Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━ 13/24 [nvidia-cublas-cu12] 2025-09-09T14:10:21.8799332Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━ 13/24 [nvidia-cublas-cu12] 2025-09-09T14:10:21.8799924Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━ 13/24 [nvidia-cublas-cu12] 2025-09-09T14:10:21.8800512Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━ 13/24 [nvidia-cublas-cu12] 2025-09-09T14:10:21.8801104Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━ 13/24 [nvidia-cublas-cu12] 2025-09-09T14:10:21.8801691Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━ 13/24 [nvidia-cublas-cu12] 2025-09-09T14:10:21.8802427Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━ 13/24 [nvidia-cublas-cu12] 2025-09-09T14:10:21.8803024Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━ 13/24 [nvidia-cublas-cu12] 2025-09-09T14:10:21.8803752Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━ 13/24 [nvidia-cublas-cu12] 2025-09-09T14:10:21.8804339Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━ 13/24 [nvidia-cublas-cu12] 2025-09-09T14:10:21.8804934Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━ 13/24 [nvidia-cublas-cu12] 2025-09-09T14:10:21.8805518Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━ 13/24 [nvidia-cublas-cu12] 2025-09-09T14:10:21.8806125Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━ 13/24 [nvidia-cublas-cu12] 2025-09-09T14:10:21.8806742Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━ 13/24 [nvidia-cublas-cu12] 2025-09-09T14:10:21.8807328Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━ 13/24 [nvidia-cublas-cu12] 2025-09-09T14:10:21.8807946Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━ 13/24 [nvidia-cublas-cu12] 2025-09-09T14:10:21.8808534Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━ 13/24 [nvidia-cublas-cu12] 2025-09-09T14:10:29.0868323Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━ 13/24 [nvidia-cublas-cu12] 2025-09-09T14:10:29.0869253Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━ 13/24 [nvidia-cublas-cu12] 2025-09-09T14:10:29.0870050Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━ 13/24 [nvidia-cublas-cu12] 2025-09-09T14:10:29.0870999Z  ━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━ 14/24 [networkx] 2025-09-09T14:10:29.0871579Z  ━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━ 14/24 [networkx] 2025-09-09T14:10:29.0872340Z  ━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━ 14/24 [networkx] 2025-09-09T14:10:29.0872902Z  ━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━ 14/24 [networkx] 2025-09-09T14:10:29.0873476Z  ━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━ 14/24 [networkx] 2025-09-09T14:10:29.0874043Z  ━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━ 14/24 [networkx] 2025-09-09T14:10:29.0874604Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━ 16/24 [fsspec] 2025-09-09T14:10:29.0875164Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━ 16/24 [fsspec] 2025-09-09T14:10:29.0875788Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━ 18/24 [nvidia-cusparse-cu12] 2025-09-09T14:10:29.0876420Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━ 18/24 [nvidia-cusparse-cu12] 2025-09-09T14:10:29.0877059Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━ 18/24 [nvidia-cusparse-cu12] 2025-09-09T14:10:29.0877677Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━ 18/24 [nvidia-cusparse-cu12] 2025-09-09T14:10:29.0878284Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━ 18/24 [nvidia-cusparse-cu12] 2025-09-09T14:10:29.0878888Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━ 18/24 [nvidia-cusparse-cu12] 2025-09-09T14:10:29.0879505Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━ 18/24 [nvidia-cusparse-cu12] 2025-09-09T14:10:29.0880109Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━ 18/24 [nvidia-cusparse-cu12] 2025-09-09T14:10:29.0880741Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━ 18/24 [nvidia-cusparse-cu12] 2025-09-09T14:10:29.0881420Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━ 18/24 [nvidia-cusparse-cu12] 2025-09-09T14:10:29.0882025Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━ 18/24 [nvidia-cusparse-cu12] 2025-09-09T14:10:29.0882633Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━ 18/24 [nvidia-cusparse-cu12] 2025-09-09T14:10:29.0883239Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━ 19/24 [nvidia-cufft-cu12] 2025-09-09T14:10:29.0883830Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━ 19/24 [nvidia-cufft-cu12] 2025-09-09T14:10:29.0884420Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━ 19/24 [nvidia-cufft-cu12] 2025-09-09T14:10:29.0885760Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━ 19/24 [nvidia-cufft-cu12] 2025-09-09T14:10:29.0886364Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━ 19/24 [nvidia-cufft-cu12] 2025-09-09T14:10:29.0887043Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━ 19/24 [nvidia-cufft-cu12] 2025-09-09T14:10:29.0887631Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━ 19/24 [nvidia-cufft-cu12] 2025-09-09T14:10:29.0888218Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━ 19/24 [nvidia-cufft-cu12] 2025-09-09T14:10:29.0888796Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━ 19/24 [nvidia-cufft-cu12] 2025-09-09T14:10:29.0889380Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.0889988Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.0890576Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.0891237Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.0891826Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.0892418Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.0893000Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.0893599Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.0894193Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.0894879Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.0895570Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:36.3052413Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:36.3053810Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:36.3054449Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:36.3055208Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:36.3056051Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:36.3056688Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:36.3057293Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:36.3057914Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:36.3058510Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:36.3059116Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:36.3059710Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:36.3060320Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:36.3060947Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:36.3061545Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:36.3062166Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:36.3062757Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:36.3063387Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:36.3064010Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:36.3064599Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:36.3065199Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:36.3066071Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━ 20/24 [nvidia-cudnn-cu12] 2025-09-09T14:10:36.3066640Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━ 21/24 [jinja2] 2025-09-09T14:10:36.3067441Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━ 22/24 [nvidia-cusolver-cu12] 2025-09-09T14:10:36.3068042Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━ 22/24 [nvidia-cusolver-cu12] 2025-09-09T14:10:36.3068648Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━ 22/24 [nvidia-cusolver-cu12] 2025-09-09T14:10:36.3069245Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━ 22/24 [nvidia-cusolver-cu12] 2025-09-09T14:10:36.3069851Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━ 22/24 [nvidia-cusolver-cu12] 2025-09-09T14:10:36.3070481Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━ 22/24 [nvidia-cusolver-cu12] 2025-09-09T14:10:36.3071088Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━ 22/24 [nvidia-cusolver-cu12] 2025-09-09T14:10:36.3071711Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━ 22/24 [nvidia-cusolver-cu12] 2025-09-09T14:10:36.3072315Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━ 22/24 [nvidia-cusolver-cu12] 2025-09-09T14:10:36.3072917Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━ 22/24 [nvidia-cusolver-cu12] 2025-09-09T14:10:36.3073533Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:36.3074055Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:36.3074682Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:36.3075219Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:36.3075836Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:36.3076377Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:36.3076890Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:36.3077418Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:36.3077946Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:36.3078464Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:36.3078994Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0133284Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0134083Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0134683Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0135213Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0135748Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0136352Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0136885Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0137410Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0137947Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0138471Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0139004Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0139526Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0140043Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0140556Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0141077Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0141589Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0142386Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0142906Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0143598Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0144120Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0144632Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0145148Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0145658Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0146180Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0146719Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0147231Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0147769Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0148277Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0148797Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0149316Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0149826Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0150345Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0150961Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0151485Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0152103Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0152615Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0153130Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0153646Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0154169Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0154701Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0155236Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0155758Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0156288Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0156809Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0157330Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:44.0157844Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7591931Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7592660Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7593234Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7593754Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7594288Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7594802Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7595315Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7595822Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7596338Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7596844Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7597817Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7598354Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7599049Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7599570Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7600086Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7600590Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7601111Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7601620Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7602152Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7602664Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7603202Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7603718Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7604227Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7604738Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7605255Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7605770Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7606393Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7606904Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7607509Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7608016Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7608583Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7609099Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7609603Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7610119Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7610644Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 23/24 [torch] 2025-09-09T14:10:52.7611145Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 24/24 [torch] 2025-09-09T14:10:52.7611511Z [?25h 2025-09-09T14:10:52.7613927Z Successfully installed MarkupSafe-3.0.2 filelock-3.19.1 fsspec-2025.9.0 jinja2-3.1.6 mpmath-1.3.0 networkx-3.2.1 nvidia-cublas-cu12-12.6.4.1 nvidia-cuda-cupti-cu12-12.6.80 nvidia-cuda-nvrtc-cu12-12.6.77 nvidia-cuda-runtime-cu12-12.6.77 nvidia-cudnn-cu12-9.5.1.17 nvidia-cufft-cu12-11.3.0.4 nvidia-cufile-cu12-1.11.1.6 nvidia-curand-cu12-10.3.7.77 nvidia-cusolver-cu12-11.7.1.2 nvidia-cusparse-cu12-12.5.4.2 nvidia-cusparselt-cu12-0.6.3 nvidia-nccl-cu12-2.26.2 nvidia-nvjitlink-cu12-12.6.85 nvidia-nvtx-cu12-12.6.77 sympy-1.14.0 torch-2.7.0 triton-3.3.0 typing-extensions-4.15.0 2025-09-09T14:10:52.7617517Z WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager, possibly rendering your system unusable. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv. Use the --root-user-action option if you know what you are doing and want to suppress this warning. 2025-09-09T14:10:52.7619030Z + sed -i '' dev-requirements.txt 2025-09-09T14:10:52.7619351Z + pip install -r dev-requirements.txt 2025-09-09T14:10:52.7619709Z Collecting pytest (from -r dev-requirements.txt (line 2)) 2025-09-09T14:10:52.7620165Z Downloading pytest-8.4.2-py3-none-any.whl.metadata (7.7 kB) 2025-09-09T14:10:52.7620670Z Collecting unittest-xml-reporting (from -r dev-requirements.txt (line 3)) 2025-09-09T14:10:52.7621267Z Downloading unittest_xml_reporting-3.2.0-py2.py3-none-any.whl.metadata (11 kB) 2025-09-09T14:10:52.7621815Z Collecting parameterized (from -r dev-requirements.txt (line 4)) 2025-09-09T14:10:52.7622343Z Downloading parameterized-0.9.0-py2.py3-none-any.whl.metadata (18 kB) 2025-09-09T14:10:52.7622861Z Collecting packaging (from -r dev-requirements.txt (line 5)) 2025-09-09T14:10:52.7623454Z Downloading packaging-25.0-py3-none-any.whl.metadata (3.3 kB) 2025-09-09T14:10:52.7623936Z Collecting transformers (from -r dev-requirements.txt (line 6)) 2025-09-09T14:10:52.7624507Z Downloading transformers-4.56.1-py3-none-any.whl.metadata (42 kB) 2025-09-09T14:10:52.7624996Z Collecting hypothesis (from -r dev-requirements.txt (line 7)) 2025-09-09T14:11:03.8073027Z Downloading hypothesis-6.138.15-py3-none-any.whl.metadata (5.6 kB) 2025-09-09T14:11:03.8073782Z Collecting sentencepiece (from -r dev-requirements.txt (line 8)) 2025-09-09T14:11:03.8074480Z Downloading sentencepiece-0.2.1-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl.metadata (10 kB) 2025-09-09T14:11:03.8075122Z Collecting expecttest (from -r dev-requirements.txt (line 9)) 2025-09-09T14:11:03.8075607Z Downloading expecttest-0.3.0-py3-none-any.whl.metadata (3.8 kB) 2025-09-09T14:11:03.8076111Z Collecting bitsandbytes (from -r dev-requirements.txt (line 12)) 2025-09-09T14:11:03.8076673Z Downloading bitsandbytes-0.47.0-py3-none-manylinux_2_24_x86_64.whl.metadata (11 kB) 2025-09-09T14:11:03.8077231Z Collecting matplotlib (from -r dev-requirements.txt (line 13)) 2025-09-09T14:11:03.8077862Z Downloading matplotlib-3.9.4-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (11 kB) 2025-09-09T14:11:03.8078458Z Collecting pandas (from -r dev-requirements.txt (line 14)) 2025-09-09T14:11:03.8079039Z Downloading pandas-2.3.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (91 kB) 2025-09-09T14:11:03.8079598Z Collecting fire (from -r dev-requirements.txt (line 15)) 2025-09-09T14:11:03.8080039Z Downloading fire-0.7.1-py3-none-any.whl.metadata (5.8 kB) 2025-09-09T14:11:03.8080481Z Collecting tabulate (from -r dev-requirements.txt (line 16)) 2025-09-09T14:11:03.8080948Z Downloading tabulate-0.9.0-py3-none-any.whl.metadata (34 kB) 2025-09-09T14:11:03.8081419Z Collecting tiktoken (from -r dev-requirements.txt (line 17)) 2025-09-09T14:11:03.8082015Z Downloading tiktoken-0.11.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (6.7 kB) 2025-09-09T14:11:03.8082607Z Collecting blobfile (from -r dev-requirements.txt (line 18)) 2025-09-09T14:11:03.8083071Z Downloading blobfile-3.1.0-py3-none-any.whl.metadata (15 kB) 2025-09-09T14:11:03.8083528Z Collecting lm_eval (from -r dev-requirements.txt (line 19)) 2025-09-09T14:11:03.8083974Z Downloading lm_eval-0.4.9.1-py3-none-any.whl.metadata (53 kB) 2025-09-09T14:11:03.8084440Z Collecting diskcache (from -r dev-requirements.txt (line 21)) 2025-09-09T14:11:03.8084917Z Downloading diskcache-5.6.3-py3-none-any.whl.metadata (20 kB) 2025-09-09T14:11:03.8085389Z Collecting pycocotools (from -r dev-requirements.txt (line 22)) 2025-09-09T14:11:03.8086022Z Downloading pycocotools-2.0.10-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (1.3 kB) 2025-09-09T14:11:03.8086611Z Collecting tqdm (from -r dev-requirements.txt (line 23)) 2025-09-09T14:11:03.8087048Z Downloading tqdm-4.67.1-py3-none-any.whl.metadata (57 kB) 2025-09-09T14:11:03.8087537Z Collecting importlib_metadata (from -r dev-requirements.txt (line 24)) 2025-09-09T14:11:03.8088082Z Downloading importlib_metadata-8.7.0-py3-none-any.whl.metadata (4.8 kB) 2025-09-09T14:11:03.8088583Z Collecting ninja (from -r dev-requirements.txt (line 27)) 2025-09-09T14:11:03.8089143Z Downloading ninja-1.13.0-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (5.1 kB) 2025-09-09T14:11:03.8089744Z Collecting cmake<4.0.0,>=3.19.0 (from -r dev-requirements.txt (line 30)) 2025-09-09T14:11:03.8090334Z Downloading cmake-3.31.6-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (6.3 kB) 2025-09-09T14:11:03.8090916Z Collecting ruff==0.11.6 (from -r dev-requirements.txt (line 33)) 2025-09-09T14:11:03.8091471Z Downloading ruff-0.11.6-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (25 kB) 2025-09-09T14:11:03.8092044Z Collecting pre-commit (from -r dev-requirements.txt (line 34)) 2025-09-09T14:11:03.8092827Z Downloading pre_commit-4.3.0-py2.py3-none-any.whl.metadata (1.2 kB) 2025-09-09T14:11:03.8093375Z Collecting exceptiongroup>=1 (from pytest->-r dev-requirements.txt (line 2)) 2025-09-09T14:11:03.8093927Z Downloading exceptiongroup-1.3.0-py3-none-any.whl.metadata (6.7 kB) 2025-09-09T14:11:03.8094596Z Collecting iniconfig>=1 (from pytest->-r dev-requirements.txt (line 2)) 2025-09-09T14:11:03.8095097Z Downloading iniconfig-2.1.0-py3-none-any.whl.metadata (2.7 kB) 2025-09-09T14:11:03.8095584Z Collecting pluggy<2,>=1.5 (from pytest->-r dev-requirements.txt (line 2)) 2025-09-09T14:11:03.8096169Z Downloading pluggy-1.6.0-py3-none-any.whl.metadata (4.8 kB) 2025-09-09T14:11:03.8096663Z Collecting pygments>=2.7.2 (from pytest->-r dev-requirements.txt (line 2)) 2025-09-09T14:11:03.8097166Z Downloading pygments-2.19.2-py3-none-any.whl.metadata (2.5 kB) 2025-09-09T14:11:03.8097889Z Collecting tomli>=1 (from pytest->-r dev-requirements.txt (line 2)) 2025-09-09T14:11:03.8098346Z Downloading tomli-2.2.1-py3-none-any.whl.metadata (10 kB) 2025-09-09T14:11:03.8098866Z Collecting lxml (from unittest-xml-reporting->-r dev-requirements.txt (line 3)) 2025-09-09T14:11:03.8099499Z Downloading lxml-6.0.1-cp39-cp39-manylinux_2_26_x86_64.manylinux_2_28_x86_64.whl.metadata (3.8 kB) 2025-09-09T14:11:03.8100423Z Requirement already satisfied: filelock in /opt/conda/envs/venv/lib/python3.9/site-packages (from transformers->-r dev-requirements.txt (line 6)) (3.19.1) 2025-09-09T14:11:03.8101314Z Collecting huggingface-hub<1.0,>=0.34.0 (from transformers->-r dev-requirements.txt (line 6)) 2025-09-09T14:11:03.8101897Z Downloading huggingface_hub-0.34.4-py3-none-any.whl.metadata (14 kB) 2025-09-09T14:11:03.8102421Z Collecting numpy>=1.17 (from transformers->-r dev-requirements.txt (line 6)) 2025-09-09T14:11:03.8103018Z Downloading numpy-2.0.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (60 kB) 2025-09-09T14:11:03.8103633Z Collecting pyyaml>=5.1 (from transformers->-r dev-requirements.txt (line 6)) 2025-09-09T14:11:03.8104249Z Downloading PyYAML-6.0.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (2.1 kB) 2025-09-09T14:11:03.8104874Z Collecting regex!=2019.12.17 (from transformers->-r dev-requirements.txt (line 6)) 2025-09-09T14:11:03.8105590Z Downloading regex-2025.9.1-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl.metadata (40 kB) 2025-09-09T14:11:03.8106276Z Collecting requests (from transformers->-r dev-requirements.txt (line 6)) 2025-09-09T14:11:03.8106789Z Downloading requests-2.32.5-py3-none-any.whl.metadata (4.9 kB) 2025-09-09T14:11:03.8107329Z Collecting tokenizers<=0.23.0,>=0.22.0 (from transformers->-r dev-requirements.txt (line 6)) 2025-09-09T14:11:03.8107995Z Downloading tokenizers-0.22.0-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (6.8 kB) 2025-09-09T14:11:03.8108657Z Collecting safetensors>=0.4.3 (from transformers->-r dev-requirements.txt (line 6)) 2025-09-09T14:11:03.8109322Z Downloading safetensors-0.6.2-cp38-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (4.1 kB) 2025-09-09T14:11:03.8110381Z Requirement already satisfied: fsspec>=2023.5.0 in /opt/conda/envs/venv/lib/python3.9/site-packages (from huggingface-hub<1.0,>=0.34.0->transformers->-r dev-requirements.txt (line 6)) (2025.9.0) 2025-09-09T14:11:03.8111767Z Requirement already satisfied: typing-extensions>=3.7.4.3 in /opt/conda/envs/venv/lib/python3.9/site-packages (from huggingface-hub<1.0,>=0.34.0->transformers->-r dev-requirements.txt (line 6)) (4.15.0) 2025-09-09T14:11:03.8112835Z Collecting hf-xet<2.0.0,>=1.1.3 (from huggingface-hub<1.0,>=0.34.0->transformers->-r dev-requirements.txt (line 6)) 2025-09-09T14:11:03.8113543Z Downloading hf_xet-1.1.9-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (4.7 kB) 2025-09-09T14:11:03.8114147Z Collecting attrs>=22.2.0 (from hypothesis->-r dev-requirements.txt (line 7)) 2025-09-09T14:11:03.8114627Z Downloading attrs-25.3.0-py3-none-any.whl.metadata (10 kB) 2025-09-09T14:11:03.8115295Z Collecting sortedcontainers<3.0.0,>=2.1.0 (from hypothesis->-r dev-requirements.txt (line 7)) 2025-09-09T14:11:03.8115923Z Downloading sortedcontainers-2.4.0-py2.py3-none-any.whl.metadata (10 kB) 2025-09-09T14:11:03.8116917Z Requirement already satisfied: torch<3,>=2.2 in /opt/conda/envs/venv/lib/python3.9/site-packages (from bitsandbytes->-r dev-requirements.txt (line 12)) (2.7.0) 2025-09-09T14:11:03.8118087Z Requirement already satisfied: sympy>=1.13.3 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (1.14.0) 2025-09-09T14:11:03.8119301Z Requirement already satisfied: networkx in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (3.2.1) 2025-09-09T14:11:03.8120493Z Requirement already satisfied: jinja2 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (3.1.6) 2025-09-09T14:11:03.8121763Z Requirement already satisfied: nvidia-cuda-nvrtc-cu12==12.6.77 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (12.6.77) 2025-09-09T14:11:03.8123139Z Requirement already satisfied: nvidia-cuda-runtime-cu12==12.6.77 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (12.6.77) 2025-09-09T14:11:03.8124508Z Requirement already satisfied: nvidia-cuda-cupti-cu12==12.6.80 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (12.6.80) 2025-09-09T14:11:03.8125842Z Requirement already satisfied: nvidia-cudnn-cu12==9.5.1.17 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (9.5.1.17) 2025-09-09T14:11:03.8127176Z Requirement already satisfied: nvidia-cublas-cu12==12.6.4.1 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (12.6.4.1) 2025-09-09T14:11:03.8128499Z Requirement already satisfied: nvidia-cufft-cu12==11.3.0.4 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (11.3.0.4) 2025-09-09T14:11:03.8129826Z Requirement already satisfied: nvidia-curand-cu12==10.3.7.77 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (10.3.7.77) 2025-09-09T14:11:03.8131174Z Requirement already satisfied: nvidia-cusolver-cu12==11.7.1.2 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (11.7.1.2) 2025-09-09T14:11:16.8466732Z Requirement already satisfied: nvidia-cusparse-cu12==12.5.4.2 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (12.5.4.2) 2025-09-09T14:11:16.8468231Z Requirement already satisfied: nvidia-cusparselt-cu12==0.6.3 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (0.6.3) 2025-09-09T14:11:16.8469568Z Requirement already satisfied: nvidia-nccl-cu12==2.26.2 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (2.26.2) 2025-09-09T14:11:16.8470900Z Requirement already satisfied: nvidia-nvtx-cu12==12.6.77 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (12.6.77) 2025-09-09T14:11:16.8472266Z Requirement already satisfied: nvidia-nvjitlink-cu12==12.6.85 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (12.6.85) 2025-09-09T14:11:16.8473627Z Requirement already satisfied: nvidia-cufile-cu12==1.11.1.6 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (1.11.1.6) 2025-09-09T14:11:16.8475333Z Requirement already satisfied: triton==3.3.0 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (3.3.0) 2025-09-09T14:11:16.8476806Z Requirement already satisfied: setuptools>=40.8.0 in /opt/conda/envs/venv/lib/python3.9/site-packages (from triton==3.3.0->torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (78.1.1) 2025-09-09T14:11:16.8477757Z Collecting contourpy>=1.0.1 (from matplotlib->-r dev-requirements.txt (line 13)) 2025-09-09T14:11:16.8478419Z Downloading contourpy-1.3.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (5.4 kB) 2025-09-09T14:11:16.8479059Z Collecting cycler>=0.10 (from matplotlib->-r dev-requirements.txt (line 13)) 2025-09-09T14:11:16.8479570Z Downloading cycler-0.12.1-py3-none-any.whl.metadata (3.8 kB) 2025-09-09T14:11:16.8480093Z Collecting fonttools>=4.22.0 (from matplotlib->-r dev-requirements.txt (line 13)) 2025-09-09T14:11:16.8480759Z Downloading fonttools-4.59.2-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (109 kB) 2025-09-09T14:11:16.8481427Z Collecting kiwisolver>=1.3.1 (from matplotlib->-r dev-requirements.txt (line 13)) 2025-09-09T14:11:16.8482087Z Downloading kiwisolver-1.4.7-cp39-cp39-manylinux_2_12_x86_64.manylinux2010_x86_64.whl.metadata (6.3 kB) 2025-09-09T14:11:16.8482732Z Collecting pillow>=8 (from matplotlib->-r dev-requirements.txt (line 13)) 2025-09-09T14:11:16.8483350Z Downloading pillow-11.3.0-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl.metadata (9.0 kB) 2025-09-09T14:11:16.8483998Z Collecting pyparsing>=2.3.1 (from matplotlib->-r dev-requirements.txt (line 13)) 2025-09-09T14:11:16.8484535Z Downloading pyparsing-3.2.3-py3-none-any.whl.metadata (5.0 kB) 2025-09-09T14:11:16.8485087Z Collecting python-dateutil>=2.7 (from matplotlib->-r dev-requirements.txt (line 13)) 2025-09-09T14:11:16.8485707Z Downloading python_dateutil-2.9.0.post0-py2.py3-none-any.whl.metadata (8.4 kB) 2025-09-09T14:11:16.8486388Z Collecting importlib-resources>=3.2.0 (from matplotlib->-r dev-requirements.txt (line 13)) 2025-09-09T14:11:16.8487014Z Downloading importlib_resources-6.5.2-py3-none-any.whl.metadata (3.9 kB) 2025-09-09T14:11:16.8487564Z Collecting pytz>=2020.1 (from pandas->-r dev-requirements.txt (line 14)) 2025-09-09T14:11:16.8488060Z Downloading pytz-2025.2-py2.py3-none-any.whl.metadata (22 kB) 2025-09-09T14:11:16.8488570Z Collecting tzdata>=2022.7 (from pandas->-r dev-requirements.txt (line 14)) 2025-09-09T14:11:16.8489081Z Downloading tzdata-2025.2-py2.py3-none-any.whl.metadata (1.4 kB) 2025-09-09T14:11:16.8489579Z Collecting termcolor (from fire->-r dev-requirements.txt (line 15)) 2025-09-09T14:11:16.8490068Z Downloading termcolor-3.1.0-py3-none-any.whl.metadata (6.4 kB) 2025-09-09T14:11:16.8490603Z Collecting pycryptodomex>=3.8 (from blobfile->-r dev-requirements.txt (line 18)) 2025-09-09T14:11:16.8491289Z Downloading pycryptodomex-3.23.0-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (3.4 kB) 2025-09-09T14:11:16.8491972Z Collecting urllib3<3,>=1.25.3 (from blobfile->-r dev-requirements.txt (line 18)) 2025-09-09T14:11:16.8492481Z Downloading urllib3-2.5.0-py3-none-any.whl.metadata (6.5 kB) 2025-09-09T14:11:16.8500836Z Collecting accelerate>=0.26.0 (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:16.8501392Z Downloading accelerate-1.10.1-py3-none-any.whl.metadata (19 kB) 2025-09-09T14:11:16.8501904Z Collecting evaluate (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:16.8502408Z Downloading evaluate-0.4.5-py3-none-any.whl.metadata (9.5 kB) 2025-09-09T14:11:16.8502918Z Collecting datasets<4.0,>=2.16.0 (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:16.8503443Z Downloading datasets-3.6.0-py3-none-any.whl.metadata (19 kB) 2025-09-09T14:11:16.8504005Z Collecting jsonlines (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:16.8504581Z Downloading jsonlines-4.0.0-py3-none-any.whl.metadata (1.6 kB) 2025-09-09T14:11:16.8505314Z Collecting numexpr (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:16.8505926Z Downloading numexpr-2.10.2-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl.metadata (8.1 kB) 2025-09-09T14:11:16.8506671Z Collecting peft>=0.2.0 (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:16.8507136Z Downloading peft-0.17.1-py3-none-any.whl.metadata (14 kB) 2025-09-09T14:11:16.8507630Z Collecting pybind11>=2.6.2 (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:16.8508136Z Downloading pybind11-3.0.1-py3-none-any.whl.metadata (10.0 kB) 2025-09-09T14:11:16.8508659Z Collecting pytablewriter (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:16.8509210Z Downloading pytablewriter-1.2.1-py3-none-any.whl.metadata (38 kB) 2025-09-09T14:11:16.8509741Z Collecting rouge-score>=0.0.4 (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:16.8510193Z Downloading rouge_score-0.1.2.tar.gz (17 kB) 2025-09-09T14:11:16.8510784Z Preparing metadata (setup.py) ... [?25l- done 2025-09-09T14:11:16.8511352Z [?25hCollecting sacrebleu>=1.5.0 (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:16.8511886Z Downloading sacrebleu-2.5.1-py3-none-any.whl.metadata (51 kB) 2025-09-09T14:11:16.8512423Z Collecting scikit-learn>=0.24.1 (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:16.8513093Z Downloading scikit_learn-1.6.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (18 kB) 2025-09-09T14:11:16.8513731Z Collecting sqlitedict (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:16.8514179Z Downloading sqlitedict-2.1.0.tar.gz (21 kB) 2025-09-09T14:11:16.8514585Z Preparing metadata (setup.py) ... [?25l- done 2025-09-09T14:11:16.8515152Z [?25hCollecting tqdm-multiprocess (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:16.8515740Z Downloading tqdm_multiprocess-0.0.11-py3-none-any.whl.metadata (5.7 kB) 2025-09-09T14:11:16.8516336Z Collecting zstandard (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:16.8517005Z Downloading zstandard-0.24.0-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (3.1 kB) 2025-09-09T14:11:16.8517632Z Collecting dill (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:16.8518097Z Downloading dill-0.4.0-py3-none-any.whl.metadata (10 kB) 2025-09-09T14:11:16.8518580Z Collecting word2number (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:16.8519017Z Downloading word2number-1.1.zip (9.7 kB) 2025-09-09T14:11:16.8519409Z Preparing metadata (setup.py) ... [?25l- done 2025-09-09T14:11:16.8519951Z [?25hCollecting more_itertools (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:16.8520502Z Downloading more_itertools-10.8.0-py3-none-any.whl.metadata (39 kB) 2025-09-09T14:11:16.8521080Z Collecting pyarrow>=15.0.0 (from datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:16.8521694Z Downloading pyarrow-21.0.0-cp39-cp39-manylinux_2_28_x86_64.whl.metadata (3.3 kB) 2025-09-09T14:11:16.8522227Z Collecting dill (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:16.8522686Z Downloading dill-0.3.8-py3-none-any.whl.metadata (10 kB) 2025-09-09T14:11:16.8523206Z Collecting xxhash (from datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:16.8523862Z Downloading xxhash-3.5.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (12 kB) 2025-09-09T14:11:16.8524560Z Collecting multiprocess<0.70.17 (from datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:16.8525188Z Downloading multiprocess-0.70.16-py39-none-any.whl.metadata (7.2 kB) 2025-09-09T14:11:16.8525833Z Collecting fsspec>=2023.5.0 (from huggingface-hub<1.0,>=0.34.0->transformers->-r dev-requirements.txt (line 6)) 2025-09-09T14:11:16.8526470Z Downloading fsspec-2025.3.0-py3-none-any.whl.metadata (11 kB) 2025-09-09T14:11:16.8527335Z Collecting aiohttp!=4.0.0a0,!=4.0.0a1 (from fsspec[http]<=2025.3.0,>=2023.1.0->datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:16.8528135Z Downloading aiohttp-3.12.15-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (7.7 kB) 2025-09-09T14:11:16.8528853Z Collecting zipp>=3.20 (from importlib_metadata->-r dev-requirements.txt (line 24)) 2025-09-09T14:11:16.8529364Z Downloading zipp-3.23.0-py3-none-any.whl.metadata (3.6 kB) 2025-09-09T14:11:16.8529841Z Collecting cfgv>=2.0.0 (from pre-commit->-r dev-requirements.txt (line 34)) 2025-09-09T14:11:16.8530334Z Downloading cfgv-3.4.0-py2.py3-none-any.whl.metadata (8.5 kB) 2025-09-09T14:11:16.8530847Z Collecting identify>=1.0.0 (from pre-commit->-r dev-requirements.txt (line 34)) 2025-09-09T14:11:16.8531384Z Downloading identify-2.6.14-py2.py3-none-any.whl.metadata (4.4 kB) 2025-09-09T14:11:16.8531912Z Collecting nodeenv>=0.11.1 (from pre-commit->-r dev-requirements.txt (line 34)) 2025-09-09T14:11:16.8532438Z Downloading nodeenv-1.9.1-py2.py3-none-any.whl.metadata (21 kB) 2025-09-09T14:11:16.8532982Z Collecting virtualenv>=20.10.0 (from pre-commit->-r dev-requirements.txt (line 34)) 2025-09-09T14:11:16.8533522Z Downloading virtualenv-20.34.0-py3-none-any.whl.metadata (4.6 kB) 2025-09-09T14:11:16.8534095Z Collecting psutil (from accelerate>=0.26.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5168811Z Downloading psutil-7.0.0-cp36-abi3-manylinux_2_12_x86_64.manylinux2010_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (22 kB) 2025-09-09T14:11:23.5170233Z Collecting aiohappyeyeballs>=2.5.0 (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2025.3.0,>=2023.1.0->datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5171317Z Downloading aiohappyeyeballs-2.6.1-py3-none-any.whl.metadata (5.9 kB) 2025-09-09T14:11:23.5172110Z Collecting aiosignal>=1.4.0 (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2025.3.0,>=2023.1.0->datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5173106Z Downloading aiosignal-1.4.0-py3-none-any.whl.metadata (3.7 kB) 2025-09-09T14:11:23.5174108Z Collecting async-timeout<6.0,>=4.0 (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2025.3.0,>=2023.1.0->datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5174864Z Downloading async_timeout-5.0.1-py3-none-any.whl.metadata (5.1 kB) 2025-09-09T14:11:23.5175684Z Collecting frozenlist>=1.1.1 (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2025.3.0,>=2023.1.0->datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5176733Z Downloading frozenlist-1.7.0-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (18 kB) 2025-09-09T14:11:23.5177687Z Collecting multidict<7.0,>=4.5 (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2025.3.0,>=2023.1.0->datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5178599Z Downloading multidict-6.6.4-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl.metadata (5.3 kB) 2025-09-09T14:11:23.5179503Z Collecting propcache>=0.2.0 (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2025.3.0,>=2023.1.0->datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5180340Z Downloading propcache-0.3.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (12 kB) 2025-09-09T14:11:23.5181157Z Collecting yarl<2.0,>=1.17.0 (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2025.3.0,>=2023.1.0->datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5181966Z Downloading yarl-1.20.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (73 kB) 2025-09-09T14:11:23.5182772Z Collecting idna>=2.0 (from yarl<2.0,>=1.17.0->aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2025.3.0,>=2023.1.0->datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5183458Z Downloading idna-3.10-py3-none-any.whl.metadata (10 kB) 2025-09-09T14:11:23.5184213Z Collecting six>=1.5 (from python-dateutil>=2.7->matplotlib->-r dev-requirements.txt (line 13)) 2025-09-09T14:11:23.5184879Z Downloading six-1.17.0-py2.py3-none-any.whl.metadata (1.7 kB) 2025-09-09T14:11:23.5185690Z Collecting charset_normalizer<4,>=2 (from requests->transformers->-r dev-requirements.txt (line 6)) 2025-09-09T14:11:23.5186675Z Downloading charset_normalizer-3.4.3-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl.metadata (36 kB) 2025-09-09T14:11:23.5187623Z Collecting certifi>=2017.4.17 (from requests->transformers->-r dev-requirements.txt (line 6)) 2025-09-09T14:11:23.5188286Z Downloading certifi-2025.8.3-py3-none-any.whl.metadata (2.4 kB) 2025-09-09T14:11:23.5188916Z Collecting absl-py (from rouge-score>=0.0.4->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5189529Z Downloading absl_py-2.3.1-py3-none-any.whl.metadata (3.3 kB) 2025-09-09T14:11:23.5190126Z Collecting nltk (from rouge-score>=0.0.4->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5190731Z Downloading nltk-3.9.1-py3-none-any.whl.metadata (2.9 kB) 2025-09-09T14:11:23.5191345Z Collecting portalocker (from sacrebleu>=1.5.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5192028Z Downloading portalocker-3.2.0-py3-none-any.whl.metadata (8.7 kB) 2025-09-09T14:11:23.5192666Z Collecting colorama (from sacrebleu>=1.5.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5193316Z Downloading colorama-0.4.6-py2.py3-none-any.whl.metadata (17 kB) 2025-09-09T14:11:23.5193976Z Collecting scipy>=1.6.0 (from scikit-learn>=0.24.1->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5194751Z Downloading scipy-1.13.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (60 kB) 2025-09-09T14:11:23.5195532Z Collecting joblib>=1.2.0 (from scikit-learn>=0.24.1->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5196169Z Downloading joblib-1.5.2-py3-none-any.whl.metadata (5.6 kB) 2025-09-09T14:11:23.5196850Z Collecting threadpoolctl>=3.1.0 (from scikit-learn>=0.24.1->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5197740Z Downloading threadpoolctl-3.6.0-py3-none-any.whl.metadata (13 kB) 2025-09-09T14:11:23.5198855Z Requirement already satisfied: mpmath<1.4,>=1.1.0 in /opt/conda/envs/venv/lib/python3.9/site-packages (from sympy>=1.13.3->torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (1.3.0) 2025-09-09T14:11:23.5200042Z Collecting distlib<1,>=0.3.7 (from virtualenv>=20.10.0->pre-commit->-r dev-requirements.txt (line 34)) 2025-09-09T14:11:23.5200726Z Downloading distlib-0.4.0-py2.py3-none-any.whl.metadata (5.2 kB) 2025-09-09T14:11:23.5201436Z Collecting platformdirs<5,>=3.9.1 (from virtualenv>=20.10.0->pre-commit->-r dev-requirements.txt (line 34)) 2025-09-09T14:11:23.5202162Z Downloading platformdirs-4.4.0-py3-none-any.whl.metadata (12 kB) 2025-09-09T14:11:23.5203263Z Requirement already satisfied: MarkupSafe>=2.0 in /opt/conda/envs/venv/lib/python3.9/site-packages (from jinja2->torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (3.0.2) 2025-09-09T14:11:23.5204427Z Collecting click (from nltk->rouge-score>=0.0.4->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5205061Z Downloading click-8.1.8-py3-none-any.whl.metadata (2.3 kB) 2025-09-09T14:11:23.5205715Z Collecting DataProperty<2,>=1.1.0 (from pytablewriter->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5206412Z Downloading DataProperty-1.1.0-py3-none-any.whl.metadata (11 kB) 2025-09-09T14:11:23.5207105Z Collecting mbstrdecoder<2,>=1.0.0 (from pytablewriter->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5207811Z Downloading mbstrdecoder-1.1.4-py3-none-any.whl.metadata (4.3 kB) 2025-09-09T14:11:23.5208503Z Collecting pathvalidate<4,>=2.3.0 (from pytablewriter->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5209204Z Downloading pathvalidate-3.3.1-py3-none-any.whl.metadata (12 kB) 2025-09-09T14:11:23.5210011Z Collecting tabledata<2,>=1.3.1 (from pytablewriter->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5210683Z Downloading tabledata-1.3.4-py3-none-any.whl.metadata (3.7 kB) 2025-09-09T14:11:23.5211353Z Collecting tcolorpy<1,>=0.0.5 (from pytablewriter->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5211914Z Downloading tcolorpy-0.1.7-py3-none-any.whl.metadata (6.3 kB) 2025-09-09T14:11:23.5212532Z Collecting typepy<2,>=1.3.2 (from typepy[datetime]<2,>=1.3.2->pytablewriter->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5213143Z Downloading typepy-1.3.4-py3-none-any.whl.metadata (9.2 kB) 2025-09-09T14:11:23.5213750Z Collecting chardet<6,>=3.0.4 (from mbstrdecoder<2,>=1.0.0->pytablewriter->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:23.5214355Z Downloading chardet-5.2.0-py3-none-any.whl.metadata (3.4 kB) 2025-09-09T14:11:23.5214896Z Downloading ruff-0.11.6-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (11.5 MB) 2025-09-09T14:11:23.5215693Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/11.5 MB ? eta -:--:-- 2025-09-09T14:11:23.5216366Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 11.5/11.5 MB 145.3 MB/s 0:00:00 2025-09-09T14:11:23.5217055Z [?25hDownloading cmake-3.31.6-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (27.8 MB) 2025-09-09T14:11:23.5217729Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/27.8 MB ? eta -:--:-- 2025-09-09T14:11:23.5218332Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 27.8/27.8 MB 154.7 MB/s 0:00:00 2025-09-09T14:11:23.5218845Z [?25hDownloading pytest-8.4.2-py3-none-any.whl (365 kB) 2025-09-09T14:11:23.5219226Z Downloading pluggy-1.6.0-py3-none-any.whl (20 kB) 2025-09-09T14:11:23.5219667Z Downloading unittest_xml_reporting-3.2.0-py2.py3-none-any.whl (20 kB) 2025-09-09T14:11:23.5220155Z Downloading parameterized-0.9.0-py2.py3-none-any.whl (20 kB) 2025-09-09T14:11:23.5220596Z Downloading packaging-25.0-py3-none-any.whl (66 kB) 2025-09-09T14:11:23.5221006Z Downloading transformers-4.56.1-py3-none-any.whl (11.6 MB) 2025-09-09T14:11:23.5221576Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/11.6 MB ? eta -:--:-- 2025-09-09T14:11:23.5222197Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 11.6/11.6 MB 223.0 MB/s 0:00:00 2025-09-09T14:11:23.5222748Z [?25hDownloading huggingface_hub-0.34.4-py3-none-any.whl (561 kB) 2025-09-09T14:11:23.5223335Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/561.5 kB ? eta -:--:-- 2025-09-09T14:11:23.5223941Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 561.5/561.5 kB 57.5 MB/s 0:00:00 2025-09-09T14:11:24.5621403Z [?25hDownloading hf_xet-1.1.9-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (3.2 MB) 2025-09-09T14:11:24.5622435Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/3.2 MB ? eta -:--:-- 2025-09-09T14:11:24.5623248Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 3.2/3.2 MB 182.1 MB/s 0:00:00 2025-09-09T14:11:24.5624188Z [?25hDownloading tokenizers-0.22.0-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (3.3 MB) 2025-09-09T14:11:24.5625163Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/3.3 MB ? eta -:--:-- 2025-09-09T14:11:24.5625778Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 3.3/3.3 MB 166.7 MB/s 0:00:00 2025-09-09T14:11:24.5626315Z [?25hDownloading hypothesis-6.138.15-py3-none-any.whl (533 kB) 2025-09-09T14:11:24.5626901Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/533.6 kB ? eta -:--:-- 2025-09-09T14:11:24.5627513Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 533.6/533.6 kB 52.0 MB/s 0:00:00 2025-09-09T14:11:24.5628290Z [?25hDownloading sortedcontainers-2.4.0-py2.py3-none-any.whl (29 kB) 2025-09-09T14:11:24.5628921Z Downloading sentencepiece-0.2.1-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (1.4 MB) 2025-09-09T14:11:24.5629825Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/1.4 MB ? eta -:--:-- 2025-09-09T14:11:24.5630428Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.4/1.4 MB 121.6 MB/s 0:00:00 2025-09-09T14:11:24.5630947Z [?25hDownloading expecttest-0.3.0-py3-none-any.whl (8.2 kB) 2025-09-09T14:11:24.5631460Z Downloading bitsandbytes-0.47.0-py3-none-manylinux_2_24_x86_64.whl (61.3 MB) 2025-09-09T14:11:24.5632093Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/61.3 MB ? eta -:--:-- 2025-09-09T14:11:24.5632768Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━ 46.9/61.3 MB 235.0 MB/s eta 0:00:01 2025-09-09T14:11:24.5633446Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 61.3/61.3 MB 154.6 MB/s 0:00:00 2025-09-09T14:11:24.5634144Z [?25hDownloading matplotlib-3.9.4-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (8.3 MB) 2025-09-09T14:11:24.5634870Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/8.3 MB ? eta -:--:-- 2025-09-09T14:11:24.5635468Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 8.3/8.3 MB 183.1 MB/s 0:00:00 2025-09-09T14:11:24.5636141Z [?25hDownloading pandas-2.3.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (12.4 MB) 2025-09-09T14:11:24.5636841Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/12.4 MB ? eta -:--:-- 2025-09-09T14:11:24.5637444Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 12.4/12.4 MB 200.0 MB/s 0:00:00 2025-09-09T14:11:24.5637954Z [?25hDownloading fire-0.7.1-py3-none-any.whl (115 kB) 2025-09-09T14:11:24.5638363Z Downloading tabulate-0.9.0-py3-none-any.whl (35 kB) 2025-09-09T14:11:24.5638920Z Downloading tiktoken-0.11.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.2 MB) 2025-09-09T14:11:24.5639652Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/1.2 MB ? eta -:--:-- 2025-09-09T14:11:24.5640248Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.2/1.2 MB 125.6 MB/s 0:00:00 2025-09-09T14:11:24.5640765Z [?25hDownloading blobfile-3.1.0-py3-none-any.whl (75 kB) 2025-09-09T14:11:24.5641166Z Downloading urllib3-2.5.0-py3-none-any.whl (129 kB) 2025-09-09T14:11:24.5641558Z Downloading lm_eval-0.4.9.1-py3-none-any.whl (7.5 MB) 2025-09-09T14:11:24.5642098Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/7.5 MB ? eta -:--:-- 2025-09-09T14:11:24.5642696Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7.5/7.5 MB 166.1 MB/s 0:00:00 2025-09-09T14:11:24.5643313Z [?25hDownloading datasets-3.6.0-py3-none-any.whl (491 kB) 2025-09-09T14:11:24.5643709Z Downloading dill-0.3.8-py3-none-any.whl (116 kB) 2025-09-09T14:11:24.5644207Z Downloading fsspec-2025.3.0-py3-none-any.whl (193 kB) 2025-09-09T14:11:24.5644632Z Downloading multiprocess-0.70.16-py39-none-any.whl (133 kB) 2025-09-09T14:11:24.5645065Z Downloading diskcache-5.6.3-py3-none-any.whl (45 kB) 2025-09-09T14:11:24.5645610Z Downloading pycocotools-2.0.10-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (453 kB) 2025-09-09T14:11:24.5646154Z Downloading tqdm-4.67.1-py3-none-any.whl (78 kB) 2025-09-09T14:11:24.5646577Z Downloading importlib_metadata-8.7.0-py3-none-any.whl (27 kB) 2025-09-09T14:11:24.5647125Z Downloading ninja-1.13.0-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (180 kB) 2025-09-09T14:11:24.5647669Z Downloading pre_commit-4.3.0-py2.py3-none-any.whl (220 kB) 2025-09-09T14:11:24.5648104Z Downloading accelerate-1.10.1-py3-none-any.whl (374 kB) 2025-09-09T14:11:24.5648626Z Downloading numpy-2.0.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (19.5 MB) 2025-09-09T14:11:24.5649312Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/19.5 MB ? eta -:--:-- 2025-09-09T14:11:24.5649933Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 19.5/19.5 MB 231.6 MB/s 0:00:00 2025-09-09T14:11:24.5650623Z [?25hDownloading aiohttp-3.12.15-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.6 MB) 2025-09-09T14:11:24.5651321Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/1.6 MB ? eta -:--:-- 2025-09-09T14:11:24.5651916Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.6/1.6 MB 136.1 MB/s 0:00:00 2025-09-09T14:11:24.5652457Z [?25hDownloading async_timeout-5.0.1-py3-none-any.whl (6.2 kB) 2025-09-09T14:11:24.5653092Z Downloading multidict-6.6.4-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl (239 kB) 2025-09-09T14:11:24.5653833Z Downloading yarl-1.20.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (327 kB) 2025-09-09T14:11:24.5654383Z Downloading aiohappyeyeballs-2.6.1-py3-none-any.whl (15 kB) 2025-09-09T14:11:24.5654841Z Downloading aiosignal-1.4.0-py3-none-any.whl (7.5 kB) 2025-09-09T14:11:24.5655232Z Downloading attrs-25.3.0-py3-none-any.whl (63 kB) 2025-09-09T14:11:24.5655619Z Downloading cfgv-3.4.0-py2.py3-none-any.whl (7.2 kB) 2025-09-09T14:11:24.5656243Z Downloading contourpy-1.3.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (321 kB) 2025-09-09T14:11:24.5656770Z Downloading cycler-0.12.1-py3-none-any.whl (8.3 kB) 2025-09-09T14:11:24.5657161Z Downloading evaluate-0.4.5-py3-none-any.whl (84 kB) 2025-09-09T14:11:24.5657575Z Downloading exceptiongroup-1.3.0-py3-none-any.whl (16 kB) 2025-09-09T14:11:24.5658137Z Downloading fonttools-4.59.2-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (4.8 MB) 2025-09-09T14:11:24.5658946Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/4.8 MB ? eta -:--:-- 2025-09-09T14:11:24.5659553Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 4.8/4.8 MB 123.8 MB/s 0:00:00 2025-09-09T14:11:24.5660496Z [?25hDownloading frozenlist-1.7.0-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl (225 kB) 2025-09-09T14:11:24.5661185Z Downloading identify-2.6.14-py2.py3-none-any.whl (99 kB) 2025-09-09T14:11:24.5661587Z Downloading idna-3.10-py3-none-any.whl (70 kB) 2025-09-09T14:11:24.5662001Z Downloading importlib_resources-6.5.2-py3-none-any.whl (37 kB) 2025-09-09T14:11:24.5662439Z Downloading iniconfig-2.1.0-py3-none-any.whl (6.0 kB) 2025-09-09T14:11:24.5662979Z Downloading kiwisolver-1.4.7-cp39-cp39-manylinux_2_12_x86_64.manylinux2010_x86_64.whl (1.6 MB) 2025-09-09T14:11:25.4874876Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/1.6 MB ? eta -:--:-- 2025-09-09T14:11:25.4875797Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.6/1.6 MB 139.1 MB/s 0:00:00 2025-09-09T14:11:25.4876609Z [?25hDownloading lxml-6.0.1-cp39-cp39-manylinux_2_26_x86_64.manylinux_2_28_x86_64.whl (5.3 MB) 2025-09-09T14:11:25.4877310Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/5.3 MB ? eta -:--:-- 2025-09-09T14:11:25.4877915Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 5.3/5.3 MB 191.3 MB/s 0:00:00 2025-09-09T14:11:25.4878448Z [?25hDownloading nodeenv-1.9.1-py2.py3-none-any.whl (22 kB) 2025-09-09T14:11:25.4878879Z Downloading peft-0.17.1-py3-none-any.whl (504 kB) 2025-09-09T14:11:25.4879424Z Downloading pillow-11.3.0-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (6.6 MB) 2025-09-09T14:11:25.4880098Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/6.6 MB ? eta -:--:-- 2025-09-09T14:11:25.4880717Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 6.6/6.6 MB 210.7 MB/s 0:00:00 2025-09-09T14:11:25.4881413Z [?25hDownloading propcache-0.3.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (200 kB) 2025-09-09T14:11:25.4882047Z Downloading pyarrow-21.0.0-cp39-cp39-manylinux_2_28_x86_64.whl (42.7 MB) 2025-09-09T14:11:25.4882675Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/42.7 MB ? eta -:--:-- 2025-09-09T14:11:25.4883278Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 42.7/42.7 MB 223.9 MB/s 0:00:00 2025-09-09T14:11:25.4883811Z [?25hDownloading pybind11-3.0.1-py3-none-any.whl (293 kB) 2025-09-09T14:11:25.4884379Z Downloading pycryptodomex-3.23.0-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (2.3 MB) 2025-09-09T14:11:25.4885099Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/2.3 MB ? eta -:--:-- 2025-09-09T14:11:25.4885914Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 2.3/2.3 MB 169.3 MB/s 0:00:00 2025-09-09T14:11:25.4886446Z [?25hDownloading pygments-2.19.2-py3-none-any.whl (1.2 MB) 2025-09-09T14:11:25.4887158Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/1.2 MB ? eta -:--:-- 2025-09-09T14:11:25.4887761Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.2/1.2 MB 124.0 MB/s 0:00:00 2025-09-09T14:11:25.4888288Z [?25hDownloading pyparsing-3.2.3-py3-none-any.whl (111 kB) 2025-09-09T14:11:25.4888781Z Downloading python_dateutil-2.9.0.post0-py2.py3-none-any.whl (229 kB) 2025-09-09T14:11:25.4889249Z Downloading pytz-2025.2-py2.py3-none-any.whl (509 kB) 2025-09-09T14:11:25.4889770Z Downloading PyYAML-6.0.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (737 kB) 2025-09-09T14:11:25.4890448Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/737.4 kB ? eta -:--:-- 2025-09-09T14:11:25.4891084Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 737.4/737.4 kB 81.1 MB/s 0:00:00 2025-09-09T14:11:25.4891873Z [?25hDownloading regex-2025.9.1-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl (789 kB) 2025-09-09T14:11:25.4892646Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/789.5 kB ? eta -:--:-- 2025-09-09T14:11:25.4893264Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 789.5/789.5 kB 48.9 MB/s 0:00:00 2025-09-09T14:11:25.4893787Z [?25hDownloading requests-2.32.5-py3-none-any.whl (64 kB) 2025-09-09T14:11:25.4894448Z Downloading charset_normalizer-3.4.3-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl (152 kB) 2025-09-09T14:11:25.4895099Z Downloading certifi-2025.8.3-py3-none-any.whl (161 kB) 2025-09-09T14:11:25.4895505Z Downloading sacrebleu-2.5.1-py3-none-any.whl (104 kB) 2025-09-09T14:11:25.4896139Z Downloading safetensors-0.6.2-cp38-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (485 kB) 2025-09-09T14:11:25.4896822Z Downloading scikit_learn-1.6.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (13.5 MB) 2025-09-09T14:11:25.4897710Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/13.5 MB ? eta -:--:-- 2025-09-09T14:11:25.4898319Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 13.5/13.5 MB 218.7 MB/s 0:00:00 2025-09-09T14:11:25.4898854Z [?25hDownloading joblib-1.5.2-py3-none-any.whl (308 kB) 2025-09-09T14:11:25.4899401Z Downloading scipy-1.13.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (38.6 MB) 2025-09-09T14:11:25.4900075Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/38.6 MB ? eta -:--:-- 2025-09-09T14:11:25.4900717Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 38.5/38.6 MB 440.3 MB/s eta 0:00:01 2025-09-09T14:11:25.4901515Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 38.6/38.6 MB 185.1 MB/s 0:00:00 2025-09-09T14:11:25.4902044Z [?25hDownloading six-1.17.0-py2.py3-none-any.whl (11 kB) 2025-09-09T14:11:25.4902613Z Downloading threadpoolctl-3.6.0-py3-none-any.whl (18 kB) 2025-09-09T14:11:25.4903012Z Downloading tomli-2.2.1-py3-none-any.whl (14 kB) 2025-09-09T14:11:25.4903404Z Downloading tzdata-2025.2-py2.py3-none-any.whl (347 kB) 2025-09-09T14:11:25.4903817Z Downloading virtualenv-20.34.0-py3-none-any.whl (6.0 MB) 2025-09-09T14:11:25.4904381Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/6.0 MB ? eta -:--:-- 2025-09-09T14:11:25.4904976Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 6.0/6.0 MB 193.4 MB/s 0:00:00 2025-09-09T14:11:25.4905509Z [?25hDownloading distlib-0.4.0-py2.py3-none-any.whl (469 kB) 2025-09-09T14:11:25.4905953Z Downloading platformdirs-4.4.0-py3-none-any.whl (18 kB) 2025-09-09T14:11:25.4906343Z Downloading zipp-3.23.0-py3-none-any.whl (10 kB) 2025-09-09T14:11:25.4906741Z Downloading absl_py-2.3.1-py3-none-any.whl (135 kB) 2025-09-09T14:11:25.4907128Z Downloading colorama-0.4.6-py2.py3-none-any.whl (25 kB) 2025-09-09T14:11:25.4907547Z Downloading jsonlines-4.0.0-py3-none-any.whl (8.7 kB) 2025-09-09T14:11:25.4907962Z Downloading more_itertools-10.8.0-py3-none-any.whl (69 kB) 2025-09-09T14:11:25.4908371Z Downloading nltk-3.9.1-py3-none-any.whl (1.5 MB) 2025-09-09T14:11:25.4908948Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/1.5 MB ? eta -:--:-- 2025-09-09T14:11:25.4909580Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.5/1.5 MB 132.8 MB/s 0:00:00 2025-09-09T14:11:25.4910086Z [?25hDownloading click-8.1.8-py3-none-any.whl (98 kB) 2025-09-09T14:11:25.4910609Z Downloading numexpr-2.10.2-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (396 kB) 2025-09-09T14:11:25.4911165Z Downloading portalocker-3.2.0-py3-none-any.whl (22 kB) 2025-09-09T14:11:25.4911845Z Downloading psutil-7.0.0-cp36-abi3-manylinux_2_12_x86_64.manylinux2010_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl (277 kB) 2025-09-09T14:11:25.4912530Z Downloading pytablewriter-1.2.1-py3-none-any.whl (91 kB) 2025-09-09T14:11:25.4912964Z Downloading DataProperty-1.1.0-py3-none-any.whl (27 kB) 2025-09-09T14:11:25.4913383Z Downloading mbstrdecoder-1.1.4-py3-none-any.whl (7.9 kB) 2025-09-09T14:11:25.4913801Z Downloading chardet-5.2.0-py3-none-any.whl (199 kB) 2025-09-09T14:11:25.4914211Z Downloading pathvalidate-3.3.1-py3-none-any.whl (24 kB) 2025-09-09T14:11:25.4914630Z Downloading tabledata-1.3.4-py3-none-any.whl (11 kB) 2025-09-09T14:11:25.4915037Z Downloading tcolorpy-0.1.7-py3-none-any.whl (8.1 kB) 2025-09-09T14:11:25.4915417Z Downloading typepy-1.3.4-py3-none-any.whl (31 kB) 2025-09-09T14:11:25.4915816Z Downloading termcolor-3.1.0-py3-none-any.whl (7.7 kB) 2025-09-09T14:11:25.4916249Z Downloading tqdm_multiprocess-0.0.11-py3-none-any.whl (9.8 kB) 2025-09-09T14:11:31.1981075Z Downloading xxhash-3.5.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (193 kB) 2025-09-09T14:11:31.1981968Z Downloading zstandard-0.24.0-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (5.6 MB) 2025-09-09T14:11:31.1983081Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/5.6 MB ? eta -:--:-- 2025-09-09T14:11:31.1983691Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 5.6/5.6 MB 123.3 MB/s 0:00:00 2025-09-09T14:11:31.1984303Z [?25hBuilding wheels for collected packages: rouge-score, sqlitedict, word2number 2025-09-09T14:11:31.1986720Z  DEPRECATION: Building 'rouge-score' using the legacy setup.py bdist_wheel mechanism, which will be removed in a future version. pip 25.3 will enforce this behaviour change. A possible replacement is to use the standardized build interface by setting the `--use-pep517` option, (possibly combined with `--no-build-isolation`), or adding a `pyproject.toml` file to the source tree of 'rouge-score'. Discussion can be found at https://github.com/pypa/pip/issues/6334 2025-09-09T14:11:31.1988581Z  Building wheel for rouge-score (setup.py) ... [?25l- done 2025-09-09T14:11:31.1989694Z [?25h Created wheel for rouge-score: filename=rouge_score-0.1.2-py3-none-any.whl size=24988 sha256=a7b33d153faeeed78eac2d44e17098ec47b15234002e699e2fa9c3a2e5d40bb2 2025-09-09T14:11:31.1990651Z Stored in directory: /root/.cache/pip/wheels/9b/3d/39/09558097d3119ca0a4d462df68f22c6f3c1b345ac63a09b86e 2025-09-09T14:11:31.1992930Z  DEPRECATION: Building 'sqlitedict' using the legacy setup.py bdist_wheel mechanism, which will be removed in a future version. pip 25.3 will enforce this behaviour change. A possible replacement is to use the standardized build interface by setting the `--use-pep517` option, (possibly combined with `--no-build-isolation`), or adding a `pyproject.toml` file to the source tree of 'sqlitedict'. Discussion can be found at https://github.com/pypa/pip/issues/6334 2025-09-09T14:11:31.1994762Z  Building wheel for sqlitedict (setup.py) ... [?25l- done 2025-09-09T14:11:31.1995677Z [?25h Created wheel for sqlitedict: filename=sqlitedict-2.1.0-py3-none-any.whl size=16958 sha256=654485bedfbee49afdd0e3fe0d13712da596a9130755c51d1546cad57b6188ca 2025-09-09T14:11:31.1996648Z Stored in directory: /root/.cache/pip/wheels/f6/48/c4/942f7a1d556fddd2348cb9ac262f251873dfd8a39afec5678e 2025-09-09T14:11:31.1999177Z  DEPRECATION: Building 'word2number' using the legacy setup.py bdist_wheel mechanism, which will be removed in a future version. pip 25.3 will enforce this behaviour change. A possible replacement is to use the standardized build interface by setting the `--use-pep517` option, (possibly combined with `--no-build-isolation`), or adding a `pyproject.toml` file to the source tree of 'word2number'. Discussion can be found at https://github.com/pypa/pip/issues/6334 2025-09-09T14:11:31.2001055Z  Building wheel for word2number (setup.py) ... [?25l- done 2025-09-09T14:11:31.2001976Z [?25h Created wheel for word2number: filename=word2number-1.1-py3-none-any.whl size=5658 sha256=544d5ab2ca96721db27d90cf5f6d621b3331889358821cc53dd588f4e656e534 2025-09-09T14:11:31.2002950Z Stored in directory: /root/.cache/pip/wheels/a0/4a/5b/d2f2df5c344ddbecb8bea759872c207ea91d93f57fb54e816e 2025-09-09T14:11:31.2003552Z Successfully built rouge-score sqlitedict word2number 2025-09-09T14:11:31.2008034Z Installing collected packages: word2number, sqlitedict, sortedcontainers, pytz, distlib, zstandard, zipp, xxhash, urllib3, tzdata, tqdm, tomli, threadpoolctl, termcolor, tcolorpy, tabulate, six, sentencepiece, safetensors, ruff, regex, pyyaml, pyparsing, pygments, pycryptodomex, pybind11, pyarrow, psutil, propcache, portalocker, pluggy, platformdirs, pillow, pathvalidate, parameterized, packaging, numpy, nodeenv, ninja, multidict, more_itertools, lxml, kiwisolver, joblib, iniconfig, idna, identify, hf-xet, fsspec, frozenlist, fonttools, expecttest, exceptiongroup, diskcache, dill, cycler, colorama, cmake, click, charset_normalizer, chardet, cfgv, certifi, attrs, async-timeout, aiohappyeyeballs, absl-py, yarl, virtualenv, unittest-xml-reporting, tqdm-multiprocess, scipy, sacrebleu, requests, python-dateutil, pytest, pycocotools, numexpr, nltk, multiprocess, mbstrdecoder, jsonlines, importlib-resources, importlib_metadata, hypothesis, fire, contourpy, blobfile, aiosignal, typepy, tiktoken, scikit-learn, rouge-score, pre-commit, pandas, matplotlib, huggingface-hub, aiohttp, tokenizers, bitsandbytes, accelerate, transformers, datasets, DataProperty, tabledata, peft, evaluate, pytablewriter, lm_eval 2025-09-09T14:11:31.2012698Z [?25l 2025-09-09T14:11:31.2013115Z  ━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/109 [distlib] 2025-09-09T14:11:31.2013676Z  ━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  7/109 [xxhash] 2025-09-09T14:11:31.2014397Z  ━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  9/109 [tzdata] 2025-09-09T14:11:31.2014978Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  18/109 [safetensors] 2025-09-09T14:11:31.2015722Z  ━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  20/109 [regex] 2025-09-09T14:11:31.2016375Z  ━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  23/109 [pygments] 2025-09-09T14:11:31.2016935Z  ━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  23/109 [pygments] 2025-09-09T14:11:31.2017500Z  ━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  23/109 [pygments] 2025-09-09T14:11:31.2018057Z  ━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  23/109 [pygments] 2025-09-09T14:11:31.2018640Z  ━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  24/109 [pycryptodomex] 2025-09-09T14:11:31.2019261Z  ━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  24/109 [pycryptodomex] 2025-09-09T14:11:31.2019832Z  ━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  26/109 [pyarrow] 2025-09-09T14:11:31.2020421Z  ━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  26/109 [pyarrow] 2025-09-09T14:11:31.2021024Z  ━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  26/109 [pyarrow] 2025-09-09T14:11:31.2021591Z  ━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  26/109 [pyarrow] 2025-09-09T14:11:31.2022155Z  ━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  26/109 [pyarrow] 2025-09-09T14:11:31.2022718Z  ━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  26/109 [pyarrow] 2025-09-09T14:11:31.2023280Z  ━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  26/109 [pyarrow] 2025-09-09T14:11:31.2023966Z  ━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━  31/109 [platformdirs] 2025-09-09T14:11:31.2024546Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  32/109 [pillow] 2025-09-09T14:11:38.0549144Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━  36/109 [numpy] 2025-09-09T14:11:38.0549754Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━  36/109 [numpy] 2025-09-09T14:11:38.0550308Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━  36/109 [numpy] 2025-09-09T14:11:38.0550849Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━  36/109 [numpy] 2025-09-09T14:11:38.0551403Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━  36/109 [numpy] 2025-09-09T14:11:38.0551943Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━  36/109 [numpy] 2025-09-09T14:11:38.0552531Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━  36/109 [numpy] 2025-09-09T14:11:38.0553084Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━  36/109 [numpy] 2025-09-09T14:11:38.0553654Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━  36/109 [numpy] 2025-09-09T14:11:38.0554207Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━  36/109 [numpy] 2025-09-09T14:11:38.0554753Z  ━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━  41/109 [lxml] 2025-09-09T14:11:38.0555307Z  ━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━  43/109 [joblib] 2025-09-09T14:11:38.0555867Z  ━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━  45/109 [idna] 2025-09-09T14:11:38.0556294Z  Attempting uninstall: fsspec 2025-09-09T14:11:38.0556788Z ━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━  45/109 [idna] 2025-09-09T14:11:38.0557251Z  Found existing installation: fsspec 2025.9.0 2025-09-09T14:11:38.0557756Z ━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━  45/109 [idna] 2025-09-09T14:11:38.0558194Z  Uninstalling fsspec-2025.9.0: 2025-09-09T14:11:38.0558655Z ━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━  45/109 [idna] 2025-09-09T14:11:38.0559104Z  Successfully uninstalled fsspec-2025.9.0 2025-09-09T14:11:38.0559586Z ━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━  45/109 [idna] 2025-09-09T14:11:38.0560133Z  ━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━  48/109 [fsspec] 2025-09-09T14:11:38.0560689Z  ━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━  50/109 [fonttools] 2025-09-09T14:11:38.0561501Z  ━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━  50/109 [fonttools] 2025-09-09T14:11:38.0562080Z  ━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━  50/109 [fonttools] 2025-09-09T14:11:38.0562803Z  ━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━  50/109 [fonttools] 2025-09-09T14:11:38.0563355Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━  57/109 [cmake] 2025-09-09T14:11:38.0563893Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━  57/109 [cmake] 2025-09-09T14:11:38.0564435Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━  57/109 [cmake] 2025-09-09T14:11:38.0564971Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━  57/109 [cmake] 2025-09-09T14:11:38.0565503Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━  57/109 [cmake] 2025-09-09T14:11:38.0566070Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━  57/109 [cmake] 2025-09-09T14:11:38.0566602Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━  57/109 [cmake] 2025-09-09T14:11:38.0567175Z  ━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━  60/109 [chardet] 2025-09-09T14:11:38.0567731Z  ━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━  66/109 [absl-py] 2025-09-09T14:11:38.0568287Z  ━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━  68/109 [virtualenv] 2025-09-09T14:11:38.0568836Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━  71/109 [scipy] 2025-09-09T14:11:38.0569375Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━  71/109 [scipy] 2025-09-09T14:11:38.0569916Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━  71/109 [scipy] 2025-09-09T14:11:38.0570572Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━  71/109 [scipy] 2025-09-09T14:11:38.0571130Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━  71/109 [scipy] 2025-09-09T14:11:38.0571767Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━  71/109 [scipy] 2025-09-09T14:11:38.0579613Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━  71/109 [scipy] 2025-09-09T14:11:38.0580258Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━  71/109 [scipy] 2025-09-09T14:11:38.0580815Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━  71/109 [scipy] 2025-09-09T14:11:38.0581358Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━  71/109 [scipy] 2025-09-09T14:11:38.0581907Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━  71/109 [scipy] 2025-09-09T14:11:38.0582470Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━  71/109 [scipy] 2025-09-09T14:11:45.5341577Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━  71/109 [scipy] 2025-09-09T14:11:45.5342235Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━  71/109 [scipy] 2025-09-09T14:11:45.5342784Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━  71/109 [scipy] 2025-09-09T14:11:45.5343326Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━  71/109 [scipy] 2025-09-09T14:11:45.5343874Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━  75/109 [pytest] 2025-09-09T14:11:45.5344423Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━  77/109 [numexpr] 2025-09-09T14:11:45.5344964Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━  78/109 [nltk] 2025-09-09T14:11:45.5345528Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━  78/109 [nltk] 2025-09-09T14:11:45.5346063Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━  78/109 [nltk] 2025-09-09T14:11:45.5346688Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━  79/109 [multiprocess] 2025-09-09T14:11:45.5347258Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━  84/109 [hypothesis] 2025-09-09T14:11:45.5347823Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━  84/109 [hypothesis] 2025-09-09T14:11:45.5348372Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━  88/109 [aiosignal] 2025-09-09T14:11:45.5348956Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━  91/109 [scikit-learn] 2025-09-09T14:11:45.5349525Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━  91/109 [scikit-learn] 2025-09-09T14:11:45.5350358Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━  91/109 [scikit-learn] 2025-09-09T14:11:45.5350943Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━  91/109 [scikit-learn] 2025-09-09T14:11:45.5351675Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━  91/109 [scikit-learn] 2025-09-09T14:11:45.5352242Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━  91/109 [scikit-learn] 2025-09-09T14:11:45.5352804Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━  91/109 [scikit-learn] 2025-09-09T14:11:45.5353373Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━  91/109 [scikit-learn] 2025-09-09T14:11:45.5353938Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━  91/109 [scikit-learn] 2025-09-09T14:11:45.5354514Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  93/109 [pre-commit] 2025-09-09T14:11:45.5355069Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  94/109 [pandas] 2025-09-09T14:11:45.5355622Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  94/109 [pandas] 2025-09-09T14:11:45.5356161Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  94/109 [pandas] 2025-09-09T14:11:45.5356701Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  94/109 [pandas] 2025-09-09T14:11:45.5357235Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  94/109 [pandas] 2025-09-09T14:11:45.5357767Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  94/109 [pandas] 2025-09-09T14:11:45.5358296Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  94/109 [pandas] 2025-09-09T14:11:45.5358936Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  94/109 [pandas] 2025-09-09T14:11:45.5359474Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  94/109 [pandas] 2025-09-09T14:11:45.5360376Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  94/109 [pandas] 2025-09-09T14:11:45.5360917Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  94/109 [pandas] 2025-09-09T14:11:45.5361448Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  94/109 [pandas] 2025-09-09T14:11:45.5361985Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  94/109 [pandas] 2025-09-09T14:11:45.5362519Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  94/109 [pandas] 2025-09-09T14:11:45.5363050Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  94/109 [pandas] 2025-09-09T14:11:45.5363603Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  94/109 [pandas] 2025-09-09T14:11:45.5364134Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  94/109 [pandas] 2025-09-09T14:11:45.5364692Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  94/109 [pandas] 2025-09-09T14:11:45.5365226Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  94/109 [pandas] 2025-09-09T14:11:45.5365758Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  94/109 [pandas] 2025-09-09T14:11:45.5366357Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━  95/109 [matplotlib] 2025-09-09T14:11:52.8799636Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━  95/109 [matplotlib] 2025-09-09T14:11:52.8800233Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━  95/109 [matplotlib] 2025-09-09T14:11:52.8800837Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━  95/109 [matplotlib] 2025-09-09T14:11:52.8801412Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━  95/109 [matplotlib] 2025-09-09T14:11:52.8801996Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━  95/109 [matplotlib] 2025-09-09T14:11:52.8802576Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━  96/109 [huggingface-hub] 2025-09-09T14:11:52.8803135Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━  97/109 [aiohttp] 2025-09-09T14:11:52.8803693Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━  99/109 [bitsandbytes] 2025-09-09T14:11:52.8804254Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━  99/109 [bitsandbytes] 2025-09-09T14:11:52.8805073Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━  99/109 [bitsandbytes] 2025-09-09T14:11:52.8805645Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━  99/109 [bitsandbytes] 2025-09-09T14:11:52.8806362Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━  99/109 [bitsandbytes] 2025-09-09T14:11:52.8806920Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━ 100/109 [accelerate] 2025-09-09T14:11:52.8807474Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8808025Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8808637Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8809190Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8809769Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8810328Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8810899Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8811460Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8812015Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8812574Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8813130Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8813839Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8814403Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8815056Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8815619Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8816237Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8816791Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8817350Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8817936Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8818531Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8819094Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8819666Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8820230Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8820785Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8821343Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8821902Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8822456Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8823031Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8823609Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:11:52.8824165Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:19:10.5085962Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 101/109 [transformers] 2025-09-09T14:19:10.5086570Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 102/109 [datasets] 2025-09-09T14:19:10.5087128Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━ 104/109 [tabledata] 2025-09-09T14:19:10.5087678Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 105/109 [peft] 2025-09-09T14:19:10.5088596Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺ 107/109 [pytablewriter] 2025-09-09T14:19:10.5089155Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 108/109 [lm_eval] 2025-09-09T14:19:10.5089874Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 108/109 [lm_eval] 2025-09-09T14:19:10.5090387Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 108/109 [lm_eval] 2025-09-09T14:19:10.5090898Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 108/109 [lm_eval] 2025-09-09T14:19:10.5091398Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 108/109 [lm_eval] 2025-09-09T14:19:10.5091906Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 108/109 [lm_eval] 2025-09-09T14:19:10.5092408Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 108/109 [lm_eval] 2025-09-09T14:19:10.5092916Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 108/109 [lm_eval] 2025-09-09T14:19:10.5093442Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 108/109 [lm_eval] 2025-09-09T14:19:10.5093953Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 108/109 [lm_eval] 2025-09-09T14:19:10.5094478Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 108/109 [lm_eval] 2025-09-09T14:19:10.5094975Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 108/109 [lm_eval] 2025-09-09T14:19:10.5095494Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 108/109 [lm_eval] 2025-09-09T14:19:10.5096073Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 108/109 [lm_eval] 2025-09-09T14:19:10.5096574Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 108/109 [lm_eval] 2025-09-09T14:19:10.5097082Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 108/109 [lm_eval] 2025-09-09T14:19:10.5098120Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 108/109 [lm_eval] 2025-09-09T14:19:10.5098619Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 109/109 [lm_eval] 2025-09-09T14:19:10.5099098Z [?25h 2025-09-09T14:19:10.5105999Z Successfully installed DataProperty-1.1.0 absl-py-2.3.1 accelerate-1.10.1 aiohappyeyeballs-2.6.1 aiohttp-3.12.15 aiosignal-1.4.0 async-timeout-5.0.1 attrs-25.3.0 bitsandbytes-0.47.0 blobfile-3.1.0 certifi-2025.8.3 cfgv-3.4.0 chardet-5.2.0 charset_normalizer-3.4.3 click-8.1.8 cmake-3.31.6 colorama-0.4.6 contourpy-1.3.0 cycler-0.12.1 datasets-3.6.0 dill-0.3.8 diskcache-5.6.3 distlib-0.4.0 evaluate-0.4.5 exceptiongroup-1.3.0 expecttest-0.3.0 fire-0.7.1 fonttools-4.59.2 frozenlist-1.7.0 fsspec-2025.3.0 hf-xet-1.1.9 huggingface-hub-0.34.4 hypothesis-6.138.15 identify-2.6.14 idna-3.10 importlib-resources-6.5.2 importlib_metadata-8.7.0 iniconfig-2.1.0 joblib-1.5.2 jsonlines-4.0.0 kiwisolver-1.4.7 lm_eval-0.4.9.1 lxml-6.0.1 matplotlib-3.9.4 mbstrdecoder-1.1.4 more_itertools-10.8.0 multidict-6.6.4 multiprocess-0.70.16 ninja-1.13.0 nltk-3.9.1 nodeenv-1.9.1 numexpr-2.10.2 numpy-2.0.2 packaging-25.0 pandas-2.3.2 parameterized-0.9.0 pathvalidate-3.3.1 peft-0.17.1 pillow-11.3.0 platformdirs-4.4.0 pluggy-1.6.0 portalocker-3.2.0 pre-commit-4.3.0 propcache-0.3.2 psutil-7.0.0 pyarrow-21.0.0 pybind11-3.0.1 pycocotools-2.0.10 pycryptodomex-3.23.0 pygments-2.19.2 pyparsing-3.2.3 pytablewriter-1.2.1 pytest-8.4.2 python-dateutil-2.9.0.post0 pytz-2025.2 pyyaml-6.0.2 regex-2025.9.1 requests-2.32.5 rouge-score-0.1.2 ruff-0.11.6 sacrebleu-2.5.1 safetensors-0.6.2 scikit-learn-1.6.1 scipy-1.13.1 sentencepiece-0.2.1 six-1.17.0 sortedcontainers-2.4.0 sqlitedict-2.1.0 tabledata-1.3.4 tabulate-0.9.0 tcolorpy-0.1.7 termcolor-3.1.0 threadpoolctl-3.6.0 tiktoken-0.11.0 tokenizers-0.22.0 tomli-2.2.1 tqdm-4.67.1 tqdm-multiprocess-0.0.11 transformers-4.56.1 typepy-1.3.4 tzdata-2025.2 unittest-xml-reporting-3.2.0 urllib3-2.5.0 virtualenv-20.34.0 word2number-1.1 xxhash-3.5.0 yarl-1.20.1 zipp-3.23.0 zstandard-0.24.0 2025-09-09T14:19:10.5113368Z WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager, possibly rendering your system unusable. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv. Use the --root-user-action option if you know what you are doing and want to suppress this warning. 2025-09-09T14:19:10.5114817Z + pip install . 2025-09-09T14:19:10.5115054Z Processing /pytorch/ao 2025-09-09T14:19:10.5115384Z Preparing metadata (setup.py) ... [?25l- done 2025-09-09T14:19:10.5115808Z [?25hBuilding wheels for collected packages: torchao 2025-09-09T14:19:10.5117839Z  DEPRECATION: Building 'torchao' using the legacy setup.py bdist_wheel mechanism, which will be removed in a future version. pip 25.3 will enforce this behaviour change. A possible replacement is to use the standardized build interface by setting the `--use-pep517` option, (possibly combined with `--no-build-isolation`), or adding a `pyproject.toml` file to the source tree of 'torchao'. Discussion can be found at https://github.com/pypa/pip/issues/6334 2025-09-09T14:19:10.5119820Z  Building wheel for torchao (setup.py) ... [?25l- \ | / - \ | / - \ | / - \ | / - \ | / - \ | done 2025-09-09T14:19:10.5120917Z [?25h Created wheel for torchao: filename=torchao-0.14.0+git7c05f81-cp39-abi3-linux_x86_64.whl size=7947574 sha256=e931edc6544aacbd90c8e05233cd016269d2b598de5a16658648dda4f4942e9c 2025-09-09T14:19:10.5122029Z Stored in directory: /tmp/pip-ephem-wheel-cache-735top0y/wheels/4d/54/dc/0c70e60a8677bf126f1486798ebe76c8770ada66c7376b401d 2025-09-09T14:19:10.5122655Z Successfully built torchao 2025-09-09T14:19:10.5122928Z Installing collected packages: torchao 2025-09-09T14:19:10.5123256Z Successfully installed torchao-0.14.0+git7c05f81 2025-09-09T14:19:10.5125042Z WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager, possibly rendering your system unusable. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv. Use the --root-user-action option if you know what you are doing and want to suppress this warning. 2025-09-09T14:19:10.5126544Z ++++ which conda 2025-09-09T14:19:10.5126793Z +++ dirname /opt/conda/condabin/conda 2025-09-09T14:19:10.5127096Z ++ dirname /opt/conda/condabin 2025-09-09T14:19:10.5127359Z + export CONDA=/opt/conda 2025-09-09T14:19:10.5127597Z + CONDA=/opt/conda 2025-09-09T14:19:10.5128076Z + export LD_LIBRARY_PATH=/opt/conda/lib/:/opt/rh/gcc-toolset-11/root/usr/lib64:/opt/rh/gcc-toolset-11/root/usr/lib: 2025-09-09T14:19:10.5128898Z + LD_LIBRARY_PATH=/opt/conda/lib/:/opt/rh/gcc-toolset-11/root/usr/lib64:/opt/rh/gcc-toolset-11/root/usr/lib: 2025-09-09T14:19:10.5129410Z + pytest test --verbose -s 2025-09-09T14:19:25.4727821Z ============================= test session starts ============================== 2025-09-09T14:19:25.4728548Z platform linux -- Python 3.9.23, pytest-8.4.2, pluggy-1.6.0 -- /opt/conda/envs/venv/bin/python3.9 2025-09-09T14:19:25.4729214Z cachedir: .pytest_cache 2025-09-09T14:19:25.4729915Z hypothesis profile 'ci' -> database=None, deadline=None, print_blob=True, derandomize=True, suppress_health_check=(HealthCheck.too_slow,) 2025-09-09T14:19:25.4730524Z rootdir: /pytorch/ao 2025-09-09T14:19:25.4730779Z plugins: hypothesis-6.138.15 2025-09-09T14:19:25.4731094Z collecting ...  2025-09-09T14:19:25.4731513Z collecting 0 items  2025-09-09T14:19:25.4732024Z collecting 26 items  2025-09-09T14:19:25.4732530Z collecting 26 items  2025-09-09T14:19:25.4733045Z collecting 273 items  2025-09-09T14:19:25.4733664Z collecting 1031 items / 3 skipped  2025-09-09T14:19:25.4734938Z collecting 1072 items / 5 skipped NOTE: Using slow Hadamard transform for SpinQuant. For better performance on GPU, install `fast_hadamard_transform`: `pip install git+https://github.com/Dao-AILab/fast-hadamard-transform.git` 2025-09-09T14:19:25.4735989Z  2025-09-09T14:19:25.4736376Z collecting 1484 items / 18 skipped  2025-09-09T14:19:25.4736951Z collecting 2154 items / 18 skipped  2025-09-09T14:19:25.4737530Z collecting 4142 items / 18 skipped  2025-09-09T14:19:25.4738094Z collected 5480 items / 18 skipped  2025-09-09T14:19:25.4738397Z 2025-09-09T14:19:25.4738793Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config0] PASSED 2025-09-09T14:19:25.4739526Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config1] PASSED 2025-09-09T14:19:25.4740266Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config2] PASSED 2025-09-09T14:19:25.4740991Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config3] PASSED 2025-09-09T14:19:25.4741708Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config4] PASSED 2025-09-09T14:19:25.4742444Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config5] PASSED 2025-09-09T14:19:25.4743163Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config6] PASSED 2025-09-09T14:19:25.4743888Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config7] PASSED 2025-09-09T14:19:25.4744616Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config8] PASSED 2025-09-09T14:19:25.4745628Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config9] PASSED 2025-09-09T14:19:25.4746371Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config10] PASSED 2025-09-09T14:19:25.4747262Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config11] PASSED 2025-09-09T14:19:25.4748000Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config12] PASSED 2025-09-09T14:19:25.4748748Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config13] PASSED 2025-09-09T14:19:25.4749484Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config14] PASSED 2025-09-09T14:19:25.4750223Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config15] PASSED 2025-09-09T14:19:25.4750956Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config16] PASSED 2025-09-09T14:19:25.4751703Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config17] PASSED 2025-09-09T14:19:25.4752436Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config18] PASSED 2025-09-09T14:19:25.4753221Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config19] PASSED 2025-09-09T14:19:25.4753957Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config20] PASSED 2025-09-09T14:19:25.4754686Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config21] PASSED 2025-09-09T14:19:25.4755425Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config22] PASSED 2025-09-09T14:19:25.4756053Z test/core/test_config.py::test_disallowed_modules PASSED 2025-09-09T14:19:25.4756594Z test/core/test_config.py::test_version_mismatch PASSED 2025-09-09T14:19:25.4757116Z test/core/test_config.py::test_default_version PASSED 2025-09-09T14:19:25.4757867Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_copy__mismatch_metadata_apply_quant0 PASSED 2025-09-09T14:19:25.4758820Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_copy__mismatch_metadata_apply_quant1 PASSED 2025-09-09T14:19:25.4759761Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_copy__mismatch_metadata_apply_quant2 PASSED 2025-09-09T14:19:25.4762819Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_copy__mismatch_metadata_apply_quant3 PASSED 2025-09-09T14:19:25.4763813Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_copy__mismatch_metadata_apply_quant4 PASSED 2025-09-09T14:19:25.4764745Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_copy__mismatch_metadata_apply_quant5 PASSED 2025-09-09T14:19:25.4765676Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_copy__mismatch_metadata_apply_quant6 PASSED 2025-09-09T14:19:25.4766555Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_print_quantized_module PASSED 2025-09-09T14:19:25.4767390Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_register_new_dispatch PASSED 2025-09-09T14:19:25.4768254Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_tensor_core_layout_transpose PASSED 2025-09-09T14:19:25.4769140Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_test_copy__apply_apply_quant0 PASSED 2025-09-09T14:19:25.4770022Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_test_copy__apply_apply_quant1 PASSED 2025-09-09T14:19:25.4770896Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_test_copy__apply_apply_quant2 PASSED 2025-09-09T14:19:25.4771777Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_test_copy__apply_apply_quant3 PASSED 2025-09-09T14:19:25.4772669Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_test_copy__apply_apply_quant4 PASSED 2025-09-09T14:19:25.4773686Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_test_copy__apply_apply_quant5 PASSED 2025-09-09T14:19:25.4774574Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_test_copy__apply_apply_quant6 PASSED 2025-09-09T14:19:25.4775564Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_to_affine_quantized_intx_static PASSED 2025-09-09T14:19:25.4776507Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_to_device_apply_quant0 PASSED 2025-09-09T14:19:25.4777340Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_to_device_apply_quant1 PASSED 2025-09-09T14:19:25.4778159Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_to_device_apply_quant2 PASSED 2025-09-09T14:19:25.4778984Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_to_device_apply_quant3 PASSED 2025-09-09T14:19:25.4779777Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_weights_only PASSED 2025-09-09T14:19:25.4780613Z test/dtypes/test_affine_quantized.py::TestAffineQuantizedBasic::test_alias_device_cpu_bfloat16 PASSED 2025-09-09T14:19:25.4781530Z test/dtypes/test_affine_quantized.py::TestAffineQuantizedBasic::test_alias_device_cuda_bfloat16 PASSED 2025-09-09T14:19:25.4782491Z test/dtypes/test_affine_quantized.py::TestAffineQuantizedBasic::test_flatten_unflatten_device_cpu_bfloat16 PASSED 2025-09-09T14:19:25.4783550Z test/dtypes/test_affine_quantized.py::TestAffineQuantizedBasic::test_flatten_unflatten_device_cuda_bfloat16 PASSED 2025-09-09T14:19:25.4784499Z test/dtypes/test_affine_quantized.py::TestAffineQuantizedBasic::test_matmul_device_cuda_bfloat16 PASSED 2025-09-09T14:19:25.4785429Z test/dtypes/test_affine_quantized.py::TestAffineQuantizedBasic::test_mm_int4wo_device_cuda_bfloat16 PASSED 2025-09-09T14:19:25.4786419Z test/dtypes/test_affine_quantized.py::TestAffineQuantizedBasic::test_slice_and_copy_int4wo_device_cuda_bfloat16 PASSED 2025-09-09T14:19:25.4787410Z test/dtypes/test_affine_quantized.py::TestAffineQuantizedBasic::test_slice_gemlite_device_cuda_bfloat16 SKIPPED 2025-09-09T14:19:25.4788390Z test/dtypes/test_affine_quantized.py::TestAffineQuantizedBasic::test_slice_gemlite_device_cuda_float16 SKIPPED 2025-09-09T14:19:25.4789352Z test/dtypes/test_affine_quantized.py::TestAffineQuantizedBasic::test_slice_int4wo_device_cuda_bfloat16 PASSED 2025-09-09T14:19:25.4790447Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_choose_scale_float8_bounds_float8_e4m3fn_bfloat16 SKIPPED 2025-09-09T14:19:25.4791660Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_choose_scale_float8_bounds_float8_e4m3fn_float32 SKIPPED 2025-09-09T14:19:25.4792862Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_choose_scale_float8_bounds_float8_e5m2_bfloat16 SKIPPED 2025-09-09T14:19:25.4794101Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_choose_scale_float8_bounds_float8_e5m2_float32 SKIPPED 2025-09-09T14:19:25.5072436Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e4m3fn_bfloat16_block_size0 SKIPPED 2025-09-09T14:19:25.5073985Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e4m3fn_bfloat16_block_size1 SKIPPED 2025-09-09T14:19:25.5075269Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e4m3fn_bfloat16_block_size2 SKIPPED 2025-09-09T14:19:25.5076567Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e4m3fn_bfloat16_block_size3 SKIPPED 2025-09-09T14:19:25.5078010Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e4m3fn_float32_block_size0 SKIPPED 2025-09-09T14:19:25.5079300Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e4m3fn_float32_block_size1 SKIPPED 2025-09-09T14:19:25.5080701Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e4m3fn_float32_block_size2 SKIPPED 2025-09-09T14:19:25.5081978Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e4m3fn_float32_block_size3 SKIPPED 2025-09-09T14:19:25.5083318Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e5m2_bfloat16_block_size0 SKIPPED 2025-09-09T14:19:25.5084592Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e5m2_bfloat16_block_size1 SKIPPED 2025-09-09T14:19:25.5085879Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e5m2_bfloat16_block_size2 SKIPPED 2025-09-09T14:19:25.5087175Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e5m2_bfloat16_block_size3 SKIPPED 2025-09-09T14:19:25.5088451Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e5m2_float32_block_size0 SKIPPED 2025-09-09T14:19:25.5089718Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e5m2_float32_block_size1 SKIPPED 2025-09-09T14:19:25.5090989Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e5m2_float32_block_size2 SKIPPED 2025-09-09T14:19:25.5092261Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e5m2_float32_block_size3 SKIPPED 2025-09-09T14:19:25.5093537Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_scale_broadcasting SKIPPED 2025-09-09T14:19:25.5094795Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_expected_kernels_on_gpu_granularity0_float8_config_version_1 SKIPPED 2025-09-09T14:19:25.5096151Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_expected_kernels_on_gpu_granularity0_float8_config_version_2 SKIPPED 2025-09-09T14:19:25.5097684Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_expected_kernels_on_gpu_granularity1_float8_config_version_1 SKIPPED 2025-09-09T14:19:25.5099005Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_expected_kernels_on_gpu_granularity1_float8_config_version_2 SKIPPED 2025-09-09T14:19:25.5100222Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_float8_tensor_slicing_basic_granularity0 SKIPPED 2025-09-09T14:19:25.5101374Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_float8_tensor_slicing_basic_granularity1 SKIPPED 2025-09-09T14:19:25.5110669Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_float8_tensor_slicing_edge_cases SKIPPED 2025-09-09T14:19:25.5112102Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_float8_tensor_slicing_functional_correctness_granularity0 SKIPPED 2025-09-09T14:19:25.5113751Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_float8_tensor_slicing_functional_correctness_granularity1 SKIPPED 2025-09-09T14:19:25.5115105Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_float8_tensor_slicing_per_row SKIPPED 2025-09-09T14:19:25.5116200Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_float8_tensor_slicing_per_tensor SKIPPED 2025-09-09T14:19:25.5117623Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity0_sizes0 SKIPPED 2025-09-09T14:19:25.5119041Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity0_sizes1 SKIPPED 2025-09-09T14:19:25.5120458Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity1_sizes0 SKIPPED 2025-09-09T14:19:25.5121874Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity1_sizes1 SKIPPED 2025-09-09T14:19:25.5123318Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity0_sizes0 SKIPPED 2025-09-09T14:19:25.5124762Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity0_sizes1 SKIPPED 2025-09-09T14:19:25.5126171Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity1_sizes0 SKIPPED 2025-09-09T14:19:25.5127579Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity1_sizes1 SKIPPED 2025-09-09T14:19:25.5128985Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_static_compile_False_granularity0_sizes0 SKIPPED 2025-09-09T14:19:25.5130408Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_static_compile_False_granularity0_sizes1 SKIPPED 2025-09-09T14:19:25.5131813Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_static_compile_False_granularity1_sizes0 SKIPPED 2025-09-09T14:19:25.5133226Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_static_compile_False_granularity1_sizes1 SKIPPED 2025-09-09T14:19:25.5134643Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_static_compile_True_granularity0_sizes0 SKIPPED 2025-09-09T14:19:25.5136129Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_static_compile_True_granularity0_sizes1 SKIPPED 2025-09-09T14:19:25.5137551Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_static_compile_True_granularity1_sizes0 SKIPPED 2025-09-09T14:19:25.5138962Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_static_compile_True_granularity1_sizes1 SKIPPED 2025-09-09T14:19:25.5140405Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity0_sizes0 SKIPPED 2025-09-09T14:19:25.5141884Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity0_sizes1 SKIPPED 2025-09-09T14:19:25.5143394Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity1_sizes0 SKIPPED 2025-09-09T14:19:25.5144953Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity1_sizes1 SKIPPED 2025-09-09T14:19:25.5146496Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity0_sizes0 SKIPPED 2025-09-09T14:19:25.5392370Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity0_sizes1 SKIPPED 2025-09-09T14:19:25.5394509Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity1_sizes0 SKIPPED 2025-09-09T14:19:25.5397656Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity1_sizes1 SKIPPED 2025-09-09T14:19:25.5401203Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity0_sizes0 SKIPPED 2025-09-09T14:19:25.5403589Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity0_sizes1 SKIPPED 2025-09-09T14:19:25.5405343Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity1_sizes0 SKIPPED 2025-09-09T14:19:25.5407098Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity1_sizes1 SKIPPED 2025-09-09T14:19:25.5408834Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity0_sizes0 SKIPPED 2025-09-09T14:19:25.5410586Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity0_sizes1 SKIPPED 2025-09-09T14:19:25.5412329Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity1_sizes0 SKIPPED 2025-09-09T14:19:25.5414118Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity1_sizes1 SKIPPED 2025-09-09T14:19:25.5415928Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_static_compile_False_granularity0_sizes0 SKIPPED 2025-09-09T14:19:25.5417323Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_static_compile_False_granularity0_sizes1 SKIPPED 2025-09-09T14:19:25.5418734Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_static_compile_False_granularity1_sizes0 SKIPPED 2025-09-09T14:19:25.5420144Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_static_compile_False_granularity1_sizes1 SKIPPED 2025-09-09T14:19:25.5421540Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_static_compile_True_granularity0_sizes0 SKIPPED 2025-09-09T14:19:25.5422942Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_static_compile_True_granularity0_sizes1 SKIPPED 2025-09-09T14:19:25.5424388Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_static_compile_True_granularity1_sizes0 SKIPPED 2025-09-09T14:19:25.5425940Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_static_compile_True_granularity1_sizes1 SKIPPED 2025-09-09T14:19:25.5427489Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity0_sizes0 SKIPPED 2025-09-09T14:19:25.5428942Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity0_sizes1 SKIPPED 2025-09-09T14:19:25.5430384Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity1_sizes0 SKIPPED 2025-09-09T14:19:25.5431831Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity1_sizes1 SKIPPED 2025-09-09T14:19:25.5433289Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity0_sizes0 SKIPPED 2025-09-09T14:19:25.5434733Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity0_sizes1 SKIPPED 2025-09-09T14:19:25.5436177Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity1_sizes0 SKIPPED 2025-09-09T14:19:25.5437617Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity1_sizes1 SKIPPED 2025-09-09T14:19:25.5438852Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_weight_dimension_warning SKIPPED 2025-09-09T14:19:25.5439904Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_invalid_granularity SKIPPED 2025-09-09T14:19:25.5440922Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mismatched_granularity SKIPPED 2025-09-09T14:19:25.5442179Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_1024_out_features_512_leading_shape0_bias_False SKIPPED 2025-09-09T14:19:25.5443653Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_1024_out_features_512_leading_shape0_bias_True SKIPPED 2025-09-09T14:19:25.5445068Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_1024_out_features_512_leading_shape1_bias_False SKIPPED 2025-09-09T14:19:25.5446490Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_1024_out_features_512_leading_shape1_bias_True SKIPPED 2025-09-09T14:19:25.5447911Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_1024_out_features_512_leading_shape2_bias_False SKIPPED 2025-09-09T14:19:25.5449326Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_1024_out_features_512_leading_shape2_bias_True SKIPPED 2025-09-09T14:19:25.5450749Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_1024_out_features_512_leading_shape3_bias_False SKIPPED 2025-09-09T14:19:25.5452170Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_1024_out_features_512_leading_shape3_bias_True SKIPPED 2025-09-09T14:19:25.5453720Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_1024_out_features_512_leading_shape4_bias_False SKIPPED 2025-09-09T14:19:25.5455141Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_1024_out_features_512_leading_shape4_bias_True SKIPPED 2025-09-09T14:19:25.5456691Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_256_out_features_768_leading_shape0_bias_False SKIPPED 2025-09-09T14:19:25.5458096Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_256_out_features_768_leading_shape0_bias_True SKIPPED 2025-09-09T14:19:25.5459695Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_256_out_features_768_leading_shape1_bias_False SKIPPED 2025-09-09T14:19:25.5461114Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_256_out_features_768_leading_shape1_bias_True SKIPPED 2025-09-09T14:19:25.5462522Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_256_out_features_768_leading_shape2_bias_False SKIPPED 2025-09-09T14:19:25.5463934Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_256_out_features_768_leading_shape2_bias_True SKIPPED 2025-09-09T14:19:25.5465331Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_256_out_features_768_leading_shape3_bias_False SKIPPED 2025-09-09T14:20:19.8742352Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_256_out_features_768_leading_shape3_bias_True SKIPPED 2025-09-09T14:20:19.8743840Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_256_out_features_768_leading_shape4_bias_False SKIPPED 2025-09-09T14:20:19.8745268Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_256_out_features_768_leading_shape4_bias_True SKIPPED 2025-09-09T14:20:19.8746692Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_512_out_features_1024_leading_shape0_bias_False SKIPPED 2025-09-09T14:20:19.8748102Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_512_out_features_1024_leading_shape0_bias_True SKIPPED 2025-09-09T14:20:19.8749515Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_512_out_features_1024_leading_shape1_bias_False SKIPPED 2025-09-09T14:20:19.8750937Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_512_out_features_1024_leading_shape1_bias_True SKIPPED 2025-09-09T14:20:19.8752362Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_512_out_features_1024_leading_shape2_bias_False SKIPPED 2025-09-09T14:20:19.8753779Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_512_out_features_1024_leading_shape2_bias_True SKIPPED 2025-09-09T14:20:19.8755194Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_512_out_features_1024_leading_shape3_bias_False SKIPPED 2025-09-09T14:20:19.8756595Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_512_out_features_1024_leading_shape3_bias_True SKIPPED 2025-09-09T14:20:19.8758413Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_512_out_features_1024_leading_shape4_bias_False SKIPPED 2025-09-09T14:20:19.8759834Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_512_out_features_1024_leading_shape4_bias_True SKIPPED 2025-09-09T14:20:19.8761196Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_per_row_with_float32 SKIPPED 2025-09-09T14:20:19.8762231Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_preprocess_scale_3d_reshape PASSED 2025-09-09T14:20:19.8763285Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_serialization_mode_dynamic SKIPPED 2025-09-09T14:20:19.8764332Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_serialization_mode_static SKIPPED 2025-09-09T14:20:19.8765404Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_serialization_mode_weight-only SKIPPED 2025-09-09T14:20:19.8766464Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_unsupported_granularity SKIPPED 2025-09-09T14:20:19.8767543Z test/dtypes/test_affine_quantized_tensor_parallel.py::TestInt8woAffineQuantizedTensorParallel::test_tp_bfloat16 PASSED 2025-09-09T14:20:19.8768618Z test/dtypes/test_affine_quantized_tensor_parallel.py::TestInt8woAffineQuantizedTensorParallel::test_tp_float16 PASSED 2025-09-09T14:20:19.8769684Z test/dtypes/test_affine_quantized_tensor_parallel.py::TestInt8woAffineQuantizedTensorParallel::test_tp_float32 PASSED 2025-09-09T14:20:19.8770813Z test/dtypes/test_affine_quantized_tensor_parallel.py::TestInt4woAffineQuantizedTensorParallel::test_tp_bfloat16 SKIPPED 2025-09-09T14:20:19.8771887Z test/dtypes/test_affine_quantized_tensor_parallel.py::TestGemliteLayoutTensorParallel::test_tp_gemlite_float16 SKIPPED 2025-09-09T14:20:19.8772956Z test/dtypes/test_affine_quantized_tensor_parallel.py::TestInt8dqAffineQuantizedTensorParallel::test_tp_bfloat16 PASSED 2025-09-09T14:20:19.8773750Z test/dtypes/test_bitpacking.py::test_CPU[0-1] PASSED 2025-09-09T14:20:19.8774256Z test/dtypes/test_bitpacking.py::test_CPU[0-2] PASSED 2025-09-09T14:20:19.8774765Z test/dtypes/test_bitpacking.py::test_CPU[0-3] PASSED 2025-09-09T14:20:19.8775267Z test/dtypes/test_bitpacking.py::test_CPU[0-4] PASSED 2025-09-09T14:20:19.8775774Z test/dtypes/test_bitpacking.py::test_CPU[0-5] PASSED 2025-09-09T14:20:19.8776346Z test/dtypes/test_bitpacking.py::test_CPU[0-6] PASSED 2025-09-09T14:20:19.8776842Z test/dtypes/test_bitpacking.py::test_CPU[0-7] PASSED 2025-09-09T14:20:19.8777348Z test/dtypes/test_bitpacking.py::test_CPU[-1-1] PASSED 2025-09-09T14:20:19.8777854Z test/dtypes/test_bitpacking.py::test_CPU[-1-2] PASSED 2025-09-09T14:20:19.8778371Z test/dtypes/test_bitpacking.py::test_CPU[-1-3] PASSED 2025-09-09T14:20:19.8778874Z test/dtypes/test_bitpacking.py::test_CPU[-1-4] PASSED 2025-09-09T14:20:19.8779390Z test/dtypes/test_bitpacking.py::test_CPU[-1-5] PASSED 2025-09-09T14:20:19.8779899Z test/dtypes/test_bitpacking.py::test_CPU[-1-6] PASSED 2025-09-09T14:20:19.8780453Z test/dtypes/test_bitpacking.py::test_CPU[-1-7] PASSED 2025-09-09T14:20:19.8780960Z test/dtypes/test_bitpacking.py::test_CPU[1-1] PASSED 2025-09-09T14:20:19.8781463Z test/dtypes/test_bitpacking.py::test_CPU[1-2] PASSED 2025-09-09T14:20:19.8781967Z test/dtypes/test_bitpacking.py::test_CPU[1-3] PASSED 2025-09-09T14:20:19.8782464Z test/dtypes/test_bitpacking.py::test_CPU[1-4] PASSED 2025-09-09T14:20:19.8782966Z test/dtypes/test_bitpacking.py::test_CPU[1-5] PASSED 2025-09-09T14:20:19.8783470Z test/dtypes/test_bitpacking.py::test_CPU[1-6] PASSED 2025-09-09T14:20:19.8784061Z test/dtypes/test_bitpacking.py::test_CPU[1-7] PASSED 2025-09-09T14:20:19.8784567Z test/dtypes/test_bitpacking.py::test_GPU[0-1] PASSED 2025-09-09T14:20:19.8785064Z test/dtypes/test_bitpacking.py::test_GPU[0-2] PASSED 2025-09-09T14:20:19.8785652Z test/dtypes/test_bitpacking.py::test_GPU[0-3] PASSED 2025-09-09T14:20:19.8786151Z test/dtypes/test_bitpacking.py::test_GPU[0-4] PASSED 2025-09-09T14:20:19.8786653Z test/dtypes/test_bitpacking.py::test_GPU[0-5] PASSED 2025-09-09T14:20:19.8787154Z test/dtypes/test_bitpacking.py::test_GPU[0-6] PASSED 2025-09-09T14:20:19.8787648Z test/dtypes/test_bitpacking.py::test_GPU[0-7] PASSED 2025-09-09T14:20:19.8788157Z test/dtypes/test_bitpacking.py::test_GPU[-1-1] PASSED 2025-09-09T14:20:19.8788667Z test/dtypes/test_bitpacking.py::test_GPU[-1-2] PASSED 2025-09-09T14:20:19.8789177Z test/dtypes/test_bitpacking.py::test_GPU[-1-3] PASSED 2025-09-09T14:20:19.8789689Z test/dtypes/test_bitpacking.py::test_GPU[-1-4] PASSED 2025-09-09T14:20:19.8790249Z test/dtypes/test_bitpacking.py::test_GPU[-1-5] PASSED 2025-09-09T14:20:19.8790762Z test/dtypes/test_bitpacking.py::test_GPU[-1-6] PASSED 2025-09-09T14:20:19.8791271Z test/dtypes/test_bitpacking.py::test_GPU[-1-7] PASSED 2025-09-09T14:20:19.8791776Z test/dtypes/test_bitpacking.py::test_GPU[1-1] PASSED 2025-09-09T14:20:19.8792274Z test/dtypes/test_bitpacking.py::test_GPU[1-2] PASSED 2025-09-09T14:20:19.8792781Z test/dtypes/test_bitpacking.py::test_GPU[1-3] PASSED 2025-09-09T14:20:19.8793275Z test/dtypes/test_bitpacking.py::test_GPU[1-4] PASSED 2025-09-09T14:20:19.8793776Z test/dtypes/test_bitpacking.py::test_GPU[1-5] PASSED 2025-09-09T14:20:19.8794276Z test/dtypes/test_bitpacking.py::test_GPU[1-6] PASSED 2025-09-09T14:20:19.8794779Z test/dtypes/test_bitpacking.py::test_GPU[1-7] PASSED 2025-09-09T14:20:19.8795306Z test/dtypes/test_bitpacking.py::test_compile[0-1] PASSED 2025-09-09T14:20:19.8795849Z test/dtypes/test_bitpacking.py::test_compile[0-2] PASSED 2025-09-09T14:20:19.8796390Z test/dtypes/test_bitpacking.py::test_compile[0-3] PASSED 2025-09-09T14:20:19.8796928Z test/dtypes/test_bitpacking.py::test_compile[0-4] PASSED 2025-09-09T14:20:19.8797837Z test/dtypes/test_bitpacking.py::test_compile[0-5] PASSED 2025-09-09T14:20:19.8798375Z test/dtypes/test_bitpacking.py::test_compile[0-6] PASSED 2025-09-09T14:20:19.8798911Z test/dtypes/test_bitpacking.py::test_compile[0-7] PASSED 2025-09-09T14:20:19.8799452Z test/dtypes/test_bitpacking.py::test_compile[-1-1] PASSED 2025-09-09T14:20:19.8799995Z test/dtypes/test_bitpacking.py::test_compile[-1-2] PASSED 2025-09-09T14:20:19.8800542Z test/dtypes/test_bitpacking.py::test_compile[-1-3] PASSED 2025-09-09T14:20:19.8801079Z test/dtypes/test_bitpacking.py::test_compile[-1-4] PASSED 2025-09-09T14:20:19.8801628Z test/dtypes/test_bitpacking.py::test_compile[-1-5] PASSED 2025-09-09T14:20:19.8802166Z test/dtypes/test_bitpacking.py::test_compile[-1-6] PASSED 2025-09-09T14:20:19.8802712Z test/dtypes/test_bitpacking.py::test_compile[-1-7] PASSED 2025-09-09T14:20:19.8803259Z test/dtypes/test_bitpacking.py::test_compile[1-1] PASSED 2025-09-09T14:20:19.8803792Z test/dtypes/test_bitpacking.py::test_compile[1-2] PASSED 2025-09-09T14:20:19.8804335Z test/dtypes/test_bitpacking.py::test_compile[1-3] PASSED 2025-09-09T14:20:19.8804867Z test/dtypes/test_bitpacking.py::test_compile[1-4] PASSED 2025-09-09T14:20:19.8805410Z test/dtypes/test_bitpacking.py::test_compile[1-5] PASSED 2025-09-09T14:20:19.8805941Z test/dtypes/test_bitpacking.py::test_compile[1-6] PASSED 2025-09-09T14:20:19.8806478Z test/dtypes/test_bitpacking.py::test_compile[1-7] PASSED 2025-09-09T14:20:19.8807387Z test/dtypes/test_bitpacking.py::test_pack_example tensor([ 0, 105, 151, 37], device='cuda:0', dtype=torch.uint8) tensor([ 39, 146], device='cuda:0', dtype=torch.uint8) 2025-09-09T14:20:19.8808079Z PASSED 2025-09-09T14:20:19.8808620Z test/dtypes/test_bitpacking.py::test_pack_example_CPU tensor([ 0, 105, 151, 37], dtype=torch.uint8) tensor([ 39, 146], dtype=torch.uint8) 2025-09-09T14:22:10.2481838Z PASSED 2025-09-09T14:22:10.2482632Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_fpx_weight_only_ebits_2_mbits_2_bias_False_bfloat16 PASSED 2025-09-09T14:22:10.2483737Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_fpx_weight_only_ebits_2_mbits_2_bias_False_float16 PASSED 2025-09-09T14:22:10.2484844Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_fpx_weight_only_ebits_2_mbits_2_bias_True_bfloat16 PASSED 2025-09-09T14:22:10.2485938Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_fpx_weight_only_ebits_2_mbits_2_bias_True_float16 PASSED 2025-09-09T14:22:10.2487052Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_fpx_weight_only_ebits_3_mbits_2_bias_False_bfloat16 PASSED 2025-09-09T14:22:10.2488455Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_fpx_weight_only_ebits_3_mbits_2_bias_False_float16 PASSED 2025-09-09T14:22:10.2489743Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_fpx_weight_only_ebits_3_mbits_2_bias_True_bfloat16 PASSED 2025-09-09T14:22:10.2490830Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_fpx_weight_only_ebits_3_mbits_2_bias_True_float16 PASSED 2025-09-09T14:22:10.2491952Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_from_scaled_tc_floatx_compile_ebits_2_mbits_2_device_cpu PASSED 2025-09-09T14:22:10.2493108Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_from_scaled_tc_floatx_compile_ebits_2_mbits_2_device_cuda PASSED 2025-09-09T14:22:10.2494261Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_from_scaled_tc_floatx_compile_ebits_3_mbits_2_device_cpu PASSED 2025-09-09T14:22:10.2495423Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_from_scaled_tc_floatx_compile_ebits_3_mbits_2_device_cuda PASSED 2025-09-09T14:22:10.2496708Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_from_tc_floatx_correctness_ebits_2_mbits_2_device_cpu PASSED 2025-09-09T14:22:10.2498132Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_from_tc_floatx_correctness_ebits_2_mbits_2_device_cuda PASSED 2025-09-09T14:22:10.2499258Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_from_tc_floatx_correctness_ebits_3_mbits_2_device_cpu PASSED 2025-09-09T14:22:10.2500378Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_from_tc_floatx_correctness_ebits_3_mbits_2_device_cuda PASSED 2025-09-09T14:22:10.2501439Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_pack_tc_fp6_correctness_device_cpu PASSED 2025-09-09T14:22:10.2502425Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_pack_tc_fp6_correctness_device_cuda PASSED 2025-09-09T14:22:10.2503393Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_to_copy_device_ebits_2_mbits_2 PASSED 2025-09-09T14:22:10.2504337Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_to_copy_device_ebits_3_mbits_2 PASSED 2025-09-09T14:22:10.2505378Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_to_scaled_tc_floatx_compile_ebits_2_mbits_2_device_cpu PASSED 2025-09-09T14:22:10.2506561Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_to_scaled_tc_floatx_compile_ebits_2_mbits_2_device_cuda PASSED 2025-09-09T14:22:10.2507685Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_to_scaled_tc_floatx_compile_ebits_3_mbits_2_device_cpu PASSED 2025-09-09T14:22:10.2510077Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_to_scaled_tc_floatx_compile_ebits_3_mbits_2_device_cuda PASSED 2025-09-09T14:22:10.2511159Z test/dtypes/test_nf4.py::TestNF4Linear::test_backward_dtype_match_bfloat16 PASSED 2025-09-09T14:22:10.2511856Z test/dtypes/test_nf4.py::TestNF4Linear::test_backward_dtype_match_float16 PASSED 2025-09-09T14:22:10.2512551Z test/dtypes/test_nf4.py::TestNF4Linear::test_backward_dtype_match_float32 PASSED 2025-09-09T14:22:10.2513345Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_bfloat16_shape0_chunk_size_16 PASSED 2025-09-09T14:22:10.2514213Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_bfloat16_shape0_chunk_size_32 PASSED 2025-09-09T14:22:10.2515088Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_bfloat16_shape0_chunk_size_8 PASSED 2025-09-09T14:22:10.2515958Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_bfloat16_shape1_chunk_size_16 PASSED 2025-09-09T14:22:10.2516826Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_bfloat16_shape1_chunk_size_32 PASSED 2025-09-09T14:22:10.2517700Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_bfloat16_shape1_chunk_size_8 PASSED 2025-09-09T14:22:10.2518561Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float16_shape0_chunk_size_16 PASSED 2025-09-09T14:22:10.2519423Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float16_shape0_chunk_size_32 PASSED 2025-09-09T14:22:10.2520276Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float16_shape0_chunk_size_8 PASSED 2025-09-09T14:22:10.2521140Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float16_shape1_chunk_size_16 PASSED 2025-09-09T14:22:10.2522007Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float16_shape1_chunk_size_32 PASSED 2025-09-09T14:22:10.2522861Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float16_shape1_chunk_size_8 PASSED 2025-09-09T14:22:10.2523732Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float32_shape0_chunk_size_16 PASSED 2025-09-09T14:22:10.2524599Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float32_shape0_chunk_size_32 PASSED 2025-09-09T14:22:10.2525595Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float32_shape0_chunk_size_8 PASSED 2025-09-09T14:22:10.2526680Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float32_shape1_chunk_size_16 PASSED 2025-09-09T14:22:10.2527741Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float32_shape1_chunk_size_32 PASSED 2025-09-09T14:22:10.2528602Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float32_shape1_chunk_size_8 PASSED 2025-09-09T14:22:10.2529359Z test/dtypes/test_nf4.py::TestNF4Linear::test_empty_like_input_size0 PASSED 2025-09-09T14:22:10.2530004Z test/dtypes/test_nf4.py::TestNF4Linear::test_empty_like_input_size1 PASSED 2025-09-09T14:22:10.2530698Z test/dtypes/test_nf4.py::TestNF4Linear::test_load_from_nf4_diff_meta_bfloat16 PASSED 2025-09-09T14:22:10.2531419Z test/dtypes/test_nf4.py::TestNF4Linear::test_load_from_nf4_diff_meta_float16 PASSED 2025-09-09T14:22:10.2532129Z test/dtypes/test_nf4.py::TestNF4Linear::test_load_from_nf4_diff_meta_float32 PASSED 2025-09-09T14:22:10.2532840Z test/dtypes/test_nf4.py::TestNF4Linear::test_load_from_nf4_same_meta_bfloat16 PASSED 2025-09-09T14:22:10.2533562Z test/dtypes/test_nf4.py::TestNF4Linear::test_load_from_nf4_same_meta_float16 PASSED 2025-09-09T14:22:10.2534272Z test/dtypes/test_nf4.py::TestNF4Linear::test_load_from_nf4_same_meta_float32 PASSED 2025-09-09T14:22:10.2535103Z test/dtypes/test_nf4.py::TestNF4Linear::test_load_from_state_dicts_bfloat16 PASSED 2025-09-09T14:22:10.2536077Z test/dtypes/test_nf4.py::TestNF4Linear::test_load_from_state_dicts_float16 PASSED 2025-09-09T14:22:10.2537042Z test/dtypes/test_nf4.py::TestNF4Linear::test_load_from_state_dicts_float32 PASSED 2025-09-09T14:22:10.2537749Z test/dtypes/test_nf4.py::TestNF4Linear::test_nf4_bnb_linear_bfloat16 SKIPPED 2025-09-09T14:22:10.2538412Z test/dtypes/test_nf4.py::TestNF4Linear::test_nf4_bnb_linear_float16 SKIPPED 2025-09-09T14:22:10.2539064Z test/dtypes/test_nf4.py::TestNF4Linear::test_nf4_bnb_linear_float32 SKIPPED 2025-09-09T14:22:10.2539738Z test/dtypes/test_nf4.py::TestNF4Linear::test_output_dtype_match_bfloat16 PASSED 2025-09-09T14:22:10.2540414Z test/dtypes/test_nf4.py::TestNF4Linear::test_output_dtype_match_float16 PASSED 2025-09-09T14:22:10.2541092Z test/dtypes/test_nf4.py::TestNF4Linear::test_output_dtype_match_float32 PASSED 2025-09-09T14:22:10.2541776Z test/dtypes/test_nf4.py::TestNF4Linear::test_quantize_api_compile_False PASSED 2025-09-09T14:22:10.2542449Z test/dtypes/test_nf4.py::TestNF4Linear::test_quantize_api_compile_True PASSED 2025-09-09T14:22:10.2543194Z test/dtypes/test_nf4.py::TestNF4Linear::test_reconstruction_qlora_vs_bnb_bfloat16 SKIPPED 2025-09-09T14:22:10.2543961Z test/dtypes/test_nf4.py::TestNF4Linear::test_reconstruction_qlora_vs_bnb_float16 SKIPPED 2025-09-09T14:22:10.2544730Z test/dtypes/test_nf4.py::TestNF4Linear::test_reconstruction_qlora_vs_bnb_float32 SKIPPED 2025-09-09T14:22:10.2545461Z test/dtypes/test_nf4.py::TestNF4Linear::test_register_nf4_as_param_bfloat16 PASSED 2025-09-09T14:22:10.2546171Z test/dtypes/test_nf4.py::TestNF4Linear::test_register_nf4_as_param_float16 PASSED 2025-09-09T14:22:10.2546872Z test/dtypes/test_nf4.py::TestNF4Linear::test_register_nf4_as_param_float32 PASSED 2025-09-09T14:22:10.2547558Z test/dtypes/test_nf4.py::TestNF4Linear::test_smoketest_linear_bfloat16 PASSED 2025-09-09T14:22:10.2548230Z test/dtypes/test_nf4.py::TestNF4Linear::test_smoketest_linear_compile_bfloat16 AUTOTUNE mm(64x32, 32x32) 2025-09-09T14:22:10.2549053Z triton_mm_1 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:22:16.6075109Z triton_mm_4 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:22:16.6076282Z triton_mm_5 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:22:16.6077391Z triton_mm_6 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:22:16.6078515Z triton_mm_7 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:22:16.6079604Z triton_mm_9 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:22:16.6080721Z triton_mm_10 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:22:16.6081821Z triton_mm_11 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=8 2025-09-09T14:22:16.6082909Z triton_mm_2 0.0216 ms 99.7% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:22:16.6084018Z triton_mm_3 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:22:16.6085568Z SingleProcess AUTOTUNE benchmarking takes 0.5303 seconds and 0.2870 seconds precompiling for 13 choices 2025-09-09T14:22:16.6086458Z PASSED 2025-09-09T14:22:16.6087224Z test/dtypes/test_nf4.py::TestNF4Linear::test_smoketest_linear_compile_float16 AUTOTUNE mm(64x32, 32x32) 2025-09-09T14:22:16.6088248Z triton_mm_23 0.0184 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=8 2025-09-09T14:22:16.6089336Z triton_mm_13 0.0195 ms 94.7% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:22:16.6090419Z triton_mm_15 0.0195 ms 94.7% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:22:16.6091501Z triton_mm_16 0.0195 ms 94.7% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:22:16.6092589Z triton_mm_17 0.0195 ms 94.7% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:22:16.6093666Z triton_mm_18 0.0195 ms 94.7% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:22:16.6094730Z triton_mm_12 0.0205 ms 90.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:22:16.6095805Z triton_mm_14 0.0205 ms 90.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:22:16.6096963Z triton_mm_19 0.0205 ms 90.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:22:16.6098240Z triton_mm_20 0.0205 ms 90.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:22:16.6099254Z SingleProcess AUTOTUNE benchmarking takes 0.3822 seconds and 0.1783 seconds precompiling for 13 choices 2025-09-09T14:22:16.6099935Z PASSED 2025-09-09T14:22:16.6100490Z test/dtypes/test_nf4.py::TestNF4Linear::test_smoketest_linear_compile_float32 AUTOTUNE mm(64x32, 32x32) 2025-09-09T14:22:16.6101500Z triton_mm_26 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:22:16.6102584Z triton_mm_27 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:22:16.6103669Z triton_mm_28 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:22:16.6104768Z triton_mm_29 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:22:16.6105847Z triton_mm_30 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:22:16.6106947Z triton_mm_31 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:22:16.6108073Z triton_mm_32 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:22:16.6109161Z triton_mm_33 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:22:16.6110378Z triton_mm_35 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=8 2025-09-09T14:22:16.6111475Z triton_mm_24 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:22:16.6112609Z SingleProcess AUTOTUNE benchmarking takes 0.2534 seconds and 0.2181 seconds precompiling for 13 choices 2025-09-09T14:22:16.6113277Z PASSED 2025-09-09T14:22:16.6113830Z test/dtypes/test_nf4.py::TestNF4Linear::test_smoketest_linear_float16 PASSED 2025-09-09T14:22:16.6114663Z test/dtypes/test_nf4.py::TestNF4Linear::test_smoketest_linear_float32 PASSED 2025-09-09T14:22:16.6115453Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_copy_bfloat16 PASSED 2025-09-09T14:22:16.6116198Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_copy_device PASSED 2025-09-09T14:22:16.6116931Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_copy_float16 PASSED 2025-09-09T14:22:16.6117669Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_copy_float32 PASSED 2025-09-09T14:22:16.6118470Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_dtype_bfloat16 PASSED 2025-09-09T14:22:16.6119225Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_dtype_float16 PASSED 2025-09-09T14:22:16.6119965Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_dtype_float32 PASSED 2025-09-09T14:22:16.6120670Z test/dtypes/test_nf4.py::TestFSDPOps::test_pin_memory PASSED 2025-09-09T14:22:16.6121459Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_2d_view_valid_input_size0 PASSED 2025-09-09T14:22:16.6122357Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_as_strided_invalid_input_size0 PASSED 2025-09-09T14:22:16.6123281Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_as_strided_invalid_input_size1 PASSED 2025-09-09T14:22:16.6124188Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_as_strided_valid_input_size1 PASSED 2025-09-09T14:22:16.6125094Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_as_strided_valid_input_size2 PASSED 2025-09-09T14:22:16.6125869Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_as_strided_valid_input_size_262144 PASSED 2025-09-09T14:22:16.6126593Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_deepcopy_input_size1 PASSED 2025-09-09T14:22:16.6127264Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_deepcopy_input_size2 PASSED 2025-09-09T14:22:16.6127963Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_deepcopy_input_size_262144 PASSED 2025-09-09T14:22:16.6128691Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_new_zeros_invalid_input_size1 PASSED 2025-09-09T14:22:16.6129435Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_new_zeros_invalid_input_size2 PASSED 2025-09-09T14:22:16.6130196Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_new_zeros_invalid_input_size_262144 PASSED 2025-09-09T14:22:16.6130957Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_new_zeros_valid_input_size1 PASSED 2025-09-09T14:22:16.6131681Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_new_zeros_valid_input_size2 PASSED 2025-09-09T14:22:16.6132430Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_new_zeros_valid_input_size_262144 PASSED 2025-09-09T14:22:16.6133145Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_slice_1d_invalid PASSED 2025-09-09T14:22:16.6133783Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_slice_2d_invalid PASSED 2025-09-09T14:22:16.6134454Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_slice_valid_input_size1 PASSED 2025-09-09T14:22:16.6135152Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_slice_valid_input_size2 PASSED 2025-09-09T14:22:16.6135935Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_slice_valid_input_size_262144 PASSED 2025-09-09T14:22:16.6136662Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_view_invalid_input_size0 PASSED 2025-09-09T14:22:16.6137448Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_view_valid_input_size0 PASSED 2025-09-09T14:22:16.6138180Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_view_valid_input_size1 PASSED 2025-09-09T14:22:16.6138861Z test/dtypes/test_nf4.py::TestFSDPOps::test_to_cpu PASSED 2025-09-09T14:22:16.6139397Z test/dtypes/test_nf4.py::TestFSDPOps::test_to_cuda PASSED 2025-09-09T14:24:44.2502605Z test/dtypes/test_nf4.py::TestFSDPOps::test_to_module PASSED 2025-09-09T14:24:44.2505617Z test/dtypes/test_nf4.py::TestFSDPOps::test_torch_chunk_invalid_3d_input_size0 PASSED 2025-09-09T14:24:44.2506409Z test/dtypes/test_nf4.py::TestFSDPOps::test_torch_chunk_invalid_divide_input_size1 PASSED 2025-09-09T14:24:44.2507167Z test/dtypes/test_nf4.py::TestFSDPOps::test_torch_chunk_invalid_divide_input_size2 PASSED 2025-09-09T14:24:44.2507957Z test/dtypes/test_nf4.py::TestFSDPOps::test_torch_chunk_invalid_divide_input_size_261632 PASSED 2025-09-09T14:24:44.2508726Z test/dtypes/test_nf4.py::TestFSDPOps::test_torch_chunk_valid_input_size1 PASSED 2025-09-09T14:24:44.2509408Z test/dtypes/test_nf4.py::TestFSDPOps::test_torch_chunk_valid_input_size2 PASSED 2025-09-09T14:24:44.2510138Z test/dtypes/test_nf4.py::TestFSDPOps::test_torch_chunk_valid_input_size_262144 PASSED 2025-09-09T14:24:44.2510750Z test/dtypes/test_nf4.py::TestQLoRA::test_qlora_fsdp2 dist init r=0, world=2 2025-09-09T14:24:44.2511145Z dist init r=1, world=2 2025-09-09T14:24:44.2511407Z PASSED 2025-09-09T14:24:44.2511718Z test/dtypes/test_nf4.py::TestComm::test_comm dist init r=1, world=2 2025-09-09T14:24:44.2512093Z dist init r=0, world=2 2025-09-09T14:24:44.2512341Z PASSED 2025-09-09T14:24:44.2512760Z test/dtypes/test_uint4.py::TestUInt4::test_basic_tensor_ops SKIPPED 2025-09-09T14:24:44.2513382Z test/dtypes/test_uint4.py::TestUInt4::test_gpu_quant SKIPPED (FAILED...) 2025-09-09T14:24:44.2514024Z test/dtypes/test_uint4.py::TestUInt4::test_pt2e_quant SKIPPED (FAILE...) 2025-09-09T14:24:44.2514718Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[32-dtype0] PASSED 2025-09-09T14:24:44.2515446Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[32-dtype1] PASSED 2025-09-09T14:24:44.2516174Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[32-dtype2] PASSED 2025-09-09T14:24:44.2516893Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[32-dtype3] PASSED 2025-09-09T14:24:44.2517621Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[32-dtype4] PASSED 2025-09-09T14:24:44.2518345Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[32-dtype5] PASSED 2025-09-09T14:24:44.2519064Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[32-dtype6] PASSED 2025-09-09T14:24:44.2519786Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[64-dtype0] PASSED 2025-09-09T14:24:44.2520506Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[64-dtype1] PASSED 2025-09-09T14:24:44.2521234Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[64-dtype2] PASSED 2025-09-09T14:24:44.2521966Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[64-dtype3] PASSED 2025-09-09T14:24:44.2522682Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[64-dtype4] PASSED 2025-09-09T14:24:44.2523414Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[64-dtype5] PASSED 2025-09-09T14:24:44.2524133Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[64-dtype6] PASSED 2025-09-09T14:24:44.2524917Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[128-dtype0] PASSED 2025-09-09T14:24:44.2525658Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[128-dtype1] PASSED 2025-09-09T14:24:44.2527748Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[128-dtype2] PASSED 2025-09-09T14:24:44.2528499Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[128-dtype3] PASSED 2025-09-09T14:24:44.2529430Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[128-dtype4] PASSED 2025-09-09T14:24:44.2530164Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[128-dtype5] PASSED 2025-09-09T14:24:44.2530903Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[128-dtype6] PASSED 2025-09-09T14:24:44.2531625Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-32-dtype0] PASSED 2025-09-09T14:24:44.2532347Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-32-dtype1] PASSED 2025-09-09T14:24:44.2533071Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-32-dtype2] PASSED 2025-09-09T14:24:44.2533802Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-32-dtype3] PASSED 2025-09-09T14:24:44.2534514Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-32-dtype4] PASSED 2025-09-09T14:24:44.2535294Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-32-dtype5] PASSED 2025-09-09T14:24:44.2536086Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-32-dtype6] PASSED 2025-09-09T14:24:44.2536803Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-64-dtype0] PASSED 2025-09-09T14:24:44.2537526Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-64-dtype1] PASSED 2025-09-09T14:24:44.2538241Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-64-dtype2] PASSED 2025-09-09T14:24:44.2538956Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-64-dtype3] PASSED 2025-09-09T14:24:44.2539677Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-64-dtype4] PASSED 2025-09-09T14:24:44.2540385Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-64-dtype5] PASSED 2025-09-09T14:24:44.2541108Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-64-dtype6] PASSED 2025-09-09T14:24:44.2541831Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-128-dtype0] PASSED 2025-09-09T14:24:44.2542556Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-128-dtype1] PASSED 2025-09-09T14:24:44.2543287Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-128-dtype2] PASSED 2025-09-09T14:24:44.2544002Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-128-dtype3] PASSED 2025-09-09T14:24:44.2544729Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-128-dtype4] PASSED 2025-09-09T14:24:44.2545456Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-128-dtype5] PASSED 2025-09-09T14:24:44.2546185Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-128-dtype6] PASSED 2025-09-09T14:24:44.2546902Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-32-dtype0] PASSED 2025-09-09T14:24:44.2547637Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-32-dtype1] PASSED 2025-09-09T14:24:44.2548364Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-32-dtype2] PASSED 2025-09-09T14:24:44.2549079Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-32-dtype3] PASSED 2025-09-09T14:24:44.2549809Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-32-dtype4] PASSED 2025-09-09T14:24:44.2550523Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-32-dtype5] PASSED 2025-09-09T14:24:44.2551242Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-32-dtype6] PASSED 2025-09-09T14:24:44.2552058Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-64-dtype0] PASSED 2025-09-09T14:24:44.2552779Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-64-dtype1] PASSED 2025-09-09T14:24:44.2553583Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-64-dtype2] PASSED 2025-09-09T14:24:44.2554298Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-64-dtype3] PASSED 2025-09-09T14:24:44.2555073Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-64-dtype4] PASSED 2025-09-09T14:24:44.2555796Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-64-dtype5] PASSED 2025-09-09T14:24:44.2556514Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-64-dtype6] PASSED 2025-09-09T14:24:44.2557249Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-128-dtype0] PASSED 2025-09-09T14:24:44.2557989Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-128-dtype1] PASSED 2025-09-09T14:24:44.2558726Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-128-dtype2] PASSED 2025-09-09T14:24:44.2559471Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-128-dtype3] PASSED 2025-09-09T14:24:44.2560202Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-128-dtype4] PASSED 2025-09-09T14:24:44.2560941Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-128-dtype5] PASSED 2025-09-09T14:24:44.2561668Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-128-dtype6] PASSED 2025-09-09T14:24:44.2562377Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-32-dtype0] PASSED 2025-09-09T14:24:44.2563049Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-32-dtype1] PASSED 2025-09-09T14:24:44.2563731Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-32-dtype2] PASSED 2025-09-09T14:24:44.2564423Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-32-dtype3] PASSED 2025-09-09T14:24:44.2565143Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-32-dtype4] PASSED 2025-09-09T14:24:44.2565830Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-32-dtype5] PASSED 2025-09-09T14:24:44.2566503Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-32-dtype6] PASSED 2025-09-09T14:24:44.2567188Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-64-dtype0] PASSED 2025-09-09T14:24:44.2567863Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-64-dtype1] PASSED 2025-09-09T14:24:44.2568526Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-64-dtype2] PASSED 2025-09-09T14:24:44.2569202Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-64-dtype3] PASSED 2025-09-09T14:24:58.4231511Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-64-dtype4] PASSED 2025-09-09T14:24:58.4232549Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-64-dtype5] PASSED 2025-09-09T14:24:58.4233345Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-64-dtype6] PASSED 2025-09-09T14:24:58.4234017Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-128-dtype0] PASSED 2025-09-09T14:24:58.4234689Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-128-dtype1] PASSED 2025-09-09T14:24:58.4235351Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-128-dtype2] PASSED 2025-09-09T14:24:58.4236021Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-128-dtype3] PASSED 2025-09-09T14:24:58.4236692Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-128-dtype4] PASSED 2025-09-09T14:24:58.4237353Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-128-dtype5] PASSED 2025-09-09T14:24:58.4238390Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-128-dtype6] PASSED 2025-09-09T14:24:58.4239065Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-32-dtype0] PASSED 2025-09-09T14:24:58.4239783Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-32-dtype1] PASSED 2025-09-09T14:24:58.4240686Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-32-dtype2] PASSED 2025-09-09T14:24:58.4241345Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-32-dtype3] PASSED 2025-09-09T14:24:58.4242007Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-32-dtype4] PASSED 2025-09-09T14:24:58.4242666Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-32-dtype5] PASSED 2025-09-09T14:24:58.4243331Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-32-dtype6] PASSED 2025-09-09T14:24:58.4243993Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-64-dtype0] PASSED 2025-09-09T14:24:58.4244665Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-64-dtype1] PASSED 2025-09-09T14:24:58.4245340Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-64-dtype2] PASSED 2025-09-09T14:24:58.4246016Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-64-dtype3] PASSED 2025-09-09T14:24:58.4246680Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-64-dtype4] PASSED 2025-09-09T14:24:58.4247344Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-64-dtype5] PASSED 2025-09-09T14:24:58.4248018Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-64-dtype6] PASSED 2025-09-09T14:24:58.4248692Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-128-dtype0] PASSED 2025-09-09T14:24:58.4249411Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-128-dtype1] PASSED 2025-09-09T14:24:58.4250084Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-128-dtype2] PASSED 2025-09-09T14:24:58.4250752Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-128-dtype3] PASSED 2025-09-09T14:24:58.4251418Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-128-dtype4] PASSED 2025-09-09T14:24:58.4252088Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-128-dtype5] PASSED 2025-09-09T14:24:58.4252754Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-128-dtype6] PASSED 2025-09-09T14:24:58.4253385Z test/dtypes/test_uintx.py::test_uintx_target_dtype[dtype0] PASSED 2025-09-09T14:24:58.4253964Z test/dtypes/test_uintx.py::test_uintx_target_dtype[dtype1] PASSED 2025-09-09T14:24:58.4254544Z test/dtypes/test_uintx.py::test_uintx_target_dtype[dtype2] PASSED 2025-09-09T14:24:58.4255113Z test/dtypes/test_uintx.py::test_uintx_target_dtype[dtype3] PASSED 2025-09-09T14:24:58.4255695Z test/dtypes/test_uintx.py::test_uintx_target_dtype[dtype4] PASSED 2025-09-09T14:24:58.4256380Z test/dtypes/test_uintx.py::test_uintx_target_dtype[dtype5] PASSED 2025-09-09T14:24:58.4256949Z test/dtypes/test_uintx.py::test_uintx_target_dtype[dtype6] PASSED 2025-09-09T14:24:58.4257554Z test/dtypes/test_uintx.py::test_uintx_target_dtype_compile[dtype0] PASSED 2025-09-09T14:24:58.4258187Z test/dtypes/test_uintx.py::test_uintx_target_dtype_compile[dtype1] PASSED 2025-09-09T14:24:58.4258810Z test/dtypes/test_uintx.py::test_uintx_target_dtype_compile[dtype2] PASSED 2025-09-09T14:24:58.4259430Z test/dtypes/test_uintx.py::test_uintx_target_dtype_compile[dtype3] PASSED 2025-09-09T14:24:58.4260054Z test/dtypes/test_uintx.py::test_uintx_target_dtype_compile[dtype4] PASSED 2025-09-09T14:24:58.4260680Z test/dtypes/test_uintx.py::test_uintx_target_dtype_compile[dtype5] PASSED 2025-09-09T14:24:58.4261295Z test/dtypes/test_uintx.py::test_uintx_target_dtype_compile[dtype6] PASSED 2025-09-09T14:24:58.4261894Z test/dtypes/test_uintx.py::test_uintx_model_size[dtype0] PASSED 2025-09-09T14:24:58.4262546Z test/dtypes/test_uintx.py::test_uintx_model_size[dtype1] PASSED 2025-09-09T14:24:58.4263117Z test/dtypes/test_uintx.py::test_uintx_model_size[dtype2] PASSED 2025-09-09T14:24:58.4263760Z test/dtypes/test_uintx.py::test_uintx_model_size[dtype3] PASSED 2025-09-09T14:24:58.4264318Z test/dtypes/test_uintx.py::test_uintx_model_size[dtype4] PASSED 2025-09-09T14:24:58.4264883Z test/dtypes/test_uintx.py::test_uintx_model_size[dtype5] PASSED 2025-09-09T14:24:58.4265440Z test/dtypes/test_uintx.py::test_uintx_model_size[dtype6] PASSED 2025-09-09T14:24:58.4266253Z test/float8/test_auto_filter.py::test_end_to_end_filtering[tensorwise-module_dims0-valid.layer-filter_fqns0-True] PASSED 2025-09-09T14:24:58.4267288Z test/float8/test_auto_filter.py::test_end_to_end_filtering[tensorwise-module_dims1-skip_layer.linear-filter_fqns1-False] PASSED 2025-09-09T14:24:58.4268333Z test/float8/test_auto_filter.py::test_end_to_end_filtering[tensorwise-module_dims2-valid.layer-filter_fqns2-False] PASSED 2025-09-09T14:24:58.4269330Z test/float8/test_auto_filter.py::test_end_to_end_filtering[rowwise-module_dims3-valid.layer-filter_fqns3-True] PASSED 2025-09-09T14:24:58.4270349Z test/float8/test_auto_filter.py::test_end_to_end_filtering[rowwise-module_dims4-skip_layer.linear-filter_fqns4-False] PASSED 2025-09-09T14:24:58.4271360Z test/float8/test_auto_filter.py::test_end_to_end_filtering[rowwise-module_dims5-valid.layer-filter_fqns5-False] PASSED 2025-09-09T14:24:58.4272183Z test/float8/test_auto_filter.py::test_exact_boundary_dimensions_rowwise PASSED 2025-09-09T14:24:58.4272870Z test/float8/test_auto_filter.py::test_exact_boundary_dimensions_tensorwise PASSED 2025-09-09T14:24:58.4273518Z test/float8/test_auto_filter.py::test_partial_fqn_matching PASSED 2025-09-09T14:24:58.4274170Z test/float8/test_base.py::TestFloat8TrainingTensor::test_preserves_dtype PASSED 2025-09-09T14:24:58.4274916Z test/float8/test_base.py::TestFloat8TrainingTensor::test_differentiable_casts PASSED 2025-09-09T14:24:58.4275616Z test/float8/test_base.py::TestFloat8TrainingTensor::test_split_cat PASSED 2025-09-09T14:24:58.4276279Z test/float8/test_base.py::TestFloat8TrainingTensor::test_index_put PASSED 2025-09-09T14:24:58.4276927Z test/float8/test_base.py::TestFloat8TrainingTensor::test_copy_ PASSED 2025-09-09T14:24:58.4277565Z test/float8/test_base.py::TestFloat8TrainingTensor::test_transpose PASSED 2025-09-09T14:24:58.4278341Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[True-0-shape0] PASSED 2025-09-09T14:24:58.4279216Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[True-0-shape1] PASSED 2025-09-09T14:24:58.4280092Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[True-0-shape2] PASSED 2025-09-09T14:24:58.4280978Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[True--1-shape0] PASSED 2025-09-09T14:24:58.4281855Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[True--1-shape1] PASSED 2025-09-09T14:24:58.4282742Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[True--1-shape2] PASSED 2025-09-09T14:24:58.4283620Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[False-0-shape0] PASSED 2025-09-09T14:24:58.4284502Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[False-0-shape1] PASSED 2025-09-09T14:24:58.4285383Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[False-0-shape2] PASSED 2025-09-09T14:24:58.4286262Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[False--1-shape0] PASSED 2025-09-09T14:24:58.4287312Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[False--1-shape1] PASSED 2025-09-09T14:24:58.4288196Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[False--1-shape2] PASSED 2025-09-09T14:24:58.4289121Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_reshape PASSED 2025-09-09T14:24:58.4290230Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_gemm[ScalingGranularity.AXISWISE-ScalingGranularity.AXISWISE-a_shape0] SKIPPED 2025-09-09T14:24:58.4291557Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_gemm[ScalingGranularity.AXISWISE-ScalingGranularity.AXISWISE-a_shape1] SKIPPED 2025-09-09T14:24:58.4292885Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_gemm[ScalingGranularity.AXISWISE-ScalingGranularity.AXISWISE-a_shape2] SKIPPED 2025-09-09T14:24:58.4294227Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_gemm[ScalingGranularity.AXISWISE-ScalingGranularity.TENSORWISE-a_shape0] SKIPPED 2025-09-09T14:24:58.4295583Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_gemm[ScalingGranularity.AXISWISE-ScalingGranularity.TENSORWISE-a_shape1] SKIPPED 2025-09-09T14:24:58.6604479Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_gemm[ScalingGranularity.AXISWISE-ScalingGranularity.TENSORWISE-a_shape2] SKIPPED 2025-09-09T14:24:58.6605999Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_gemm[ScalingGranularity.TENSORWISE-ScalingGranularity.AXISWISE-a_shape0] SKIPPED 2025-09-09T14:24:58.6607363Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_gemm[ScalingGranularity.TENSORWISE-ScalingGranularity.AXISWISE-a_shape1] SKIPPED 2025-09-09T14:24:58.6608730Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_gemm[ScalingGranularity.TENSORWISE-ScalingGranularity.AXISWISE-a_shape2] SKIPPED 2025-09-09T14:24:58.6609740Z test/float8/test_base.py::TestFloat8TrainingTensor::test_fp8_dtype PASSED 2025-09-09T14:24:58.6610960Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-False-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape0-True] PASSED 2025-09-09T14:24:58.6612629Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-False-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape1-True] PASSED 2025-09-09T14:24:58.6614285Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-False-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape2-True] PASSED 2025-09-09T14:24:58.6616032Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-False-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape0-True] PASSED 2025-09-09T14:24:58.6617691Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-False-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape1-True] PASSED 2025-09-09T14:24:58.6619333Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-False-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape2-True] PASSED 2025-09-09T14:24:58.6620979Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-True-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape0-True] PASSED 2025-09-09T14:24:58.6622622Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-True-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape1-True] PASSED 2025-09-09T14:24:58.6624257Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-True-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape2-True] PASSED 2025-09-09T14:24:58.6626065Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-True-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape0-True] PASSED 2025-09-09T14:24:58.6627828Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-True-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape1-True] PASSED 2025-09-09T14:24:58.6629462Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-True-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape2-True] PASSED 2025-09-09T14:24:58.6631160Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-False-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape0-True] PASSED 2025-09-09T14:24:58.6632809Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-False-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape1-True] PASSED 2025-09-09T14:24:58.6634448Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-False-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape2-True] PASSED 2025-09-09T14:24:58.6636106Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-False-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape0-True] PASSED 2025-09-09T14:24:58.6637746Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-False-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape1-True] PASSED 2025-09-09T14:24:58.6639423Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-False-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape2-True] PASSED 2025-09-09T14:24:58.6641093Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-True-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape0-True] PASSED 2025-09-09T14:24:58.6642724Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-True-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape1-True] PASSED 2025-09-09T14:24:58.6644360Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-True-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape2-True] PASSED 2025-09-09T14:24:58.6645997Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-True-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape0-True] PASSED 2025-09-09T14:24:58.6647628Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-True-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape1-True] PASSED 2025-09-09T14:24:58.6649264Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-True-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape2-True] PASSED 2025-09-09T14:24:58.6650666Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-True-x_shape0-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:24:58.6651906Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-True-x_shape0-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:24:58.6653129Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-True-x_shape1-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:24:58.6654367Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-True-x_shape1-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:24:58.6656846Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-True-x_shape2-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:24:58.6658080Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-True-x_shape2-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:24:58.6659400Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-False-x_shape0-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:24:58.6660640Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-False-x_shape0-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:24:58.6661882Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-False-x_shape1-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:24:58.6663118Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-False-x_shape1-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:24:58.6664352Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-False-x_shape2-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:24:58.6665591Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-False-x_shape2-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:24:58.6666832Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-True-x_shape0-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:24:58.6668058Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-True-x_shape0-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:24:58.6669284Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-True-x_shape1-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:24:58.6670516Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-True-x_shape1-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:24:58.6671737Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-True-x_shape2-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:24:58.6672971Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-True-x_shape2-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:24:58.6674195Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-False-x_shape0-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:24:59.3966818Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-False-x_shape0-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:24:59.3968220Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-False-x_shape1-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:24:59.3969468Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-False-x_shape1-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:24:59.3970688Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-False-x_shape2-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:24:59.3971924Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-False-x_shape2-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:24:59.3973139Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-True-x_shape0-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:24:59.3974355Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-True-x_shape0-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:24:59.3975594Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-True-x_shape1-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:24:59.3977061Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-True-x_shape1-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:24:59.3978289Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-True-x_shape2-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:24:59.3979687Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-True-x_shape2-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:24:59.3980909Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-False-x_shape0-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:24:59.3982141Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-False-x_shape0-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:24:59.3983369Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-False-x_shape1-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:24:59.3984603Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-False-x_shape1-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:24:59.3985833Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-False-x_shape2-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:24:59.3987065Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-False-x_shape2-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:24:59.3988264Z test/float8/test_base.py::TestFloat8Linear::test_autocast_outputs[Float8LinearRecipeName.TENSORWISE-linear_dtype0-True] PASSED 2025-09-09T14:24:59.3989401Z test/float8/test_base.py::TestFloat8Linear::test_autocast_outputs[Float8LinearRecipeName.TENSORWISE-linear_dtype1-True] PASSED 2025-09-09T14:24:59.3990583Z test/float8/test_base.py::TestFloat8Linear::test_autocast_outputs[Float8LinearRecipeName.TENSORWISE-linear_dtype2-True] PASSED 2025-09-09T14:24:59.3991703Z test/float8/test_base.py::TestFloat8Linear::test_autocast_outputs[Float8LinearRecipeName.ROWWISE-linear_dtype0-True] PASSED 2025-09-09T14:24:59.3992806Z test/float8/test_base.py::TestFloat8Linear::test_autocast_outputs[Float8LinearRecipeName.ROWWISE-linear_dtype1-True] PASSED 2025-09-09T14:24:59.3993904Z test/float8/test_base.py::TestFloat8Linear::test_autocast_outputs[Float8LinearRecipeName.ROWWISE-linear_dtype2-True] PASSED 2025-09-09T14:24:59.3995050Z test/float8/test_base.py::TestFloat8Linear::test_autocast_outputs[Float8LinearRecipeName.ROWWISE_WITH_GW_HP-linear_dtype0-True] PASSED 2025-09-09T14:24:59.3996232Z test/float8/test_base.py::TestFloat8Linear::test_autocast_outputs[Float8LinearRecipeName.ROWWISE_WITH_GW_HP-linear_dtype1-True] PASSED 2025-09-09T14:24:59.3997568Z test/float8/test_base.py::TestFloat8Linear::test_autocast_outputs[Float8LinearRecipeName.ROWWISE_WITH_GW_HP-linear_dtype2-True] PASSED 2025-09-09T14:24:59.3998429Z test/float8/test_base.py::TestFloat8Linear::test_repr PASSED 2025-09-09T14:24:59.3999037Z test/float8/test_base.py::TestFloat8Linear::test_inference_mode SKIPPED 2025-09-09T14:24:59.3999697Z test/float8/test_base.py::TestFloat8Linear::test_quantize SKIPPED (C...) 2025-09-09T14:24:59.4000418Z test/float8/test_base.py::TestScaledMM::test_scaled_mm_vs_emulated[True-base_dtype0] SKIPPED 2025-09-09T14:24:59.4001217Z test/float8/test_base.py::TestScaledMM::test_scaled_mm_vs_emulated[True-base_dtype1] SKIPPED 2025-09-09T14:24:59.4002005Z test/float8/test_base.py::TestScaledMM::test_scaled_mm_vs_emulated[True-base_dtype2] SKIPPED 2025-09-09T14:24:59.4002791Z test/float8/test_base.py::TestScaledMM::test_scaled_mm_vs_emulated[False-base_dtype0] SKIPPED 2025-09-09T14:24:59.4003597Z test/float8/test_base.py::TestScaledMM::test_scaled_mm_vs_emulated[False-base_dtype1] SKIPPED 2025-09-09T14:24:59.4004543Z test/float8/test_base.py::TestScaledMM::test_scaled_mm_vs_emulated[False-base_dtype2] SKIPPED 2025-09-09T14:24:59.4005286Z test/float8/test_base.py::TestScaledMM::test_different_configs_error SKIPPED 2025-09-09T14:24:59.4006114Z test/float8/test_base.py::TestScaledMM::test_pad_inner_dim[True-base_dtype0] SKIPPED 2025-09-09T14:24:59.4006843Z test/float8/test_base.py::TestScaledMM::test_pad_inner_dim[True-base_dtype1] SKIPPED 2025-09-09T14:24:59.4007583Z test/float8/test_base.py::TestScaledMM::test_pad_inner_dim[True-base_dtype2] SKIPPED 2025-09-09T14:24:59.4008314Z test/float8/test_base.py::TestScaledMM::test_pad_inner_dim[False-base_dtype0] SKIPPED 2025-09-09T14:24:59.4009053Z test/float8/test_base.py::TestScaledMM::test_pad_inner_dim[False-base_dtype1] SKIPPED 2025-09-09T14:24:59.4009796Z test/float8/test_base.py::TestScaledMM::test_pad_inner_dim[False-base_dtype2] SKIPPED 2025-09-09T14:24:59.4010526Z test/float8/test_base.py::TestNumerics::test_small_amax_float16[float8_dtype0] PASSED 2025-09-09T14:24:59.4011263Z test/float8/test_base.py::TestNumerics::test_small_amax_float16[float8_dtype1] PASSED 2025-09-09T14:24:59.4011987Z test/float8/test_base.py::TestNumerics::test_small_amax_float16[float8_dtype2] PASSED 2025-09-09T14:24:59.4012725Z test/float8/test_base.py::TestNumerics::test_small_amax_float16[float8_dtype3] PASSED 2025-09-09T14:24:59.4013460Z test/float8/test_base.py::TestFloat8LinearUtils::test_fp8_tensor_statistics PASSED 2025-09-09T14:24:59.4014214Z test/float8/test_base.py::TestFloat8LinearUtils::test_swap_linears_with_filters PASSED 2025-09-09T14:24:59.4014945Z test/float8/test_base.py::TestFloat8LinearUtils::test_swap_root_linear PASSED 2025-09-09T14:24:59.4015729Z test/float8/test_base.py::TestFloat8LinearUtils::test_swap_root_linear_with_children_raises PASSED 2025-09-09T14:24:59.4016594Z test/float8/test_base.py::TestFloat8LinearUtils::test_swap_submodule_linears PASSED 2025-09-09T14:24:59.4017389Z test/float8/test_base.py::TestFloat8LinearUtils::test_swap_submodule_linears_with_skip PASSED 2025-09-09T14:24:59.4018436Z test/float8/test_compile.py::test_eager_only[dtype0-True-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-True] PASSED 2025-09-09T14:24:59.4019692Z test/float8/test_compile.py::test_eager_only[dtype1-True-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-True] PASSED 2025-09-09T14:24:59.4020938Z test/float8/test_compile.py::test_aot_eager[dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-True-True] PASSED 2025-09-09T14:24:59.4022174Z test/float8/test_compile.py::test_aot_eager[dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-True-True] PASSED 2025-09-09T14:24:59.4023504Z test/float8/test_compile.py::test_inductor_from_config_params[dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-False-True] SKIPPED 2025-09-09T14:24:59.4024900Z test/float8/test_compile.py::test_inductor_from_config_params[dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-False-True] SKIPPED 2025-09-09T14:24:59.4026012Z test/float8/test_compile.py::test_inductor_from_recipe[Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:24:59.4026916Z test/float8/test_compile.py::test_inductor_from_recipe[Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:24:59.4027720Z test/float8/test_compile.py::TestGraphBreaks::test_float8_graph_input SKIPPED 2025-09-09T14:24:59.4028430Z test/float8/test_compile.py::TestGraphBreaks::test_float8_graph_output SKIPPED 2025-09-09T14:24:59.4029205Z test/float8/test_compile.py::TestGraphBreaks::test_float8_with_graph_break_in_the_middle SKIPPED 2025-09-09T14:24:59.4029987Z test/float8/test_compile.py::test_dynamic_scale_numeric_parity[True-dtype0] SKIPPED 2025-09-09T14:24:59.4030713Z test/float8/test_compile.py::test_dynamic_scale_numeric_parity[True-dtype1] SKIPPED 2025-09-09T14:24:59.4031532Z test/float8/test_compile.py::test_dynamic_scale_numeric_parity[True-dtype2] SKIPPED 2025-09-09T14:24:59.4032266Z test/float8/test_compile.py::test_dynamic_scale_numeric_parity[False-dtype0] SKIPPED 2025-09-09T14:24:59.4033080Z test/float8/test_compile.py::test_dynamic_scale_numeric_parity[False-dtype1] SKIPPED 2025-09-09T14:25:18.0570325Z test/float8/test_compile.py::test_dynamic_scale_numeric_parity[False-dtype2] SKIPPED 2025-09-09T14:25:18.0571178Z test/float8/test_float8_utils.py::test_round_scale_down_to_power_of_2_valid_inputs[test_case0] PASSED 2025-09-09T14:25:18.0572005Z test/float8/test_float8_utils.py::test_round_scale_down_to_power_of_2_valid_inputs[test_case1] PASSED 2025-09-09T14:25:18.0572823Z test/float8/test_float8_utils.py::test_round_scale_down_to_power_of_2_valid_inputs[test_case2] PASSED 2025-09-09T14:25:18.0573678Z test/float8/test_float8_utils.py::test_round_scale_down_to_power_of_2_valid_inputs[test_case3] PASSED 2025-09-09T14:25:18.0574489Z test/float8/test_float8_utils.py::test_round_scale_down_to_power_of_2_valid_inputs[test_case4] PASSED 2025-09-09T14:25:18.0575319Z test/float8/test_float8_utils.py::test_round_scale_down_to_power_of_2_valid_inputs[test_case5] PASSED 2025-09-09T14:25:18.0576256Z test/float8/test_float8_utils.py::test_round_scale_down_to_power_of_2_valid_inputs[test_case6] PASSED 2025-09-09T14:25:18.0577066Z test/float8/test_float8_utils.py::test_round_scale_down_to_power_of_2_valid_inputs[test_case7] PASSED 2025-09-09T14:25:18.0577814Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype0] PASSED 2025-09-09T14:25:18.0578486Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype1] PASSED 2025-09-09T14:25:18.0579157Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype2] PASSED 2025-09-09T14:25:18.0579833Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype3] PASSED 2025-09-09T14:25:18.0580494Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype4] PASSED 2025-09-09T14:25:18.0581169Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype5] PASSED 2025-09-09T14:25:18.0581829Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype6] PASSED 2025-09-09T14:25:18.0582498Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype7] PASSED 2025-09-09T14:25:18.0583706Z test/float8/test_numerics_integration.py::TestFloat8NumericsIntegrationTest::test_encoder_fw_bw_from_config_params[ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC] SKIPPED 2025-09-09T14:25:18.0585169Z test/float8/test_numerics_integration.py::TestFloat8NumericsIntegrationTest::test_encoder_fw_bw_from_recipe[Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:25:18.0586534Z test/float8/test_numerics_integration.py::TestFloat8NumericsIntegrationTest::test_encoder_fw_bw_from_recipe[Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:25:18.0587501Z test/hqq/test_hqq_affine.py::TestHQQ::test_hqq_plain_2bit PASSED 2025-09-09T14:25:18.0588086Z test/hqq/test_hqq_affine.py::TestHQQ::test_hqq_plain_3bit PASSED 2025-09-09T14:25:18.0588655Z test/hqq/test_hqq_affine.py::TestHQQ::test_hqq_plain_4bit PASSED 2025-09-09T14:25:18.0589217Z test/hqq/test_hqq_affine.py::TestHQQ::test_hqq_plain_5bit PASSED 2025-09-09T14:25:18.0589794Z test/hqq/test_hqq_affine.py::TestHQQ::test_hqq_plain_6bit PASSED 2025-09-09T14:25:18.0590353Z test/hqq/test_hqq_affine.py::TestHQQ::test_hqq_plain_7bit PASSED 2025-09-09T14:25:18.0590917Z test/hqq/test_hqq_affine.py::TestHQQ::test_hqq_plain_8bit PASSED 2025-09-09T14:25:18.0591601Z test/integration/test_integration.py::SmoothquantUnitTest::test_debug_x_absmax PASSED 2025-09-09T14:25:18.0592641Z test/integration/test_integration.py::SmoothquantUnitTest::test_figure_4 PASSED 2025-09-09T14:25:18.0593450Z test/integration/test_integration.py::SmoothquantUnitTest::test_selective_torch_compile PASSED 2025-09-09T14:25:18.0594937Z test/integration/test_integration.py::SmoothquantUnitTest::test_smooth_linear_cpu [W909 14:25:07.432442830 qlinear_dynamic.cpp:251] Warning: Currently, qnnpack incorrectly ignores reduce_range when it is set to true; this may change in a future release. (function operator()) 2025-09-09T14:25:18.0596154Z PASSED 2025-09-09T14:25:18.0604043Z test/integration/test_integration.py::SmoothquantUnitTest::test_smooth_linear_cuda PASSED 2025-09-09T14:25:18.0604989Z test/integration/test_integration.py::SmoothquantUnitTest::test_smooth_linear_edge_cases PASSED 2025-09-09T14:25:18.0605778Z test/integration/test_integration.py::SmoothquantUnitTest::test_swap PASSED 2025-09-09T14:25:18.0606549Z test/integration/test_integration.py::SmoothquantUnitTest::test_tensors PASSED 2025-09-09T14:25:18.0607341Z test/integration/test_integration.py::SmoothquantUnitTest::test_weight_t_and_non_t_numerics_match AUTOTUNE int_mm(32x32, 32x16) 2025-09-09T14:25:18.0608258Z triton_mm_39 0.0195 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:25:18.0609122Z triton_mm_38 0.0205 ms 95.3% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=2 2025-09-09T14:25:18.0609971Z triton_mm_36 0.0215 ms 90.9% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:25:18.0610809Z triton_mm_37 0.0215 ms 90.8% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=2 2025-09-09T14:25:18.0611368Z _int_mm 0.0348 ms 56.1% 2025-09-09T14:25:18.0611853Z SingleProcess AUTOTUNE benchmarking takes 0.3926 seconds and 0.1348 seconds precompiling for 5 choices 2025-09-09T14:25:18.0612383Z PASSED 2025-09-09T14:25:18.0612872Z test/integration/test_integration.py::PythonQuantUtilOpUnitTest::test__int_mm AUTOTUNE int_mm(32x32, 32x16) 2025-09-09T14:25:18.0613712Z triton_mm_40 0.0195 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:25:18.0614566Z triton_mm_42 0.0195 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=2 2025-09-09T14:25:18.0615409Z triton_mm_43 0.0195 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:25:18.0616301Z triton_mm_41 0.0205 ms 95.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=2 2025-09-09T14:25:18.0616867Z _int_mm 0.0338 ms 57.6% 2025-09-09T14:25:18.0617331Z SingleProcess AUTOTUNE benchmarking takes 0.0953 seconds and 0.1556 seconds precompiling for 5 choices 2025-09-09T14:25:18.0617861Z PASSED 2025-09-09T14:25:18.0618479Z test/integration/test_integration.py::PythonQuantUtilOpUnitTest::test__int_mm_eager_and_torch_compile_numerics AUTOTUNE int_mm(17x1536, 1536x1536) 2025-09-09T14:25:18.0619436Z triton_mm_53 0.0707 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=False, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:25:18.0620283Z triton_mm_56 0.0860 ms 82.1% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=64, EVEN_K=False, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:25:18.0621117Z triton_mm_54 0.0942 ms 75.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=False, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:25:18.0622120Z triton_mm_52 0.0952 ms 74.2% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=False, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:25:18.0622962Z triton_mm_48 0.1044 ms 67.6% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=False, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:25:18.0623909Z triton_mm_50 0.1044 ms 67.6% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=False, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:25:18.0624468Z _int_mm 0.1219 ms 58.0% 2025-09-09T14:25:18.0625007Z triton_mm_58 0.1444 ms 48.9% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=128, BLOCK_M=32, BLOCK_N=128, EVEN_K=False, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:25:18.0625881Z triton_mm_51 0.1618 ms 43.7% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=False, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:25:18.0626761Z triton_mm_55 0.1618 ms 43.7% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=False, GROUP_M=8, num_stages=2, num_warps=8 2025-09-09T14:25:18.0627550Z SingleProcess AUTOTUNE benchmarking takes 0.6571 seconds and 0.9409 seconds precompiling for 12 choices 2025-09-09T14:25:18.0628074Z AUTOTUNE int_mm(136x4096, 4096x1536) 2025-09-09T14:25:18.0628638Z triton_mm_64 0.2171 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=False, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:25:18.0629477Z triton_mm_67 0.2304 ms 94.2% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=False, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:25:18.0630314Z triton_mm_59 0.2693 ms 80.6% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=False, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:25:18.0631146Z triton_mm_63 0.2918 ms 74.4% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=128, BLOCK_N=64, EVEN_K=False, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:25:18.0631996Z triton_mm_61 0.3062 ms 70.9% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=128, BLOCK_N=64, EVEN_K=False, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:25:18.0632554Z _int_mm 0.3154 ms 68.8% 2025-09-09T14:25:18.0633075Z triton_mm_65 0.3461 ms 62.7% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=False, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:25:40.3429600Z triton_mm_62 0.4393 ms 49.4% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=False, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:25:40.3430496Z triton_mm_60 0.4516 ms 48.1% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=False, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:25:40.3431355Z triton_mm_66 0.4966 ms 43.7% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=128, BLOCK_N=128, EVEN_K=False, GROUP_M=8, num_stages=2, num_warps=8 2025-09-09T14:25:40.3432189Z SingleProcess AUTOTUNE benchmarking takes 0.8458 seconds and 1.8549 seconds precompiling for 12 choices 2025-09-09T14:25:40.3432915Z PASSED 2025-09-09T14:25:40.3433645Z test/integration/test_integration.py::PythonQuantUtilOpUnitTest::test_dynamic_quant_per_channel_numerics_cpu PASSED 2025-09-09T14:25:40.3434708Z test/integration/test_integration.py::PythonQuantUtilOpUnitTest::test_dynamic_quant_per_channel_numerics_cuda SKIPPED 2025-09-09T14:25:40.3435697Z test/integration/test_integration.py::PythonQuantUtilOpUnitTest::test_per_token_linear_cpu PASSED 2025-09-09T14:25:40.3436602Z test/integration/test_integration.py::PythonQuantUtilOpUnitTest::test_per_token_linear_cuda PASSED 2025-09-09T14:25:40.3437504Z test/integration/test_integration.py::PythonQuantUtilOpUnitTest::test_quantize_per_token_cpu PASSED 2025-09-09T14:25:40.3438420Z test/integration/test_integration.py::PythonQuantUtilOpUnitTest::test_quantize_per_token_cuda PASSED 2025-09-09T14:25:40.3439581Z test/integration/test_integration.py::PythonQuantUtilOpUnitTest::test_quantize_per_token_xpu SKIPPED 2025-09-09T14:25:40.3440574Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_rowwise_scaling_subclass_0_cpu SKIPPED 2025-09-09T14:25:40.3441759Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_rowwise_scaling_subclass_1_cpu SKIPPED 2025-09-09T14:25:40.3442774Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_rowwise_scaling_subclass_2_cpu SKIPPED 2025-09-09T14:25:40.3443851Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_rowwise_scaling_subclass_3_cuda SKIPPED 2025-09-09T14:25:40.3444882Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_rowwise_scaling_subclass_4_cuda SKIPPED 2025-09-09T14:25:40.3445900Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_rowwise_scaling_subclass_5_cuda SKIPPED 2025-09-09T14:25:40.3446947Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_tensorwise_scaling_subclass_0_cpu SKIPPED 2025-09-09T14:25:40.3447996Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_tensorwise_scaling_subclass_1_cpu SKIPPED 2025-09-09T14:25:40.3449043Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_tensorwise_scaling_subclass_2_cpu SKIPPED 2025-09-09T14:25:40.3450095Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_tensorwise_scaling_subclass_3_cuda SKIPPED 2025-09-09T14:25:40.3451143Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_tensorwise_scaling_subclass_4_cuda SKIPPED 2025-09-09T14:25:40.3452197Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_tensorwise_scaling_subclass_5_cuda SKIPPED 2025-09-09T14:25:40.3453194Z test/integration/test_integration.py::TestSubclass::test_aq_float8_weight_only_quant_subclass_0_cpu SKIPPED 2025-09-09T14:25:40.3454150Z test/integration/test_integration.py::TestSubclass::test_aq_float8_weight_only_quant_subclass_1_cpu SKIPPED 2025-09-09T14:25:40.3455091Z test/integration/test_integration.py::TestSubclass::test_aq_float8_weight_only_quant_subclass_2_cpu SKIPPED 2025-09-09T14:25:40.3456103Z test/integration/test_integration.py::TestSubclass::test_aq_float8_weight_only_quant_subclass_3_cuda SKIPPED 2025-09-09T14:25:40.3457039Z test/integration/test_integration.py::TestSubclass::test_aq_float8_weight_only_quant_subclass_4_cuda SKIPPED 2025-09-09T14:25:40.3457973Z test/integration/test_integration.py::TestSubclass::test_aq_float8_weight_only_quant_subclass_5_cuda SKIPPED 2025-09-09T14:25:40.3458873Z test/integration/test_integration.py::TestSubclass::test_aq_int8_dynamic_quant_subclass_0_cpu SKIPPED 2025-09-09T14:25:40.3459756Z test/integration/test_integration.py::TestSubclass::test_aq_int8_dynamic_quant_subclass_1_cpu SKIPPED 2025-09-09T14:25:40.3460629Z test/integration/test_integration.py::TestSubclass::test_aq_int8_dynamic_quant_subclass_2_cpu SKIPPED 2025-09-09T14:25:40.3461477Z test/integration/test_integration.py::TestSubclass::test_aq_int8_dynamic_quant_subclass_3_cuda AUTOTUNE int_mm(32x64, 64x32) 2025-09-09T14:25:40.3462363Z triton_mm_72 0.0195 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:25:40.3463214Z triton_mm_74 0.0195 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:25:40.3464064Z triton_mm_73 0.0204 ms 95.3% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:25:40.3465079Z triton_mm_71 0.0205 ms 95.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:25:40.3465921Z triton_mm_70 0.0246 ms 79.2% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:25:40.3466562Z _int_mm 0.0358 ms 54.3% 2025-09-09T14:25:40.3467032Z SingleProcess AUTOTUNE benchmarking takes 0.4207 seconds and 0.1726 seconds precompiling for 6 choices 2025-09-09T14:25:40.3467577Z PASSED 2025-09-09T14:25:40.3468147Z test/integration/test_integration.py::TestSubclass::test_aq_int8_dynamic_quant_subclass_4_cuda PASSED 2025-09-09T14:25:40.3469027Z test/integration/test_integration.py::TestSubclass::test_aq_int8_dynamic_quant_subclass_5_cuda PASSED 2025-09-09T14:25:40.3469941Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_2_subclass_0_cpu SKIPPED 2025-09-09T14:25:40.3470873Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_2_subclass_1_cpu SKIPPED 2025-09-09T14:25:40.3471804Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_2_subclass_2_cpu SKIPPED 2025-09-09T14:25:40.3472733Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_2_subclass_3_cuda PASSED 2025-09-09T14:25:40.3473651Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_2_subclass_4_cuda PASSED 2025-09-09T14:25:40.3474573Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_2_subclass_5_cuda PASSED 2025-09-09T14:25:40.3475493Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_3_subclass_0_cpu SKIPPED 2025-09-09T14:25:40.3476414Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_3_subclass_1_cpu SKIPPED 2025-09-09T14:25:40.3477346Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_3_subclass_2_cpu SKIPPED 2025-09-09T14:25:40.3478202Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_3_subclass_3_cuda AUTOTUNE mm(32x64, 64x32) 2025-09-09T14:25:40.3479100Z triton_mm_85 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:25:40.3479973Z triton_mm_86 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:25:40.3480845Z triton_mm_87 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:25:40.3481717Z triton_mm_88 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:25:40.3482595Z triton_mm_89 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:25:40.3483465Z triton_mm_90 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:25:40.3484335Z triton_mm_92 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:25:40.3485198Z triton_mm_93 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:25:40.3486069Z triton_mm_94 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:25:40.3486935Z triton_mm_91 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:25:40.3487821Z SingleProcess AUTOTUNE benchmarking takes 0.2126 seconds and 0.2434 seconds precompiling for 11 choices 2025-09-09T14:25:40.3488439Z PASSED 2025-09-09T14:25:48.6134482Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_3_subclass_4_cuda AUTOTUNE mm(32x64, 64x32) 2025-09-09T14:25:48.6135470Z triton_mm_95 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:25:48.6137365Z triton_mm_96 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:25:48.6139096Z triton_mm_97 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:25:48.6140830Z triton_mm_98 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:25:48.6142673Z triton_mm_100 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:25:48.6144857Z triton_mm_101 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:25:48.6146117Z triton_mm_102 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:25:48.6147217Z triton_mm_103 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:25:48.6148303Z triton_mm_104 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:25:48.6149380Z triton_mm_99 0.0225 ms 90.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:25:48.6150397Z SingleProcess AUTOTUNE benchmarking takes 0.2155 seconds and 0.1917 seconds precompiling for 11 choices 2025-09-09T14:25:48.6151254Z PASSED 2025-09-09T14:25:48.6151924Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_3_subclass_5_cuda AUTOTUNE mm(32x64, 64x32) 2025-09-09T14:25:48.6153044Z triton_mm_107 0.0214 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:25:48.6154131Z triton_mm_106 0.0215 ms 99.7% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:25:48.6155255Z triton_mm_108 0.0215 ms 99.7% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:25:48.6156358Z triton_mm_109 0.0215 ms 99.7% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:25:48.6157459Z triton_mm_110 0.0215 ms 99.7% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:25:48.6158549Z triton_mm_111 0.0215 ms 99.7% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:25:48.6159623Z triton_mm_112 0.0215 ms 99.7% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:25:48.6160706Z triton_mm_113 0.0215 ms 99.7% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:25:48.6162129Z triton_mm_114 0.0215 ms 99.7% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:25:48.6163227Z triton_mm_105 0.0225 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:25:48.6164438Z SingleProcess AUTOTUNE benchmarking takes 0.2141 seconds and 0.1998 seconds precompiling for 11 choices 2025-09-09T14:25:48.6165110Z PASSED 2025-09-09T14:25:48.6165886Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_subclass_0_cpu SKIPPED 2025-09-09T14:25:48.6167033Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_subclass_1_cpu SKIPPED 2025-09-09T14:25:48.6168170Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_subclass_2_cpu SKIPPED 2025-09-09T14:25:48.6169320Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_subclass_3_cuda SKIPPED 2025-09-09T14:25:48.6170476Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_subclass_4_cuda SKIPPED 2025-09-09T14:25:48.6171621Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_subclass_5_cuda SKIPPED 2025-09-09T14:25:48.6172751Z test/integration/test_integration.py::TestSubclass::test_autoquantizable_flatten_unflatten PASSED 2025-09-09T14:25:48.6173921Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_0_cpu SKIPPED 2025-09-09T14:25:48.6175155Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_1_cpu SKIPPED 2025-09-09T14:25:48.6176486Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_2_cpu SKIPPED 2025-09-09T14:25:48.6177721Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_3_cuda SKIPPED 2025-09-09T14:25:48.6178963Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_4_cuda SKIPPED 2025-09-09T14:25:48.6180205Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_5_cuda PASSED 2025-09-09T14:25:48.6181476Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_grouped_0_cpu SKIPPED 2025-09-09T14:25:48.6182787Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_grouped_1_cpu SKIPPED 2025-09-09T14:25:48.6184086Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_grouped_2_cpu SKIPPED 2025-09-09T14:25:48.6185460Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_grouped_3_cuda SKIPPED 2025-09-09T14:25:48.6186783Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_grouped_4_cuda SKIPPED 2025-09-09T14:25:48.6188100Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_grouped_5_cuda PASSED 2025-09-09T14:25:48.6189344Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_dynamic_quant_subclass_0_cpu PASSED 2025-09-09T14:25:48.6190530Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_dynamic_quant_subclass_1_cpu PASSED 2025-09-09T14:25:48.6191720Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_dynamic_quant_subclass_2_cpu PASSED 2025-09-09T14:25:48.6192918Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_dynamic_quant_subclass_3_cuda PASSED 2025-09-09T14:25:48.6194111Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_dynamic_quant_subclass_4_cuda PASSED 2025-09-09T14:25:48.6195457Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_dynamic_quant_subclass_5_cuda PASSED 2025-09-09T14:25:48.6196752Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_weight_only_quant_subclass_0_cpu PASSED 2025-09-09T14:25:48.6198162Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_weight_only_quant_subclass_1_cpu PASSED 2025-09-09T14:25:48.6199397Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_weight_only_quant_subclass_2_cpu PASSED 2025-09-09T14:25:48.6200626Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_weight_only_quant_subclass_3_cuda PASSED 2025-09-09T14:25:48.6201859Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_weight_only_quant_subclass_4_cuda PASSED 2025-09-09T14:25:48.6203091Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_weight_only_quant_subclass_5_cuda PASSED 2025-09-09T14:25:48.6204173Z test/integration/test_integration.py::TestSubclass::test_gemlite_layout_0_cpu SKIPPED 2025-09-09T14:25:48.6205137Z test/integration/test_integration.py::TestSubclass::test_gemlite_layout_1_cpu SKIPPED 2025-09-09T14:25:48.6206078Z test/integration/test_integration.py::TestSubclass::test_gemlite_layout_2_cpu SKIPPED 2025-09-09T14:25:48.6207037Z test/integration/test_integration.py::TestSubclass::test_gemlite_layout_3_cuda SKIPPED 2025-09-09T14:25:48.6207991Z test/integration/test_integration.py::TestSubclass::test_gemlite_layout_4_cuda SKIPPED 2025-09-09T14:25:48.6208965Z test/integration/test_integration.py::TestSubclass::test_gemlite_layout_5_cuda SKIPPED 2025-09-09T14:25:48.6210057Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_hqq_quant_subclass_api_0_cpu SKIPPED 2025-09-09T14:26:08.7015515Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_hqq_quant_subclass_api_1_cpu SKIPPED 2025-09-09T14:26:08.7017974Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_hqq_quant_subclass_api_2_cpu PASSED 2025-09-09T14:26:08.7019006Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_hqq_quant_subclass_api_3_cuda SKIPPED 2025-09-09T14:26:08.7019968Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_hqq_quant_subclass_api_4_cuda SKIPPED 2025-09-09T14:26:08.7020888Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_hqq_quant_subclass_api_5_cuda AUTOTUNE addmm(16x16, 16x16, 16x16) 2025-09-09T14:26:08.7021841Z triton_mm_116 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=1 2025-09-09T14:26:08.7022726Z triton_mm_117 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=1 2025-09-09T14:26:08.7023606Z triton_mm_118 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=1 2025-09-09T14:26:08.7024481Z triton_mm_119 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=1 2025-09-09T14:26:08.7025341Z triton_mm_115 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=1 2025-09-09T14:26:08.7025910Z addmm 0.0481 ms 44.7% 2025-09-09T14:26:08.7026149Z bias_addmm 0.0758 ms 28.4% 2025-09-09T14:26:08.7026633Z SingleProcess AUTOTUNE benchmarking takes 0.1350 seconds and 0.1622 seconds precompiling for 7 choices 2025-09-09T14:26:08.7027170Z PASSED 2025-09-09T14:26:08.7027749Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_0_cpu SKIPPED 2025-09-09T14:26:08.7029058Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_1_cpu SKIPPED 2025-09-09T14:26:08.7030116Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_2_cpu SKIPPED 2025-09-09T14:26:08.7031015Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_3_cuda SKIPPED 2025-09-09T14:26:08.7031915Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_4_cuda SKIPPED 2025-09-09T14:26:08.7032797Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_5_cuda PASSED 2025-09-09T14:26:08.7033706Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_0_cpu SKIPPED 2025-09-09T14:26:08.7034626Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_1_cpu SKIPPED 2025-09-09T14:26:08.7035555Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_2_cpu PASSED 2025-09-09T14:26:08.7036479Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_3_cuda SKIPPED 2025-09-09T14:26:08.7037410Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_4_cuda SKIPPED 2025-09-09T14:26:08.7038333Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_5_cuda PASSED 2025-09-09T14:26:08.7039284Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_grouped_0_cpu SKIPPED 2025-09-09T14:26:08.7040447Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_grouped_1_cpu SKIPPED 2025-09-09T14:26:08.7041425Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_grouped_2_cpu PASSED 2025-09-09T14:26:08.7042403Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_grouped_3_cuda SKIPPED 2025-09-09T14:26:08.7043393Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_grouped_4_cuda SKIPPED 2025-09-09T14:26:08.7044343Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_grouped_5_cuda AUTOTUNE addmm(256x16, 256x16, 16x16) 2025-09-09T14:26:08.7045294Z triton_mm_121 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:26:08.7046178Z triton_mm_123 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:26:08.7047051Z triton_mm_125 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:26:08.7047930Z triton_mm_128 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:26:08.7048824Z triton_mm_129 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=8 2025-09-09T14:26:08.7049706Z triton_mm_130 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:26:08.7050584Z triton_mm_127 0.0215 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:26:08.7051454Z triton_mm_120 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:26:08.7052409Z triton_mm_122 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:26:08.7053271Z triton_mm_124 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:26:08.7054278Z SingleProcess AUTOTUNE benchmarking takes 0.2638 seconds and 0.2039 seconds precompiling for 13 choices 2025-09-09T14:26:08.7054787Z AUTOTUNE addmm(256x8, 256x8, 8x8) 2025-09-09T14:26:08.7055361Z triton_mm_173 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=2, num_warps=8 2025-09-09T14:26:08.7056335Z triton_mm_165 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:26:08.7057211Z triton_mm_166 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:26:08.7058081Z triton_mm_167 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:26:08.7058945Z triton_mm_168 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:26:08.7059809Z triton_mm_169 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:26:08.7060675Z triton_mm_170 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:26:08.7061540Z triton_mm_172 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:26:08.7062422Z triton_mm_174 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:26:08.7063293Z triton_mm_164 0.0216 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:26:08.7064105Z SingleProcess AUTOTUNE benchmarking takes 0.2656 seconds and 0.2049 seconds precompiling for 13 choices 2025-09-09T14:26:08.7064648Z PASSED 2025-09-09T14:26:08.7065252Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_grouped_0_cpu SKIPPED 2025-09-09T14:26:08.7066211Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_grouped_1_cpu SKIPPED 2025-09-09T14:26:08.7067161Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_grouped_2_cpu SKIPPED 2025-09-09T14:26:08.7068124Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_grouped_3_cuda SKIPPED 2025-09-09T14:26:08.7069081Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_grouped_4_cuda SKIPPED 2025-09-09T14:26:08.7070034Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_grouped_5_cuda SKIPPED 2025-09-09T14:26:08.7070944Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_0_cpu SKIPPED 2025-09-09T14:26:08.7071806Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_1_cpu SKIPPED 2025-09-09T14:26:29.6823727Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_2_cpu SKIPPED 2025-09-09T14:26:29.6826079Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_3_cuda PASSED 2025-09-09T14:26:29.6827281Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_4_cuda PASSED 2025-09-09T14:26:29.6828166Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_5_cuda PASSED 2025-09-09T14:26:29.6829278Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_00_cpu SKIPPED 2025-09-09T14:26:29.6830170Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_01_cpu SKIPPED 2025-09-09T14:26:29.6831073Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_02_cpu SKIPPED 2025-09-09T14:26:29.6831960Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_03_cpu SKIPPED 2025-09-09T14:26:29.6832852Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_04_cpu SKIPPED 2025-09-09T14:26:29.6833753Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_05_cpu SKIPPED 2025-09-09T14:26:29.6834644Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_06_cuda SKIPPED 2025-09-09T14:26:29.6835550Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_07_cuda SKIPPED 2025-09-09T14:26:29.6836444Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_08_cuda SKIPPED 2025-09-09T14:26:29.6837345Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_09_cuda SKIPPED 2025-09-09T14:26:29.6838243Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_10_cuda SKIPPED 2025-09-09T14:26:29.6839184Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_11_cuda SKIPPED 2025-09-09T14:26:29.6840079Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_0_cpu SKIPPED 2025-09-09T14:26:29.6840970Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_1_cpu SKIPPED 2025-09-09T14:26:29.6841856Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_2_cpu SKIPPED 2025-09-09T14:26:29.6842699Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_3_cuda AUTOTUNE mm(32x64, 64x32) 2025-09-09T14:26:29.6843592Z triton_mm_224 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:26:29.6844492Z triton_mm_226 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:26:29.6845372Z triton_mm_229 0.0204 ms 95.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:26:29.6846251Z triton_mm_228 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:26:29.6847131Z triton_mm_232 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:26:29.6848011Z triton_mm_223 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:26:29.6848885Z triton_mm_225 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:26:29.6849768Z triton_mm_227 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:26:29.6850737Z triton_mm_230 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:26:29.6851613Z triton_mm_231 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:26:29.6852503Z SingleProcess AUTOTUNE benchmarking takes 0.2125 seconds and 0.3154 seconds precompiling for 11 choices 2025-09-09T14:26:29.6853047Z PASSED 2025-09-09T14:26:29.6853570Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_4_cuda AUTOTUNE mm(32x64, 64x32) 2025-09-09T14:26:29.6854452Z triton_mm_241 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:26:29.6855364Z triton_mm_233 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:26:29.6856373Z triton_mm_235 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:26:29.6857248Z triton_mm_236 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:26:29.6858128Z triton_mm_237 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:26:29.6859001Z triton_mm_238 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:26:29.6859866Z triton_mm_239 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:26:29.6860741Z triton_mm_240 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:26:29.6861606Z triton_mm_234 0.0215 ms 95.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:26:29.6862484Z triton_mm_242 0.0225 ms 90.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:26:29.6863299Z SingleProcess AUTOTUNE benchmarking takes 0.2131 seconds and 0.1936 seconds precompiling for 11 choices 2025-09-09T14:26:29.6863836Z PASSED 2025-09-09T14:26:29.6864353Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_5_cuda AUTOTUNE mm(32x64, 64x32) 2025-09-09T14:26:29.6865235Z triton_mm_250 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:26:29.6866125Z triton_mm_246 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:26:29.6867009Z triton_mm_243 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:26:29.6867880Z triton_mm_244 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:26:29.6868757Z triton_mm_245 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:26:29.6869629Z triton_mm_247 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:26:29.6870646Z triton_mm_248 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:26:29.6871522Z triton_mm_249 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:26:29.6872473Z triton_mm_251 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:26:29.6873338Z triton_mm_252 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:26:29.6874145Z SingleProcess AUTOTUNE benchmarking takes 0.2121 seconds and 0.2035 seconds precompiling for 11 choices 2025-09-09T14:26:29.6874685Z PASSED 2025-09-09T14:26:29.6875269Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_api_0_cpu PASSED 2025-09-09T14:26:29.6876188Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_api_1_cpu PASSED 2025-09-09T14:26:29.6877093Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_api_2_cpu PASSED 2025-09-09T14:26:29.6877955Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_api_3_cuda AUTOTUNE mm(32x32, 32x32) 2025-09-09T14:27:05.8794725Z triton_mm_263 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:27:05.8795656Z triton_mm_264 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:27:05.8796551Z triton_mm_265 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:27:05.8797671Z triton_mm_267 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:27:05.8798559Z triton_mm_266 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:27:05.8799450Z triton_mm_268 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:27:05.8800016Z mm 0.0348 ms 61.8% 2025-09-09T14:27:05.8800512Z SingleProcess AUTOTUNE benchmarking takes 0.1375 seconds and 0.0002 seconds precompiling for 7 choices 2025-09-09T14:27:05.8801277Z PASSED 2025-09-09T14:27:05.8801827Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_api_4_cuda AUTOTUNE mm(32x32, 32x32) 2025-09-09T14:27:05.8802731Z triton_mm_279 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:27:05.8803622Z triton_mm_280 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:27:05.8804516Z triton_mm_281 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:27:05.8805403Z triton_mm_282 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:27:05.8806287Z triton_mm_283 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:27:05.8807169Z triton_mm_284 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:27:05.8807752Z mm 0.0369 ms 58.3% 2025-09-09T14:27:05.8810340Z SingleProcess AUTOTUNE benchmarking takes 0.1318 seconds and 0.0005 seconds precompiling for 7 choices 2025-09-09T14:27:05.8810957Z PASSED 2025-09-09T14:27:05.8811486Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_api_5_cuda AUTOTUNE mm(32x32, 32x32) 2025-09-09T14:27:05.8812597Z triton_mm_295 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:27:05.8813497Z triton_mm_296 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:27:05.8814377Z triton_mm_297 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:27:05.8815270Z triton_mm_298 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:27:05.8816251Z triton_mm_299 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:27:05.8817135Z triton_mm_300 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:27:05.8817709Z mm 0.0358 ms 60.0% 2025-09-09T14:27:05.8818163Z SingleProcess AUTOTUNE benchmarking takes 0.1315 seconds and 0.0005 seconds precompiling for 7 choices 2025-09-09T14:27:05.8818707Z PASSED 2025-09-09T14:27:05.8819303Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_with_freeze_0_cpu AUTOTUNE packed_linear(32x64, 1459233x1, 32x64) 2025-09-09T14:27:05.8819983Z cpp_CppMicroGemmFP32Vec_0 0.0074 ms 100.0% 2025-09-09T14:27:05.8820292Z _mkl_linear 0.0321 ms 22.9% 2025-09-09T14:27:05.8820805Z SingleProcess AUTOTUNE benchmarking takes 0.2483 seconds and 2.0980 seconds precompiling for 2 choices 2025-09-09T14:27:05.8821350Z AUTOTUNE packed_linear(32x32, 1459233x1, 32x32) 2025-09-09T14:27:05.8821683Z cpp_CppMicroGemmFP32Vec_1 0.0060 ms 100.0% 2025-09-09T14:27:05.8829459Z _mkl_linear 0.0258 ms 23.1% 2025-09-09T14:27:05.8829997Z SingleProcess AUTOTUNE benchmarking takes 0.2480 seconds and 2.1032 seconds precompiling for 2 choices 2025-09-09T14:27:05.8830586Z PASSED 2025-09-09T14:27:05.8831139Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_with_freeze_1_cpu AUTOTUNE mm(32x64, 64x32) 2025-09-09T14:27:05.8831761Z cpp_CppMicroGemmFP32Vec_2 0.0063 ms 100.0% 2025-09-09T14:27:05.8832055Z mm 0.0273 ms 23.2% 2025-09-09T14:27:05.8832576Z SingleProcess AUTOTUNE benchmarking takes 0.2482 seconds and 2.2456 seconds precompiling for 2 choices 2025-09-09T14:27:05.8833163Z AUTOTUNE mm(32x32, 32x32) 2025-09-09T14:27:05.8833456Z cpp_CppMicroGemmFP32Vec_3 0.0062 ms 100.0% 2025-09-09T14:27:05.8833779Z mm 0.0297 ms 21.0% 2025-09-09T14:27:05.8834239Z SingleProcess AUTOTUNE benchmarking takes 0.2483 seconds and 2.2494 seconds precompiling for 2 choices 2025-09-09T14:27:05.8834774Z PASSED 2025-09-09T14:27:05.8835376Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_with_freeze_2_cpu AUTOTUNE _weight_int8pack_mm(32x64, 32x64, 32) 2025-09-09T14:27:05.8836064Z cpp_CppMicroGemmFP32Vec_4 0.0063 ms 100.0% 2025-09-09T14:27:05.8836372Z _weight_int8pack_mm 0.0167 ms 37.9% 2025-09-09T14:27:05.8836894Z SingleProcess AUTOTUNE benchmarking takes 0.2489 seconds and 2.2215 seconds precompiling for 2 choices 2025-09-09T14:27:05.8837437Z AUTOTUNE _weight_int8pack_mm(32x32, 32x32, 32) 2025-09-09T14:27:05.8837765Z cpp_CppMicroGemmFP32Vec_5 0.0063 ms 100.0% 2025-09-09T14:27:05.8838077Z _weight_int8pack_mm 0.0166 ms 37.7% 2025-09-09T14:27:05.8838576Z SingleProcess AUTOTUNE benchmarking takes 0.2486 seconds and 2.2298 seconds precompiling for 2 choices 2025-09-09T14:27:05.8839114Z PASSED 2025-09-09T14:27:05.8839760Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_with_freeze_3_cuda AUTOTUNE mm(32x64, 64x32) 2025-09-09T14:27:05.8840807Z triton_mm_302 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:27:05.8841716Z triton_mm_303 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:27:05.8842609Z triton_mm_304 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:27:05.8843502Z triton_mm_305 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:27:05.8844403Z triton_mm_306 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:27:05.8845292Z triton_mm_307 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:27:05.8846188Z triton_mm_308 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:27:05.8847073Z triton_mm_309 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:27:05.8847966Z triton_mm_310 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:27:05.8848850Z triton_mm_301 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:27:05.8849665Z SingleProcess AUTOTUNE benchmarking takes 0.2164 seconds and 0.2931 seconds precompiling for 11 choices 2025-09-09T14:27:05.8850185Z AUTOTUNE mm(32x32, 32x32) 2025-09-09T14:27:05.8850747Z triton_mm_311 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:27:05.8851638Z triton_mm_312 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:27:05.8852522Z triton_mm_313 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:27:05.8853402Z triton_mm_314 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:27:05.8854297Z triton_mm_315 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:27:05.8855188Z triton_mm_316 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:27:05.8855764Z mm 0.0338 ms 63.6% 2025-09-09T14:27:05.8856288Z SingleProcess AUTOTUNE benchmarking takes 0.1377 seconds and 0.0002 seconds precompiling for 7 choices 2025-09-09T14:27:05.8856819Z PASSED 2025-09-09T14:27:16.2232635Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_with_freeze_4_cuda AUTOTUNE mm(32x64, 64x32) 2025-09-09T14:27:16.2233573Z triton_mm_317 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:27:16.2234766Z triton_mm_318 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:27:16.2235657Z triton_mm_319 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:27:16.2236703Z triton_mm_320 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:27:16.2237578Z triton_mm_321 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:27:16.2238452Z triton_mm_322 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:27:16.2239326Z triton_mm_324 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:27:16.2240201Z triton_mm_325 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:27:16.2241074Z triton_mm_323 0.0215 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:27:16.2241945Z triton_mm_326 0.0216 ms 99.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:27:16.2242748Z SingleProcess AUTOTUNE benchmarking takes 0.2231 seconds and 0.0922 seconds precompiling for 11 choices 2025-09-09T14:27:16.2243267Z AUTOTUNE mm(32x32, 32x32) 2025-09-09T14:27:16.2243845Z triton_mm_327 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:27:16.2244728Z triton_mm_329 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:27:16.2245603Z triton_mm_331 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:27:16.2246473Z triton_mm_332 0.0225 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:27:16.2247346Z triton_mm_328 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:27:16.2248222Z triton_mm_330 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:27:16.2248792Z mm 0.0368 ms 58.4% 2025-09-09T14:27:16.2249243Z SingleProcess AUTOTUNE benchmarking takes 0.1399 seconds and 0.0002 seconds precompiling for 7 choices 2025-09-09T14:27:16.2249951Z PASSED 2025-09-09T14:27:16.2250501Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_with_freeze_5_cuda AUTOTUNE mm(32x64, 64x32) 2025-09-09T14:27:16.2251397Z triton_mm_333 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:27:16.2252271Z triton_mm_336 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:27:16.2253148Z triton_mm_339 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:27:16.2254077Z triton_mm_342 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:27:16.2255034Z triton_mm_340 0.0215 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:27:16.2255973Z triton_mm_334 0.0225 ms 95.6% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:27:16.2256928Z triton_mm_341 0.0225 ms 95.6% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:27:16.2257795Z triton_mm_335 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:27:16.2258665Z triton_mm_337 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:27:16.2259524Z triton_mm_338 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:27:16.2260339Z SingleProcess AUTOTUNE benchmarking takes 0.2230 seconds and 0.1078 seconds precompiling for 11 choices 2025-09-09T14:27:16.2260846Z AUTOTUNE mm(32x32, 32x32) 2025-09-09T14:27:16.2261394Z triton_mm_347 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:27:16.2262278Z triton_mm_344 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:27:16.2263143Z triton_mm_345 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:27:16.2264012Z triton_mm_348 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:27:16.2264880Z triton_mm_343 0.0215 ms 95.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:27:16.2265747Z triton_mm_346 0.0216 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:27:16.2266319Z mm 0.0368 ms 55.7% 2025-09-09T14:27:16.2266779Z SingleProcess AUTOTUNE benchmarking takes 0.1390 seconds and 0.0002 seconds precompiling for 7 choices 2025-09-09T14:27:16.2267310Z PASSED 2025-09-09T14:27:16.2267812Z test/integration/test_integration.py::TestDynamicQuant::test_dynamic_quant PASSED 2025-09-09T14:27:16.2268696Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_groupwise_embedding_quant PASSED 2025-09-09T14:27:16.2269664Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_groupwise_quant PASSED 2025-09-09T14:27:16.2270548Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant PASSED 2025-09-09T14:27:16.2271483Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_force_mixed_mm_0_cpu SKIPPED 2025-09-09T14:27:16.2272497Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_force_mixed_mm_1_cpu SKIPPED 2025-09-09T14:27:16.2273532Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_force_mixed_mm_2_cpu SKIPPED 2025-09-09T14:27:16.2274481Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_force_mixed_mm_3_cuda AUTOTUNE mm(2x4, 4x5) 2025-09-09T14:27:16.2275410Z triton_mm_350 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=2, num_warps=1 2025-09-09T14:27:16.2276477Z triton_mm_351 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=5, num_warps=1 2025-09-09T14:27:16.2277367Z triton_mm_352 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=3, num_warps=1 2025-09-09T14:27:16.2278317Z triton_mm_353 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=4, num_warps=1 2025-09-09T14:27:16.2279194Z triton_mm_349 0.0215 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=1, num_warps=1 2025-09-09T14:27:16.2279757Z mm 0.0338 ms 63.6% 2025-09-09T14:27:16.2280201Z SingleProcess AUTOTUNE benchmarking takes 0.1182 seconds and 0.1822 seconds precompiling for 6 choices 2025-09-09T14:27:16.2280703Z AUTOTUNE mm(125x4, 4x5) 2025-09-09T14:27:16.2281243Z triton_mm_362 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:27:16.2282136Z triton_mm_354 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:27:32.9438250Z triton_mm_355 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:27:32.9439222Z triton_mm_356 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:27:32.9440102Z triton_mm_358 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:27:32.9440976Z triton_mm_359 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:27:32.9441885Z triton_mm_360 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:27:32.9442770Z triton_mm_361 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:27:32.9443669Z triton_mm_363 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=2, num_warps=8 2025-09-09T14:27:32.9444558Z triton_mm_364 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:27:32.9445386Z SingleProcess AUTOTUNE benchmarking takes 0.2436 seconds and 0.2355 seconds precompiling for 12 choices 2025-09-09T14:27:32.9445894Z AUTOTUNE mm(4x4, 4x5) 2025-09-09T14:27:32.9446447Z triton_mm_366 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=2, num_warps=1 2025-09-09T14:27:32.9447338Z triton_mm_367 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=5, num_warps=1 2025-09-09T14:27:32.9448224Z triton_mm_368 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=3, num_warps=1 2025-09-09T14:27:32.9449161Z triton_mm_369 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=4, num_warps=1 2025-09-09T14:27:32.9450034Z triton_mm_365 0.0225 ms 90.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=1, num_warps=1 2025-09-09T14:27:32.9450607Z mm 0.0338 ms 60.6% 2025-09-09T14:27:32.9451061Z SingleProcess AUTOTUNE benchmarking takes 0.1170 seconds and 0.2129 seconds precompiling for 6 choices 2025-09-09T14:27:32.9451788Z PASSED 2025-09-09T14:27:32.9452687Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_force_mixed_mm_4_cuda AUTOTUNE mm(2x4, 4x5) 2025-09-09T14:27:32.9453626Z triton_mm_371 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=2, num_warps=1 2025-09-09T14:27:32.9454682Z triton_mm_372 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=5, num_warps=1 2025-09-09T14:27:32.9455580Z triton_mm_373 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=3, num_warps=1 2025-09-09T14:27:32.9456555Z triton_mm_374 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=4, num_warps=1 2025-09-09T14:27:32.9457448Z triton_mm_370 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=1, num_warps=1 2025-09-09T14:27:32.9458021Z mm 0.0410 ms 52.5% 2025-09-09T14:27:32.9458478Z SingleProcess AUTOTUNE benchmarking takes 0.1191 seconds and 0.1951 seconds precompiling for 6 choices 2025-09-09T14:27:32.9458992Z AUTOTUNE mm(125x4, 4x5) 2025-09-09T14:27:32.9459535Z triton_mm_377 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:27:32.9460425Z triton_mm_379 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:27:32.9461298Z triton_mm_378 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:27:32.9462176Z triton_mm_381 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:27:32.9463054Z triton_mm_382 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:27:32.9463937Z triton_mm_384 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=2, num_warps=8 2025-09-09T14:27:32.9464825Z triton_mm_385 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:27:32.9465701Z triton_mm_380 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:27:32.9466571Z triton_mm_383 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:27:32.9467455Z triton_mm_375 0.0225 ms 86.4% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:27:32.9468267Z SingleProcess AUTOTUNE benchmarking takes 0.2412 seconds and 0.2005 seconds precompiling for 12 choices 2025-09-09T14:27:32.9468768Z AUTOTUNE mm(4x4, 4x5) 2025-09-09T14:27:32.9469312Z triton_mm_387 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=2, num_warps=1 2025-09-09T14:27:32.9470187Z triton_mm_386 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=1, num_warps=1 2025-09-09T14:27:32.9471065Z triton_mm_388 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=5, num_warps=1 2025-09-09T14:27:32.9472028Z triton_mm_389 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=3, num_warps=1 2025-09-09T14:27:32.9472898Z triton_mm_390 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=4, num_warps=1 2025-09-09T14:27:32.9473551Z mm 0.0399 ms 51.3% 2025-09-09T14:27:32.9474004Z SingleProcess AUTOTUNE benchmarking takes 0.1170 seconds and 0.2047 seconds precompiling for 6 choices 2025-09-09T14:27:32.9474554Z PASSED 2025-09-09T14:27:32.9475122Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_force_mixed_mm_5_cuda AUTOTUNE mm(2x4, 4x5) 2025-09-09T14:27:32.9476047Z triton_mm_392 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=2, num_warps=1 2025-09-09T14:27:32.9476929Z triton_mm_391 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=1, num_warps=1 2025-09-09T14:27:32.9477807Z triton_mm_393 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=5, num_warps=1 2025-09-09T14:27:32.9478726Z triton_mm_394 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=3, num_warps=1 2025-09-09T14:27:32.9479627Z triton_mm_395 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=4, num_warps=1 2025-09-09T14:27:32.9480188Z mm 0.0338 ms 60.6% 2025-09-09T14:27:32.9480642Z SingleProcess AUTOTUNE benchmarking takes 0.1170 seconds and 0.1610 seconds precompiling for 6 choices 2025-09-09T14:27:32.9481140Z AUTOTUNE mm(125x4, 4x5) 2025-09-09T14:27:32.9481697Z triton_mm_403 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:27:32.9482605Z triton_mm_404 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:27:32.9483495Z triton_mm_401 0.0195 ms 99.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:27:32.9484375Z triton_mm_397 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:27:32.9485252Z triton_mm_399 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:27:32.9486122Z triton_mm_402 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:28:07.4225379Z triton_mm_398 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:28:07.4226306Z triton_mm_400 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:28:07.4227203Z triton_mm_405 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=2, num_warps=8 2025-09-09T14:28:07.4228086Z triton_mm_396 0.0225 ms 86.4% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:28:07.4228914Z SingleProcess AUTOTUNE benchmarking takes 0.2394 seconds and 0.2228 seconds precompiling for 12 choices 2025-09-09T14:28:07.4229429Z AUTOTUNE mm(4x4, 4x5) 2025-09-09T14:28:07.4229970Z triton_mm_409 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=5, num_warps=1 2025-09-09T14:28:07.4231262Z triton_mm_407 0.0215 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=1, num_warps=1 2025-09-09T14:28:07.4232312Z triton_mm_408 0.0215 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=2, num_warps=1 2025-09-09T14:28:07.4233202Z triton_mm_410 0.0215 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=3, num_warps=1 2025-09-09T14:28:07.4234081Z triton_mm_411 0.0215 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, num_stages=4, num_warps=1 2025-09-09T14:28:07.4234645Z mm 0.0348 ms 59.0% 2025-09-09T14:28:07.4235126Z SingleProcess AUTOTUNE benchmarking takes 0.1168 seconds and 0.2031 seconds precompiling for 6 choices 2025-09-09T14:28:07.4235839Z PASSED 2025-09-09T14:28:07.4236501Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_use_mixed_mm_0_cpu SKIPPED 2025-09-09T14:28:07.4237510Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_use_mixed_mm_1_cpu SKIPPED 2025-09-09T14:28:07.4238510Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_use_mixed_mm_2_cpu SKIPPED 2025-09-09T14:28:07.4239511Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_use_mixed_mm_3_cuda PASSED 2025-09-09T14:28:07.4240508Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_use_mixed_mm_4_cuda PASSED 2025-09-09T14:28:07.4241493Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_use_mixed_mm_5_cuda PASSED 2025-09-09T14:28:07.4242413Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_dqtensors_0_cpu SKIPPED 2025-09-09T14:28:07.4243263Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_dqtensors_1_cpu SKIPPED 2025-09-09T14:28:07.4244111Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_dqtensors_2_cpu SKIPPED 2025-09-09T14:28:07.4244928Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_dqtensors_3_cuda AUTOTUNE int_mm(32x32, 32x32) 2025-09-09T14:28:07.4245792Z triton_mm_482 0.0205 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:28:07.4246659Z triton_mm_481 0.0215 ms 95.2% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:28:07.4247512Z triton_mm_483 0.0215 ms 95.2% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:28:07.4248370Z triton_mm_480 0.0225 ms 90.9% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:28:07.4248934Z _int_mm 0.0348 ms 58.8% 2025-09-09T14:28:07.4249406Z SingleProcess AUTOTUNE benchmarking takes 0.0945 seconds and 0.1410 seconds precompiling for 5 choices 2025-09-09T14:28:07.4249918Z AUTOTUNE int_mm(32x64, 64x32) 2025-09-09T14:28:07.4250475Z triton_mm_476 0.0205 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:28:07.4251329Z triton_mm_478 0.0215 ms 95.4% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:28:07.4252183Z triton_mm_475 0.0215 ms 95.2% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:28:07.4253126Z triton_mm_477 0.0215 ms 95.2% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:28:07.4253985Z triton_mm_479 0.0215 ms 95.2% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:28:07.4254632Z _int_mm 0.0348 ms 58.8% 2025-09-09T14:28:07.4255104Z SingleProcess AUTOTUNE benchmarking takes 0.1127 seconds and 0.0002 seconds precompiling for 6 choices 2025-09-09T14:28:07.4255643Z PASSED 2025-09-09T14:28:07.4256270Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_dqtensors_4_cuda PASSED 2025-09-09T14:28:07.4257120Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_dqtensors_5_cuda PASSED 2025-09-09T14:28:07.4257994Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int4woqtensors_0_cpu SKIPPED 2025-09-09T14:28:07.4258876Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int4woqtensors_1_cpu SKIPPED 2025-09-09T14:28:07.4259767Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int4woqtensors_2_cpu PASSED 2025-09-09T14:28:07.4260707Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int4woqtensors_3_cuda SKIPPED 2025-09-09T14:28:07.4261601Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int4woqtensors_4_cuda SKIPPED 2025-09-09T14:28:07.4262434Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int4woqtensors_5_cuda AUTOTUNE mm(32x64, 64x32) 2025-09-09T14:28:07.4263315Z triton_mm_511 0.0204 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:28:07.4264209Z triton_mm_510 0.0205 ms 99.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:28:07.4265094Z triton_mm_502 0.0215 ms 95.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:28:07.4265978Z triton_mm_503 0.0215 ms 95.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:28:07.4266863Z triton_mm_504 0.0215 ms 95.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:28:07.4267735Z triton_mm_507 0.0215 ms 95.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:28:07.4268614Z triton_mm_508 0.0215 ms 95.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:28:07.4269509Z triton_mm_509 0.0215 ms 95.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:28:07.4270443Z triton_mm_505 0.0216 ms 94.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:28:07.4271326Z triton_mm_506 0.0225 ms 90.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:28:07.4272142Z SingleProcess AUTOTUNE benchmarking takes 0.2223 seconds and 0.1289 seconds precompiling for 11 choices 2025-09-09T14:28:07.4272666Z AUTOTUNE addmm(32x32, 32x32, 32x32) 2025-09-09T14:28:07.4273254Z triton_mm_516 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:28:07.4274136Z triton_mm_512 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:28:07.4275102Z triton_mm_513 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:28:07.4277571Z triton_mm_514 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:28:07.4278457Z triton_mm_517 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:28:07.4279336Z triton_mm_515 0.0215 ms 90.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:28:07.4279914Z addmm 0.0471 ms 41.3% 2025-09-09T14:28:21.4380411Z bias_addmm 0.0758 ms 25.7% 2025-09-09T14:28:21.4380962Z SingleProcess AUTOTUNE benchmarking takes 0.1542 seconds and 0.0002 seconds precompiling for 8 choices 2025-09-09T14:28:21.4381695Z PASSED 2025-09-09T14:28:21.4382303Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int8woqtensors_0_cpu PASSED 2025-09-09T14:28:21.4383203Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int8woqtensors_1_cpu PASSED 2025-09-09T14:28:21.4384103Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int8woqtensors_2_cpu PASSED 2025-09-09T14:28:21.4384922Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int8woqtensors_3_cuda AUTOTUNE mm(32x64, 64x32) 2025-09-09T14:28:21.4385826Z triton_mm_522 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:28:21.4386726Z triton_mm_520 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:28:21.4387610Z triton_mm_524 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:28:21.4388483Z triton_mm_527 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:28:21.4389365Z triton_mm_518 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:28:21.4390237Z triton_mm_519 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:28:21.4391134Z triton_mm_521 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:28:21.4392006Z triton_mm_523 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:28:21.4392889Z triton_mm_525 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:28:21.4393766Z triton_mm_526 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:28:21.4394614Z SingleProcess AUTOTUNE benchmarking takes 0.2117 seconds and 0.2529 seconds precompiling for 11 choices 2025-09-09T14:28:21.4395150Z AUTOTUNE mm(32x32, 32x32) 2025-09-09T14:28:21.4395708Z triton_mm_530 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:28:21.4396590Z triton_mm_528 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:28:21.4397993Z triton_mm_529 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:28:21.4398880Z triton_mm_531 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:28:21.4399913Z triton_mm_532 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:28:21.4400792Z triton_mm_533 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:28:21.4401360Z mm 0.0338 ms 60.6% 2025-09-09T14:28:21.4401824Z SingleProcess AUTOTUNE benchmarking takes 0.1315 seconds and 0.0002 seconds precompiling for 7 choices 2025-09-09T14:28:21.4402375Z PASSED 2025-09-09T14:28:21.4402894Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int8woqtensors_4_cuda AUTOTUNE mm(32x64, 64x32) 2025-09-09T14:28:21.4403806Z triton_mm_537 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:28:21.4404752Z triton_mm_538 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:28:21.4405643Z triton_mm_539 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:28:21.4406528Z triton_mm_540 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:28:21.4407412Z triton_mm_534 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:28:21.4408301Z triton_mm_535 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:28:21.4409180Z triton_mm_536 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:28:21.4410066Z triton_mm_541 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:28:21.4410951Z triton_mm_542 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:28:21.4411827Z triton_mm_543 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:28:21.4412637Z SingleProcess AUTOTUNE benchmarking takes 0.2150 seconds and 0.1040 seconds precompiling for 11 choices 2025-09-09T14:28:21.4413164Z AUTOTUNE mm(32x32, 32x32) 2025-09-09T14:28:21.4413715Z triton_mm_544 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:28:21.4414664Z triton_mm_545 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:28:21.4415551Z triton_mm_546 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:28:21.4416510Z triton_mm_547 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:28:21.4417396Z triton_mm_548 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:28:21.4418364Z triton_mm_549 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:28:21.4419006Z mm 0.0339 ms 60.5% 2025-09-09T14:28:21.4419468Z SingleProcess AUTOTUNE benchmarking takes 0.1352 seconds and 0.0002 seconds precompiling for 7 choices 2025-09-09T14:28:21.4420006Z PASSED 2025-09-09T14:28:21.4420581Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int8woqtensors_5_cuda PASSED 2025-09-09T14:28:21.4421398Z test/integration/test_integration.py::TorchCompileUnitTest::test_fullgraph PASSED 2025-09-09T14:28:21.4422134Z test/integration/test_integration.py::UtilsUnitTest::test_shape_logger PASSED 2025-09-09T14:28:21.4423022Z test/integration/test_integration.py::SmoothquantIntegrationTest::test_non_dynamically_quantizable_linear SKIPPED 2025-09-09T14:28:21.4423835Z test/integration/test_integration.py::SmoothquantIntegrationTest::test_on_dummy_distilbert 2025-09-09T14:28:21.4424429Z tokenizer_config.json: 0% 0.00/48.0 [00:00>time: 0.010ms for , to_beat: infms 2025-09-09T14:28:42.8145911Z AUTOTUNE mm(32x128, 128x128) 2025-09-09T14:28:42.8146484Z triton_mm_583 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:28:42.8147378Z triton_mm_587 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:28:42.8155642Z triton_mm_589 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:28:42.8156563Z triton_mm_591 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:28:42.8157480Z triton_mm_592 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:28:42.8158379Z triton_mm_593 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:28:42.8159278Z triton_mm_596 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:28:42.8160174Z triton_mm_597 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:28:42.8161090Z triton_mm_598 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=8 2025-09-09T14:28:42.8162127Z triton_mm_590 0.0205 ms 99.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:28:42.8162958Z SingleProcess AUTOTUNE benchmarking takes 0.3502 seconds and 1.8913 seconds precompiling for 18 choices 2025-09-09T14:28:42.8163908Z >>time: 0.014ms for , to_beat: 0.010ms 2025-09-09T14:28:42.8164839Z >>time: 0.004ms for , to_beat: 0.010ms 2025-09-09T14:28:42.8165456Z AUTOTUNE int_mm(32x128, 128x128) 2025-09-09T14:28:42.8166021Z triton_mm_607 0.0205 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=8 2025-09-09T14:28:42.8166899Z triton_mm_603 0.0205 ms 99.8% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:28:42.8167762Z triton_mm_600 0.0215 ms 95.2% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:28:42.8168622Z triton_mm_601 0.0215 ms 95.2% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:28:42.8169490Z triton_mm_602 0.0215 ms 95.2% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:28:42.8170339Z triton_mm_604 0.0215 ms 95.2% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:28:42.8171195Z triton_mm_605 0.0215 ms 95.2% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:28:42.8172058Z triton_mm_609 0.0215 ms 95.2% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:28:42.8172917Z triton_mm_606 0.0225 ms 90.9% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:28:42.8173770Z triton_mm_608 0.0225 ms 90.9% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:28:42.8174566Z SingleProcess AUTOTUNE benchmarking takes 0.2148 seconds and 0.2451 seconds precompiling for 11 choices 2025-09-09T14:28:42.8175431Z >>time: 0.007ms for matmul, to_beat: 0.004ms 2025-09-09T14:28:42.8176360Z best_cls= 2025-09-09T14:28:42.8176786Z 2025-09-09T14:28:42.8176959Z PASSED 2025-09-09T14:28:42.8177511Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_compile_12_cuda SKIPPED 2025-09-09T14:28:42.8178334Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_compile_13_cuda SKIPPED 2025-09-09T14:28:42.8179248Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_compile_14_cuda activation_shapes: torch.Size([32, 128]), times_seen: 2 2025-09-09T14:28:42.8180045Z weight_shape: torch.Size([128, 128]), dtype: torch.float16, bias_shape: torch.Size([128]) 2025-09-09T14:28:42.8180492Z AUTOTUNE addmm(32x128, 32x128, 128x128) 2025-09-09T14:28:42.8181096Z triton_mm_622 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:28:42.8181995Z triton_mm_614 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:28:42.8182889Z triton_mm_620 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:28:42.8183857Z triton_mm_610 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:28:42.8184815Z triton_mm_611 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:28:42.8185694Z triton_mm_612 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:28:42.8186572Z triton_mm_615 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:29:13.5416279Z triton_mm_617 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:29:13.5418510Z triton_mm_618 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:29:13.5419719Z triton_mm_619 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:29:13.5420563Z SingleProcess AUTOTUNE benchmarking takes 0.3861 seconds and 0.3536 seconds precompiling for 19 choices 2025-09-09T14:29:13.5421309Z >>time: 0.006ms for , to_beat: infms 2025-09-09T14:29:13.5421825Z AUTOTUNE mm(32x128, 128x128) 2025-09-09T14:29:13.5422394Z triton_mm_641 0.0184 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:29:13.5423410Z triton_mm_635 0.0195 ms 94.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:29:13.5424301Z triton_mm_628 0.0205 ms 90.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:29:13.5425187Z triton_mm_630 0.0205 ms 90.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:29:13.5426076Z triton_mm_631 0.0205 ms 90.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:29:13.5426958Z triton_mm_633 0.0205 ms 90.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:29:13.5427827Z triton_mm_634 0.0205 ms 90.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:29:13.5428709Z triton_mm_636 0.0205 ms 90.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:29:13.5429588Z triton_mm_637 0.0205 ms 90.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:29:13.5430483Z triton_mm_638 0.0205 ms 90.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:29:13.5431361Z SingleProcess AUTOTUNE benchmarking takes 0.3521 seconds and 0.3038 seconds precompiling for 18 choices 2025-09-09T14:29:13.5432197Z >>time: 0.006ms for , to_beat: 0.006ms 2025-09-09T14:29:13.5433128Z >>time: 0.005ms for , to_beat: 0.006ms 2025-09-09T14:29:13.5434084Z >>time: 0.007ms for matmul, to_beat: 0.005ms 2025-09-09T14:29:13.5435249Z best_cls= 2025-09-09T14:29:13.5435687Z 2025-09-09T14:29:13.5435985Z PASSED 2025-09-09T14:29:13.5436718Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_compile_15_cuda SKIPPED 2025-09-09T14:29:13.5437548Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_compile_16_cuda SKIPPED 2025-09-09T14:29:13.5438424Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_compile_17_cuda activation_shapes: torch.Size([32, 128]), times_seen: 2 2025-09-09T14:29:13.5439220Z weight_shape: torch.Size([128, 128]), dtype: torch.bfloat16, bias_shape: torch.Size([128]) 2025-09-09T14:29:13.5439674Z AUTOTUNE addmm(32x128, 32x128, 128x128) 2025-09-09T14:29:13.5440266Z triton_mm_654 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:29:13.5441219Z triton_mm_665 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:29:13.5442117Z triton_mm_655 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:29:13.5442991Z triton_mm_656 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:29:13.5443870Z triton_mm_658 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:29:13.5444758Z triton_mm_660 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:29:13.5445632Z triton_mm_661 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:29:13.5446519Z triton_mm_662 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:29:13.5447401Z triton_mm_663 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:29:13.5448284Z triton_mm_664 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:29:13.5449110Z SingleProcess AUTOTUNE benchmarking takes 0.3827 seconds and 0.3622 seconds precompiling for 19 choices 2025-09-09T14:29:13.5449847Z >>time: 0.007ms for , to_beat: infms 2025-09-09T14:29:13.5450372Z AUTOTUNE mm(32x128, 128x128) 2025-09-09T14:29:13.5450942Z triton_mm_684 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:29:13.5451842Z triton_mm_672 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:29:13.5452733Z triton_mm_673 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:29:13.5453612Z triton_mm_674 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:29:13.5454498Z triton_mm_675 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:29:13.5455467Z triton_mm_676 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:29:13.5456437Z triton_mm_677 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:29:13.5457406Z triton_mm_680 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:29:13.5458293Z triton_mm_681 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:29:13.5459171Z triton_mm_682 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:29:13.5459999Z SingleProcess AUTOTUNE benchmarking takes 0.3523 seconds and 0.3113 seconds precompiling for 18 choices 2025-09-09T14:29:13.5460870Z >>time: 0.005ms for , to_beat: 0.007ms 2025-09-09T14:29:13.5461828Z >>time: 0.005ms for , to_beat: 0.005ms 2025-09-09T14:29:13.5462793Z >>time: 0.006ms for matmul, to_beat: 0.005ms 2025-09-09T14:29:13.5463659Z best_cls= 2025-09-09T14:29:13.5464092Z 2025-09-09T14:29:13.5464222Z PASSED 2025-09-09T14:29:13.5464784Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_double_access_0_cpu SKIPPED 2025-09-09T14:29:13.5465643Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_double_access_1_cpu SKIPPED 2025-09-09T14:29:13.5466494Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_double_access_2_cpu SKIPPED 2025-09-09T14:29:13.5467391Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_double_access_3_cuda activation_shapes: torch.Size([16, 128]), times_seen: 1 2025-09-09T14:29:13.5468205Z weight_shape: torch.Size([128, 128]), dtype: torch.float32, bias_shape: torch.Size([128]) 2025-09-09T14:29:13.5468653Z AUTOTUNE addmm(16x128, 16x128, 128x128) 2025-09-09T14:29:13.5469256Z triton_mm_701 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:29:30.6020377Z triton_mm_702 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:29:30.6021492Z triton_mm_703 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:29:30.6022495Z triton_mm_704 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:29:30.6023494Z triton_mm_705 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:29:30.6024490Z triton_mm_711 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:29:30.6025457Z triton_mm_698 0.0215 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:29:30.6026466Z triton_mm_699 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:29:30.6027480Z triton_mm_700 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:29:30.6028776Z triton_mm_706 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:29:30.6029708Z SingleProcess AUTOTUNE benchmarking takes 0.3790 seconds and 1.5606 seconds precompiling for 19 choices 2025-09-09T14:29:30.6030728Z >>time: 0.008ms for , to_beat: infms 2025-09-09T14:29:30.6031333Z AUTOTUNE mm(16x128, 128x128) 2025-09-09T14:29:30.6031951Z triton_mm_716 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:29:30.6032969Z triton_mm_718 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:29:30.6033956Z triton_mm_720 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:29:30.6034947Z triton_mm_723 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:29:30.6035921Z triton_mm_728 0.0215 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:29:30.6036950Z triton_mm_726 0.0225 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:29:30.6037940Z triton_mm_715 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:29:30.6038951Z triton_mm_717 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:29:30.6039963Z triton_mm_719 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:29:30.6040930Z triton_mm_721 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:29:30.6041841Z SingleProcess AUTOTUNE benchmarking takes 0.3434 seconds and 0.7262 seconds precompiling for 18 choices 2025-09-09T14:29:30.6042765Z >>time: 0.007ms for , to_beat: 0.008ms 2025-09-09T14:29:30.6043775Z >>time: 0.004ms for , to_beat: 0.007ms 2025-09-09T14:29:30.6044476Z AUTOTUNE int_mm(16x128, 128x128) 2025-09-09T14:29:30.6045099Z triton_mm_739 0.0195 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=8 2025-09-09T14:29:30.6046083Z triton_mm_740 0.0195 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:29:30.6047034Z triton_mm_734 0.0205 ms 95.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:29:30.6047982Z triton_mm_736 0.0205 ms 95.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:29:30.6048928Z triton_mm_738 0.0205 ms 95.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:29:30.6049867Z triton_mm_733 0.0215 ms 90.5% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:29:30.6052744Z triton_mm_735 0.0215 ms 90.5% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:29:30.6053763Z triton_mm_741 0.0215 ms 90.5% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:29:30.6054857Z triton_mm_732 0.0225 ms 86.4% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:29:30.6055805Z triton_mm_737 0.0225 ms 86.4% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:29:30.6056814Z SingleProcess AUTOTUNE benchmarking takes 0.2053 seconds and 0.2221 seconds precompiling for 11 choices 2025-09-09T14:29:30.6057772Z >>time: 0.006ms for matmul, to_beat: 0.004ms 2025-09-09T14:29:30.6058738Z best_cls= 2025-09-09T14:29:30.6059249Z 2025-09-09T14:29:30.6059572Z PASSED 2025-09-09T14:29:30.6060274Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_double_access_4_cuda activation_shapes: torch.Size([16, 128]), times_seen: 1 2025-09-09T14:29:30.6061124Z weight_shape: torch.Size([128, 128]), dtype: torch.float16, bias_shape: torch.Size([128]) 2025-09-09T14:29:30.6061579Z AUTOTUNE addmm(16x128, 16x128, 128x128) 2025-09-09T14:29:30.6062256Z triton_mm_749 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:29:30.6063161Z triton_mm_746 0.0205 ms 95.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:29:30.6064131Z triton_mm_751 0.0205 ms 95.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:29:30.6065033Z triton_mm_756 0.0205 ms 95.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:29:30.6065914Z triton_mm_758 0.0205 ms 95.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:29:30.6066858Z triton_mm_742 0.0215 ms 90.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:29:30.6067763Z triton_mm_743 0.0215 ms 90.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:29:30.6068721Z triton_mm_744 0.0215 ms 90.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:29:30.6069609Z triton_mm_745 0.0215 ms 90.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:29:30.6070515Z triton_mm_747 0.0215 ms 90.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:29:30.6071332Z SingleProcess AUTOTUNE benchmarking takes 0.3792 seconds and 0.2800 seconds precompiling for 19 choices 2025-09-09T14:29:30.6072080Z >>time: 0.005ms for , to_beat: infms 2025-09-09T14:29:30.6072593Z AUTOTUNE mm(16x128, 128x128) 2025-09-09T14:29:30.6073172Z triton_mm_771 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:29:30.6074087Z triton_mm_773 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:29:30.6075155Z triton_mm_759 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:29:54.3038873Z triton_mm_760 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:29:54.3040152Z triton_mm_761 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:29:54.3041041Z triton_mm_762 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:29:54.3042213Z triton_mm_763 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:29:54.3043463Z triton_mm_764 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:29:54.3044601Z triton_mm_765 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:29:54.3045493Z triton_mm_766 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:29:54.3046307Z SingleProcess AUTOTUNE benchmarking takes 0.3485 seconds and 0.2684 seconds precompiling for 18 choices 2025-09-09T14:29:54.3047154Z >>time: 0.006ms for , to_beat: 0.005ms 2025-09-09T14:29:54.3048083Z >>time: 0.004ms for , to_beat: 0.005ms 2025-09-09T14:29:54.3049059Z >>time: 0.006ms for matmul, to_beat: 0.004ms 2025-09-09T14:29:54.3049937Z best_cls= 2025-09-09T14:29:54.3050364Z 2025-09-09T14:29:54.3050657Z PASSED 2025-09-09T14:29:54.3051287Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_double_access_5_cuda activation_shapes: torch.Size([16, 128]), times_seen: 1 2025-09-09T14:29:54.3052112Z weight_shape: torch.Size([128, 128]), dtype: torch.bfloat16, bias_shape: torch.Size([128]) 2025-09-09T14:29:54.3052560Z AUTOTUNE addmm(16x128, 16x128, 128x128) 2025-09-09T14:29:54.3053159Z triton_mm_800 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:29:54.3054058Z triton_mm_801 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=8 2025-09-09T14:29:54.3054965Z triton_mm_786 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:29:54.3055847Z triton_mm_787 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:29:54.3056817Z triton_mm_788 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:29:54.3057703Z triton_mm_789 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:29:54.3058588Z triton_mm_790 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:29:54.3059472Z triton_mm_791 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:29:54.3060589Z triton_mm_793 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:29:54.3061558Z triton_mm_794 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:29:54.3062377Z SingleProcess AUTOTUNE benchmarking takes 0.3821 seconds and 0.2667 seconds precompiling for 19 choices 2025-09-09T14:29:54.3063116Z >>time: 0.005ms for , to_beat: infms 2025-09-09T14:29:54.3063625Z AUTOTUNE mm(16x128, 128x128) 2025-09-09T14:29:54.3064258Z triton_mm_812 0.0184 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:29:54.3065151Z triton_mm_807 0.0185 ms 99.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:29:54.3066036Z triton_mm_811 0.0195 ms 94.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:29:54.3066931Z triton_mm_815 0.0195 ms 94.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:29:54.3067817Z triton_mm_803 0.0205 ms 90.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:29:54.3068699Z triton_mm_804 0.0205 ms 90.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:29:54.3069580Z triton_mm_805 0.0205 ms 90.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:29:54.3070468Z triton_mm_806 0.0205 ms 90.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:29:54.3071352Z triton_mm_808 0.0205 ms 90.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:29:54.3072245Z triton_mm_809 0.0205 ms 90.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:29:54.3073061Z SingleProcess AUTOTUNE benchmarking takes 0.3510 seconds and 0.2491 seconds precompiling for 18 choices 2025-09-09T14:29:54.3073903Z >>time: 0.006ms for , to_beat: 0.005ms 2025-09-09T14:29:54.3074880Z >>time: 0.004ms for , to_beat: 0.005ms 2025-09-09T14:29:54.3075849Z >>time: 0.006ms for matmul, to_beat: 0.004ms 2025-09-09T14:29:54.3076719Z best_cls= 2025-09-09T14:29:54.3077155Z 2025-09-09T14:29:54.3077283Z PASSED 2025-09-09T14:29:54.3077819Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_float8_0_cpu SKIPPED 2025-09-09T14:29:54.3078619Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_float8_1_cpu SKIPPED 2025-09-09T14:29:54.3079420Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_float8_2_cpu SKIPPED 2025-09-09T14:29:54.3080219Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_float8_3_cuda SKIPPED 2025-09-09T14:29:54.3081022Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_float8_4_cuda SKIPPED 2025-09-09T14:29:54.3081820Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_float8_5_cuda SKIPPED 2025-09-09T14:29:54.3082757Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_hp_float activation_shapes: torch.Size([128, 128]), times_seen: 1 2025-09-09T14:29:54.3083620Z weight_shape: torch.Size([128, 128]), dtype: torch.float32, bias_shape: torch.Size([128]) 2025-09-09T14:29:54.3084288Z >>time: 0.008ms for , to_beat: infms 2025-09-09T14:29:54.3084937Z best_cls= 2025-09-09T14:29:54.3085274Z 2025-09-09T14:29:54.3085439Z activation_shapes: torch.Size([128, 128]), times_seen: 1 2025-09-09T14:29:54.3085924Z weight_shape: torch.Size([128, 128]), dtype: torch.float32, bias_shape: torch.Size([128]) 2025-09-09T14:29:54.3086589Z >>time: 0.012ms for , to_beat: infms 2025-09-09T14:29:54.3087240Z best_cls= 2025-09-09T14:29:54.3087584Z 2025-09-09T14:29:54.3087746Z activation_shapes: torch.Size([128, 128]), times_seen: 1 2025-09-09T14:29:54.3088230Z weight_shape: torch.Size([128, 128]), dtype: torch.float32, bias_shape: torch.Size([128]) 2025-09-09T14:29:54.3088886Z >>time: 0.011ms for , to_beat: infms 2025-09-09T14:29:54.3089540Z best_cls= 2025-09-09T14:29:54.3089873Z 2025-09-09T14:29:54.3089996Z PASSED 2025-09-09T14:29:54.3090523Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_int4wo_0_cpu SKIPPED 2025-09-09T14:29:54.3091319Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_int4wo_1_cpu SKIPPED 2025-09-09T14:29:54.3092110Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_int4wo_2_cpu SKIPPED 2025-09-09T14:29:54.3092911Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_int4wo_3_cuda SKIPPED 2025-09-09T14:29:54.3093716Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_int4wo_4_cuda SKIPPED 2025-09-09T14:30:13.5394336Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_int4wo_5_cuda SKIPPED 2025-09-09T14:30:13.5396088Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_00_cpu SKIPPED 2025-09-09T14:30:13.5397971Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_01_cpu SKIPPED 2025-09-09T14:30:13.5399567Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_02_cpu SKIPPED 2025-09-09T14:30:13.5400468Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_03_cpu SKIPPED 2025-09-09T14:30:13.5401264Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_04_cpu SKIPPED 2025-09-09T14:30:13.5402061Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_05_cpu SKIPPED 2025-09-09T14:30:13.5402861Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_06_cpu SKIPPED 2025-09-09T14:30:13.5403666Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_07_cpu SKIPPED 2025-09-09T14:30:13.5404461Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_08_cpu SKIPPED 2025-09-09T14:30:13.5405262Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_09_cuda SKIPPED 2025-09-09T14:30:13.5406065Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_10_cuda SKIPPED 2025-09-09T14:30:13.5406861Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_11_cuda PASSED 2025-09-09T14:30:13.5407668Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_12_cuda SKIPPED 2025-09-09T14:30:13.5408469Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_13_cuda SKIPPED 2025-09-09T14:30:13.5409582Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_14_cuda PASSED 2025-09-09T14:30:13.5410438Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_15_cuda SKIPPED 2025-09-09T14:30:13.5411502Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_16_cuda SKIPPED 2025-09-09T14:30:13.5412299Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_17_cuda PASSED 2025-09-09T14:30:13.5413083Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_manual_0_cpu SKIPPED 2025-09-09T14:30:13.5413874Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_manual_1_cpu SKIPPED 2025-09-09T14:30:13.5414657Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_manual_2_cpu SKIPPED 2025-09-09T14:30:13.5415443Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_manual_3_cuda PASSED 2025-09-09T14:30:13.5416293Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_manual_4_cuda PASSED 2025-09-09T14:30:13.5417075Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_manual_5_cuda PASSED 2025-09-09T14:30:13.5417861Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_mha_0_cpu SKIPPED 2025-09-09T14:30:13.5418615Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_mha_1_cpu SKIPPED 2025-09-09T14:30:13.5419376Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_mha_2_cpu SKIPPED 2025-09-09T14:30:13.5420205Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_mha_3_cuda activation_shapes: torch.Size([1, 4096]), times_seen: 1 2025-09-09T14:30:13.5421019Z weight_shape: torch.Size([4096, 4096]), dtype: torch.float32, bias_shape: torch.Size([4096]) 2025-09-09T14:30:13.5421486Z AUTOTUNE addmm(1x4096, 1x4096, 4096x4096) 2025-09-09T14:30:13.5421772Z bias_addmm 0.1434 ms 100.0% 2025-09-09T14:30:13.5422025Z addmm 0.1495 ms 95.9% 2025-09-09T14:30:13.5422573Z triton_mm_837 0.2017 ms 71.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:30:13.5423460Z triton_mm_833 0.2171 ms 66.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:30:13.5424331Z triton_mm_844 0.2171 ms 66.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:30:13.5425195Z triton_mm_832 0.2181 ms 65.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:30:13.5426076Z triton_mm_843 0.2202 ms 65.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:30:13.5426973Z triton_mm_831 0.2540 ms 56.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:30:13.5427851Z triton_mm_834 0.2642 ms 54.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:30:13.5428729Z triton_mm_836 0.2796 ms 51.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:30:13.5429543Z SingleProcess AUTOTUNE benchmarking takes 0.6537 seconds and 1.2016 seconds precompiling for 19 choices 2025-09-09T14:30:13.5430322Z >>time: 0.144ms for , to_beat: infms 2025-09-09T14:30:13.5431155Z >>time: 0.039ms for , to_beat: 0.144ms 2025-09-09T14:30:13.5432210Z >>time: 0.038ms for , to_beat: 0.039ms 2025-09-09T14:30:13.5432827Z AUTOTUNE int_mm(1x4096, 4096x4096) 2025-09-09T14:30:13.5433408Z triton_mm_857 0.0471 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:30:13.5434391Z triton_mm_856 0.0481 ms 97.9% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=256, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:30:13.5435263Z triton_mm_855 0.0532 ms 88.5% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:30:13.5436110Z triton_mm_852 0.0543 ms 86.8% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:30:13.5436980Z triton_mm_853 0.0563 ms 83.6% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:30:13.5437827Z triton_mm_851 0.0614 ms 76.7% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:30:13.5438674Z triton_mm_849 0.0758 ms 62.2% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:30:13.5439538Z triton_mm_848 0.0798 ms 59.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:30:13.5440438Z triton_mm_847 0.1321 ms 35.7% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:30:13.5441283Z triton_mm_850 0.1331 ms 35.4% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:30:13.5442090Z SingleProcess AUTOTUNE benchmarking takes 0.2667 seconds and 0.3887 seconds precompiling for 12 choices 2025-09-09T14:30:13.5442955Z >>time: 0.047ms for matmul, to_beat: 0.038ms 2025-09-09T14:30:13.5443814Z best_cls= 2025-09-09T14:30:13.5444239Z 2025-09-09T14:30:13.5444370Z PASSED 2025-09-09T14:30:13.5444932Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_mha_4_cuda activation_shapes: torch.Size([1, 4096]), times_seen: 1 2025-09-09T14:30:13.5445718Z weight_shape: torch.Size([4096, 4096]), dtype: torch.float16, bias_shape: torch.Size([4096]) 2025-09-09T14:30:13.5446173Z AUTOTUNE addmm(1x4096, 1x4096, 4096x4096) 2025-09-09T14:30:13.5446782Z triton_mm_866 0.0809 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:30:13.5447682Z triton_mm_874 0.0840 ms 96.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:30:13.5448578Z triton_mm_869 0.0860 ms 94.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:30:13.5449456Z triton_mm_865 0.0870 ms 92.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:30:13.5450040Z addmm 0.0881 ms 91.9% 2025-09-09T14:30:13.5450624Z triton_mm_860 0.0881 ms 91.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:30:13.5451508Z triton_mm_862 0.0891 ms 90.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:30:13.5452466Z triton_mm_872 0.0942 ms 85.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:30:13.5453049Z bias_addmm 0.0983 ms 82.3% 2025-09-09T14:30:43.9561513Z triton_mm_868 0.0983 ms 82.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:30:43.9563843Z SingleProcess AUTOTUNE benchmarking takes 0.5025 seconds and 0.4135 seconds precompiling for 19 choices 2025-09-09T14:30:43.9564621Z >>time: 0.081ms for , to_beat: infms 2025-09-09T14:30:43.9565464Z >>time: 0.039ms for , to_beat: 0.081ms 2025-09-09T14:30:43.9566396Z >>time: 0.039ms for , to_beat: 0.039ms 2025-09-09T14:30:43.9567393Z >>time: 0.048ms for matmul, to_beat: 0.039ms 2025-09-09T14:30:43.9568271Z best_cls= 2025-09-09T14:30:43.9568715Z 2025-09-09T14:30:43.9569022Z PASSED 2025-09-09T14:30:43.9569841Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_mha_5_cuda activation_shapes: torch.Size([1, 4096]), times_seen: 1 2025-09-09T14:30:43.9570885Z weight_shape: torch.Size([4096, 4096]), dtype: torch.bfloat16, bias_shape: torch.Size([4096]) 2025-09-09T14:30:43.9571367Z AUTOTUNE addmm(1x4096, 1x4096, 4096x4096) 2025-09-09T14:30:43.9571994Z triton_mm_894 0.0809 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:30:43.9572905Z triton_mm_902 0.0829 ms 97.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:30:43.9573820Z triton_mm_897 0.0850 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:30:43.9574713Z triton_mm_893 0.0860 ms 94.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:30:43.9575615Z triton_mm_888 0.0870 ms 92.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:30:43.9576312Z addmm 0.0881 ms 91.9% 2025-09-09T14:30:43.9576865Z triton_mm_890 0.0891 ms 90.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:30:43.9577755Z triton_mm_900 0.0922 ms 87.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:30:43.9578653Z triton_mm_896 0.0963 ms 84.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:30:43.9579236Z bias_addmm 0.0973 ms 83.2% 2025-09-09T14:30:43.9579744Z SingleProcess AUTOTUNE benchmarking takes 0.4988 seconds and 0.3984 seconds precompiling for 19 choices 2025-09-09T14:30:43.9580549Z >>time: 0.081ms for , to_beat: infms 2025-09-09T14:30:43.9581391Z >>time: 0.039ms for , to_beat: 0.081ms 2025-09-09T14:30:43.9582341Z >>time: 0.039ms for , to_beat: 0.039ms 2025-09-09T14:30:43.9583406Z >>time: 0.047ms for matmul, to_beat: 0.039ms 2025-09-09T14:30:43.9584271Z best_cls= 2025-09-09T14:30:43.9592288Z 2025-09-09T14:30:43.9592468Z PASSED 2025-09-09T14:30:43.9593097Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_min_sqnr_0_cuda activation_shapes: torch.Size([128, 128]), times_seen: 1 2025-09-09T14:30:43.9594076Z weight_shape: torch.Size([128, 128]), dtype: torch.float32, bias_shape: torch.Size([128]) 2025-09-09T14:30:43.9594528Z AUTOTUNE addmm(128x128, 128x128, 128x128) 2025-09-09T14:30:43.9595153Z triton_mm_916 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:30:43.9596052Z triton_mm_921 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:30:43.9596940Z triton_mm_924 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:30:43.9598279Z triton_mm_925 0.0216 ms 99.7% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:30:43.9599178Z triton_mm_914 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:30:43.9600107Z triton_mm_915 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:30:43.9600993Z triton_mm_917 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:30:43.9601872Z triton_mm_918 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=128, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:30:43.9602763Z triton_mm_919 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:30:43.9603652Z triton_mm_920 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:30:43.9604480Z SingleProcess AUTOTUNE benchmarking takes 0.3950 seconds and 0.8820 seconds precompiling for 21 choices 2025-09-09T14:30:43.9605239Z >>time: 0.006ms for , to_beat: infms 2025-09-09T14:30:43.9606288Z skipping q_cls: because the sqnr is too small, minimum expected sqnr: 60, got 53.82148742675781 2025-09-09T14:30:43.9607662Z skipping q_cls: because the sqnr is too small, minimum expected sqnr: 60, got 54.006710052490234 2025-09-09T14:30:43.9609061Z skipping q_cls: because the sqnr is too small, minimum expected sqnr: 60, got 47.44115447998047 2025-09-09T14:30:43.9610184Z best_cls= 2025-09-09T14:30:43.9610549Z 2025-09-09T14:30:43.9610686Z PASSED 2025-09-09T14:30:43.9611286Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_min_sqnr_1_cuda activation_shapes: torch.Size([128, 128]), times_seen: 1 2025-09-09T14:30:43.9612076Z weight_shape: torch.Size([128, 128]), dtype: torch.float16, bias_shape: torch.Size([128]) 2025-09-09T14:30:43.9612536Z AUTOTUNE addmm(128x128, 128x128, 128x128) 2025-09-09T14:30:43.9613136Z triton_mm_941 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=128, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:30:43.9614034Z triton_mm_933 0.0215 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:30:43.9615065Z triton_mm_934 0.0215 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:30:43.9616159Z triton_mm_935 0.0215 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:30:43.9617040Z triton_mm_936 0.0215 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:30:43.9617929Z triton_mm_937 0.0215 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=128, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:30:43.9618814Z triton_mm_938 0.0215 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:30:43.9619697Z triton_mm_939 0.0215 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:30:43.9620577Z triton_mm_940 0.0215 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:30:43.9621458Z triton_mm_942 0.0215 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:30:43.9622280Z SingleProcess AUTOTUNE benchmarking takes 0.4154 seconds and 0.6175 seconds precompiling for 21 choices 2025-09-09T14:30:43.9623014Z >>time: 0.006ms for , to_beat: infms 2025-09-09T14:30:43.9624032Z skipping q_cls: because the sqnr is too small, minimum expected sqnr: 60, got 54.25 2025-09-09T14:30:43.9625325Z skipping q_cls: because the sqnr is too small, minimum expected sqnr: 60, got 54.40625 2025-09-09T14:31:05.6626160Z skipping q_cls: because the sqnr is too small, minimum expected sqnr: 60, got 48.0625 2025-09-09T14:31:05.6627158Z best_cls= 2025-09-09T14:31:05.6627495Z 2025-09-09T14:31:05.6627787Z PASSED 2025-09-09T14:31:05.6628397Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_min_sqnr_2_cuda activation_shapes: torch.Size([128, 128]), times_seen: 1 2025-09-09T14:31:05.6629195Z weight_shape: torch.Size([128, 128]), dtype: torch.bfloat16, bias_shape: torch.Size([128]) 2025-09-09T14:31:05.6629641Z AUTOTUNE addmm(128x128, 128x128, 128x128) 2025-09-09T14:31:05.6630247Z triton_mm_963 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:31:05.6631148Z triton_mm_969 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=128, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:31:05.6632039Z triton_mm_958 0.0195 ms 99.7% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:31:05.6632902Z triton_mm_954 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:31:05.6633766Z triton_mm_968 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=128, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:31:05.6634642Z triton_mm_956 0.0205 ms 94.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=128, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:31:05.6635814Z triton_mm_959 0.0214 ms 90.7% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:31:05.6636708Z triton_mm_952 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:31:05.6638745Z triton_mm_953 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:31:05.6639612Z triton_mm_955 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:31:05.6640412Z SingleProcess AUTOTUNE benchmarking takes 0.4137 seconds and 0.6166 seconds precompiling for 21 choices 2025-09-09T14:31:05.6641151Z >>time: 0.008ms for , to_beat: infms 2025-09-09T14:31:05.6642153Z skipping q_cls: because the sqnr is too small, minimum expected sqnr: 60, got 49.0 2025-09-09T14:31:05.6643429Z skipping q_cls: because the sqnr is too small, minimum expected sqnr: 60, got 48.75 2025-09-09T14:31:05.6644710Z skipping q_cls: because the sqnr is too small, minimum expected sqnr: 60, got 46.0 2025-09-09T14:31:05.6645628Z best_cls= 2025-09-09T14:31:05.6645973Z 2025-09-09T14:31:05.6646103Z PASSED 2025-09-09T14:31:05.6646583Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_00_cpu (m, k, n): (16, 128, 128) 2025-09-09T14:31:05.6647204Z SKIPPED 2025-09-09T14:31:05.6647689Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_01_cpu (m, k, n): (64, 128, 128) 2025-09-09T14:31:05.6648243Z SKIPPED 2025-09-09T14:31:05.6648724Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_02_cpu (m, k, n): (16, 128, 256) 2025-09-09T14:31:05.6649275Z SKIPPED 2025-09-09T14:31:05.6649758Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_03_cpu (m, k, n): (16, 256, 128) 2025-09-09T14:31:05.6650310Z SKIPPED 2025-09-09T14:31:05.6650783Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_04_cpu (m, k, n): (64, 256, 128) 2025-09-09T14:31:05.6651340Z SKIPPED 2025-09-09T14:31:05.6651806Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_05_cpu (m, k, n): (16, 128, 128) 2025-09-09T14:31:05.6652369Z SKIPPED 2025-09-09T14:31:05.6652837Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_06_cpu (m, k, n): (64, 128, 128) 2025-09-09T14:31:05.6653389Z SKIPPED 2025-09-09T14:31:05.6653857Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_07_cpu (m, k, n): (16, 128, 256) 2025-09-09T14:31:05.6654415Z SKIPPED 2025-09-09T14:31:05.6654887Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_08_cpu (m, k, n): (16, 256, 128) 2025-09-09T14:31:05.6655444Z SKIPPED 2025-09-09T14:31:05.6655989Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_09_cpu (m, k, n): (64, 256, 128) 2025-09-09T14:31:05.6656540Z SKIPPED 2025-09-09T14:31:05.6657013Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_10_cpu (m, k, n): (16, 128, 128) 2025-09-09T14:31:05.6657560Z SKIPPED 2025-09-09T14:31:05.6658037Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_11_cpu (m, k, n): (64, 128, 128) 2025-09-09T14:31:05.6658596Z SKIPPED 2025-09-09T14:31:05.6659058Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_12_cpu (m, k, n): (16, 128, 256) 2025-09-09T14:31:05.6659613Z SKIPPED 2025-09-09T14:31:05.6660174Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_13_cpu (m, k, n): (16, 256, 128) 2025-09-09T14:31:05.6660809Z SKIPPED 2025-09-09T14:31:05.6661273Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_14_cpu (m, k, n): (64, 256, 128) 2025-09-09T14:31:05.6661829Z SKIPPED 2025-09-09T14:31:05.6662303Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_15_cuda (m, k, n): (16, 128, 128) 2025-09-09T14:31:05.6662854Z PASSED 2025-09-09T14:31:05.6663327Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_16_cuda (m, k, n): (64, 128, 128) 2025-09-09T14:31:05.6663920Z activation_shapes: torch.Size([64, 128]), times_seen: 1 2025-09-09T14:31:05.6664403Z weight_shape: torch.Size([128, 128]), dtype: torch.float32, bias_shape: torch.Size([128]) 2025-09-09T14:31:05.6664838Z AUTOTUNE addmm(64x128, 64x128, 128x128) 2025-09-09T14:31:05.6665459Z triton_mm_971 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:31:05.6666361Z triton_mm_972 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:31:05.6667293Z triton_mm_974 0.0236 ms 91.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:31:05.6668166Z triton_mm_973 0.0246 ms 87.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:31:05.6669039Z triton_mm_984 0.0246 ms 87.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:31:05.6669906Z triton_mm_977 0.0256 ms 84.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:31:05.6670779Z triton_mm_976 0.0276 ms 77.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:31:05.6671664Z triton_mm_980 0.0297 ms 72.4% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:31:05.6672542Z triton_mm_978 0.0338 ms 63.6% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:31:05.6673423Z triton_mm_982 0.0338 ms 63.6% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:31:05.6674232Z SingleProcess AUTOTUNE benchmarking takes 0.4202 seconds and 9.9760 seconds precompiling for 19 choices 2025-09-09T14:31:05.6674983Z >>time: 0.014ms for , to_beat: infms 2025-09-09T14:31:05.6675501Z AUTOTUNE mm(64x128, 128x128) 2025-09-09T14:31:05.6676061Z triton_mm_989 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:31:05.6677015Z triton_mm_992 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:31:05.6677893Z triton_mm_988 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:31:05.6678777Z triton_mm_990 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:31:05.6679739Z triton_mm_991 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:31:05.6680607Z triton_mm_993 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:31:31.3758222Z triton_mm_994 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:31:31.3761300Z triton_mm_995 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:31:31.3762250Z triton_mm_997 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:31:31.3763142Z triton_mm_998 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:31:31.3763994Z SingleProcess AUTOTUNE benchmarking takes 0.3645 seconds and 4.6057 seconds precompiling for 18 choices 2025-09-09T14:31:31.3764836Z >>time: 0.011ms for , to_beat: 0.014ms 2025-09-09T14:31:31.3765513Z AUTOTUNE int_mm(64x128, 128x128) 2025-09-09T14:31:31.3766085Z triton_mm_1007 0.0195 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:31:31.3766966Z triton_mm_1009 0.0195 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:31:31.3767838Z triton_mm_1005 0.0205 ms 95.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:31:31.3768715Z triton_mm_1006 0.0205 ms 95.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:31:31.3769588Z triton_mm_1011 0.0205 ms 95.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:31:31.3770466Z triton_mm_1012 0.0205 ms 95.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=8 2025-09-09T14:31:31.3771343Z triton_mm_1013 0.0205 ms 95.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:31:31.3772217Z triton_mm_1014 0.0205 ms 95.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:31:31.3773100Z triton_mm_1008 0.0215 ms 90.5% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:31:31.3773969Z triton_mm_1010 0.0215 ms 90.5% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:31:31.3774805Z SingleProcess AUTOTUNE benchmarking takes 0.2142 seconds and 0.2757 seconds precompiling for 11 choices 2025-09-09T14:31:31.3775715Z >>time: 0.006ms for matmul, to_beat: 0.011ms 2025-09-09T14:31:31.3776796Z >>time: 0.009ms for , to_beat: 0.040ms 2025-09-09T14:31:31.3777821Z >>time: 0.009ms for interpolated, breakeven constant: 1.67 2025-09-09T14:31:31.3778747Z best_cls= 2025-09-09T14:31:31.3779184Z 2025-09-09T14:31:31.3779478Z PASSED 2025-09-09T14:31:31.3780375Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_17_cuda (m, k, n): (16, 128, 256) 2025-09-09T14:31:31.3780988Z activation_shapes: torch.Size([16, 128]), times_seen: 1 2025-09-09T14:31:31.3781474Z weight_shape: torch.Size([256, 128]), dtype: torch.float32, bias_shape: torch.Size([256]) 2025-09-09T14:31:31.3782084Z AUTOTUNE addmm(16x256, 16x128, 128x256) 2025-09-09T14:31:31.3782687Z triton_mm_1025 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:31:31.3783616Z triton_mm_1026 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:31:31.3784521Z triton_mm_1027 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:31:31.3785434Z triton_mm_1028 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:31:31.3786342Z triton_mm_1029 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:31:31.3787247Z triton_mm_1031 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:31:31.3788147Z triton_mm_1032 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:31:31.3789042Z triton_mm_1033 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:31:31.3789948Z triton_mm_1038 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:31:31.3790850Z triton_mm_1030 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:31:31.3791670Z SingleProcess AUTOTUNE benchmarking takes 0.3800 seconds and 1.5537 seconds precompiling for 19 choices 2025-09-09T14:31:31.3792415Z >>time: 0.011ms for , to_beat: infms 2025-09-09T14:31:31.3792934Z AUTOTUNE mm(16x128, 128x256) 2025-09-09T14:31:31.3793511Z triton_mm_1054 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:31:31.3794426Z triton_mm_1055 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:31:31.3795380Z triton_mm_1042 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:31:31.3796280Z triton_mm_1043 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:31:31.3797194Z triton_mm_1044 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:31:31.3798337Z triton_mm_1045 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:31:31.3799228Z triton_mm_1047 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:31:31.3800113Z triton_mm_1049 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:31:31.3801128Z triton_mm_1050 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:31:31.3802141Z triton_mm_1051 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:31:31.3802973Z SingleProcess AUTOTUNE benchmarking takes 0.3421 seconds and 0.7368 seconds precompiling for 18 choices 2025-09-09T14:31:31.3803816Z >>time: 0.011ms for , to_beat: 0.011ms 2025-09-09T14:31:31.3804740Z >>time: 0.005ms for , to_beat: 0.011ms 2025-09-09T14:31:31.3805414Z AUTOTUNE int_mm(16x128, 128x256) 2025-09-09T14:31:31.3805982Z triton_mm_1059 0.0205 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:31:31.3806874Z triton_mm_1060 0.0205 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:31:31.3807761Z triton_mm_1062 0.0205 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:31:31.3808633Z triton_mm_1065 0.0205 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:31:31.3809510Z triton_mm_1066 0.0205 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=8 2025-09-09T14:31:31.3810394Z triton_mm_1068 0.0205 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=256, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:31:51.4670453Z triton_mm_1069 0.0205 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:31:51.4671371Z triton_mm_1061 0.0205 ms 99.7% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:31:51.4672244Z triton_mm_1063 0.0215 ms 95.2% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:31:51.4673099Z triton_mm_1064 0.0215 ms 95.2% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:31:51.4673901Z SingleProcess AUTOTUNE benchmarking takes 0.2258 seconds and 0.2221 seconds precompiling for 12 choices 2025-09-09T14:31:51.4674769Z >>time: 0.007ms for matmul, to_beat: 0.005ms 2025-09-09T14:31:51.4675636Z best_cls= 2025-09-09T14:31:51.4676060Z 2025-09-09T14:31:51.4676371Z PASSED 2025-09-09T14:31:51.4676884Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_18_cuda (m, k, n): (16, 256, 128) 2025-09-09T14:31:51.4677497Z activation_shapes: torch.Size([16, 256]), times_seen: 1 2025-09-09T14:31:51.4677981Z weight_shape: torch.Size([128, 256]), dtype: torch.float32, bias_shape: torch.Size([128]) 2025-09-09T14:31:51.4678432Z AUTOTUNE addmm(16x128, 16x256, 256x128) 2025-09-09T14:31:51.4679032Z triton_mm_1071 0.0225 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:31:51.4679944Z triton_mm_1073 0.0225 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:31:51.4681230Z triton_mm_1074 0.0225 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:31:51.4682165Z triton_mm_1084 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:31:51.4683285Z triton_mm_1072 0.0256 ms 88.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:31:51.4684175Z triton_mm_1077 0.0256 ms 88.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:31:51.4685066Z triton_mm_1083 0.0256 ms 88.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:31:51.4685961Z triton_mm_1070 0.0266 ms 84.6% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:31:51.4686855Z triton_mm_1076 0.0287 ms 78.6% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:31:51.4687757Z triton_mm_1075 0.0328 ms 68.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:31:51.4688583Z SingleProcess AUTOTUNE benchmarking takes 0.3544 seconds and 1.1700 seconds precompiling for 19 choices 2025-09-09T14:31:51.4689321Z >>time: 0.011ms for , to_beat: infms 2025-09-09T14:31:51.4689833Z AUTOTUNE mm(16x256, 256x128) 2025-09-09T14:31:51.4690402Z triton_mm_1089 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:31:51.4691323Z triton_mm_1088 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:31:51.4692260Z triton_mm_1094 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:31:51.4693147Z triton_mm_1087 0.0225 ms 90.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:31:51.4694033Z triton_mm_1090 0.0225 ms 90.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:31:51.4694921Z triton_mm_1091 0.0225 ms 90.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:31:51.4695804Z triton_mm_1092 0.0225 ms 90.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:31:51.4696771Z triton_mm_1093 0.0225 ms 90.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:31:51.4697837Z triton_mm_1096 0.0225 ms 90.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:31:51.4698726Z triton_mm_1097 0.0225 ms 90.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:31:51.4699546Z SingleProcess AUTOTUNE benchmarking takes 0.3251 seconds and 0.8354 seconds precompiling for 18 choices 2025-09-09T14:31:51.4700378Z >>time: 0.017ms for , to_beat: 0.011ms 2025-09-09T14:31:51.4701310Z >>time: 0.005ms for , to_beat: 0.011ms 2025-09-09T14:31:51.4701922Z AUTOTUNE int_mm(16x256, 256x128) 2025-09-09T14:31:51.4703504Z triton_mm_1110 0.0195 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:31:51.4704509Z triton_mm_1105 0.0205 ms 95.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:31:51.4705379Z triton_mm_1107 0.0205 ms 95.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:31:51.4706250Z triton_mm_1108 0.0205 ms 95.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:31:51.4707122Z triton_mm_1111 0.0205 ms 95.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=8 2025-09-09T14:31:51.4707992Z triton_mm_1112 0.0205 ms 95.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:31:51.4708867Z triton_mm_1113 0.0205 ms 95.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:31:51.4709748Z triton_mm_1109 0.0205 ms 94.9% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:31:51.4710605Z triton_mm_1104 0.0215 ms 90.5% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:31:51.4711516Z triton_mm_1106 0.0215 ms 90.5% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:31:51.4712332Z SingleProcess AUTOTUNE benchmarking takes 0.2072 seconds and 0.1878 seconds precompiling for 11 choices 2025-09-09T14:31:51.4713194Z >>time: 0.006ms for matmul, to_beat: 0.005ms 2025-09-09T14:31:51.4714058Z best_cls= 2025-09-09T14:31:51.4714494Z 2025-09-09T14:31:51.4714628Z PASSED 2025-09-09T14:31:51.4715128Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_19_cuda (m, k, n): (64, 256, 128) 2025-09-09T14:31:51.4715740Z activation_shapes: torch.Size([64, 256]), times_seen: 1 2025-09-09T14:31:51.4716226Z weight_shape: torch.Size([128, 256]), dtype: torch.float32, bias_shape: torch.Size([128]) 2025-09-09T14:31:51.4716672Z AUTOTUNE addmm(64x128, 64x256, 256x128) 2025-09-09T14:31:51.4717274Z triton_mm_1115 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:31:51.4718188Z triton_mm_1114 0.0328 ms 75.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:31:51.4719083Z triton_mm_1116 0.0358 ms 68.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:31:51.4719974Z triton_mm_1117 0.0358 ms 68.6% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:31:51.4720875Z triton_mm_1127 0.0369 ms 66.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:10.5735597Z triton_mm_1119 0.0389 ms 63.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:10.5736612Z triton_mm_1120 0.0389 ms 63.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:10.5737756Z triton_mm_1123 0.0399 ms 61.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:10.5738814Z triton_mm_1121 0.0471 ms 52.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:32:10.5739705Z triton_mm_1125 0.0471 ms 52.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:10.5740537Z SingleProcess AUTOTUNE benchmarking takes 0.3532 seconds and 3.4199 seconds precompiling for 19 choices 2025-09-09T14:32:10.5741285Z >>time: 0.016ms for , to_beat: infms 2025-09-09T14:32:10.5741801Z AUTOTUNE mm(64x256, 256x128) 2025-09-09T14:32:10.5742389Z triton_mm_1132 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:10.5743301Z triton_mm_1144 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:10.5744205Z triton_mm_1133 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:32:10.5745110Z triton_mm_1134 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:32:10.5745992Z triton_mm_1138 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:32:10.5746892Z triton_mm_1145 0.0225 ms 90.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:32:10.5747832Z triton_mm_1131 0.0236 ms 87.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:32:10.5748724Z triton_mm_1137 0.0236 ms 87.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:10.5749620Z triton_mm_1136 0.0276 ms 74.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:10.5750507Z triton_mm_1140 0.0287 ms 71.4% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:10.5751332Z SingleProcess AUTOTUNE benchmarking takes 0.3213 seconds and 2.9225 seconds precompiling for 18 choices 2025-09-09T14:32:10.5752178Z >>time: 0.012ms for , to_beat: 0.016ms 2025-09-09T14:32:10.5752789Z AUTOTUNE int_mm(64x256, 256x128) 2025-09-09T14:32:10.5753367Z triton_mm_1149 0.0205 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:10.5754257Z triton_mm_1150 0.0205 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:10.5755132Z triton_mm_1152 0.0205 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:32:10.5756012Z triton_mm_1153 0.0205 ms 99.8% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:32:10.5756883Z triton_mm_1154 0.0205 ms 99.7% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:32:10.5758600Z triton_mm_1157 0.0214 ms 95.5% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:32:10.5759558Z triton_mm_1148 0.0215 ms 95.2% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:10.5760423Z triton_mm_1151 0.0215 ms 95.2% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:32:10.5761301Z triton_mm_1155 0.0215 ms 95.2% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=8 2025-09-09T14:32:10.5762164Z triton_mm_1156 0.0215 ms 95.2% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:32:10.5762985Z SingleProcess AUTOTUNE benchmarking takes 0.2151 seconds and 0.2599 seconds precompiling for 11 choices 2025-09-09T14:32:10.5763862Z >>time: 0.009ms for matmul, to_beat: 0.012ms 2025-09-09T14:32:10.5764829Z >>time: 0.012ms for , to_beat: 0.030ms 2025-09-09T14:32:10.5765857Z >>time: 0.012ms for interpolated, breakeven constant: 1.00 2025-09-09T14:32:10.5766799Z best_cls= 2025-09-09T14:32:10.5767229Z 2025-09-09T14:32:10.5767521Z PASSED 2025-09-09T14:32:10.5768038Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_20_cuda (m, k, n): (16, 128, 128) 2025-09-09T14:32:10.5768606Z PASSED 2025-09-09T14:32:10.5769092Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_21_cuda (m, k, n): (64, 128, 128) 2025-09-09T14:32:10.5769703Z activation_shapes: torch.Size([64, 128]), times_seen: 1 2025-09-09T14:32:10.5770198Z weight_shape: torch.Size([128, 128]), dtype: torch.float16, bias_shape: torch.Size([128]) 2025-09-09T14:32:10.5770659Z AUTOTUNE addmm(64x128, 64x128, 128x128) 2025-09-09T14:32:10.5771258Z triton_mm_1168 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:32:10.5772170Z triton_mm_1169 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:10.5773074Z triton_mm_1170 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:32:10.5773986Z triton_mm_1171 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:32:10.5774899Z triton_mm_1172 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:32:10.5775798Z triton_mm_1173 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:10.5776745Z triton_mm_1174 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:10.5777681Z triton_mm_1175 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:32:10.5778594Z triton_mm_1176 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:32:10.5779591Z triton_mm_1177 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:10.5780432Z SingleProcess AUTOTUNE benchmarking takes 0.3845 seconds and 0.4617 seconds precompiling for 19 choices 2025-09-09T14:32:10.5781255Z >>time: 0.007ms for , to_beat: infms 2025-09-09T14:32:10.5781774Z AUTOTUNE mm(64x128, 128x128) 2025-09-09T14:32:10.5782351Z triton_mm_1191 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:10.5783261Z triton_mm_1194 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:10.5784167Z triton_mm_1196 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:10.5785073Z triton_mm_1201 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:32:10.5785984Z triton_mm_1185 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:32:34.2599480Z triton_mm_1186 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:34.2601868Z triton_mm_1187 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:32:34.2602797Z triton_mm_1188 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:32:34.2603730Z triton_mm_1189 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:32:34.2604619Z triton_mm_1190 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:34.2605447Z SingleProcess AUTOTUNE benchmarking takes 0.3515 seconds and 0.4096 seconds precompiling for 18 choices 2025-09-09T14:32:34.2606287Z >>time: 0.008ms for , to_beat: 0.007ms 2025-09-09T14:32:34.2607236Z >>time: 0.006ms for matmul, to_beat: 0.007ms 2025-09-09T14:32:34.2608195Z >>time: 0.009ms for , to_beat: 0.013ms 2025-09-09T14:32:34.2609213Z >>time: 0.009ms for interpolated, breakeven constant: 0.33 2025-09-09T14:32:34.2610048Z best_cls= 2025-09-09T14:32:34.2610393Z 2025-09-09T14:32:34.2610690Z PASSED 2025-09-09T14:32:34.2611191Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_22_cuda (m, k, n): (16, 128, 256) 2025-09-09T14:32:34.2611800Z activation_shapes: torch.Size([16, 128]), times_seen: 1 2025-09-09T14:32:34.2612284Z weight_shape: torch.Size([256, 128]), dtype: torch.float16, bias_shape: torch.Size([256]) 2025-09-09T14:32:34.2612719Z AUTOTUNE addmm(16x256, 16x128, 128x256) 2025-09-09T14:32:34.2613318Z triton_mm_1227 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:34.2614210Z triton_mm_1223 0.0225 ms 90.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:32:34.2615410Z triton_mm_1225 0.0225 ms 90.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:32:34.2617982Z triton_mm_1226 0.0225 ms 90.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:32:34.2618865Z triton_mm_1230 0.0225 ms 90.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:32:34.2619763Z triton_mm_1231 0.0225 ms 90.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:34.2620659Z triton_mm_1232 0.0225 ms 90.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:32:34.2621547Z triton_mm_1235 0.0225 ms 90.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:34.2622438Z triton_mm_1238 0.0225 ms 90.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:32:34.2623331Z triton_mm_1228 0.0226 ms 90.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:34.2624143Z SingleProcess AUTOTUNE benchmarking takes 0.3814 seconds and 0.2952 seconds precompiling for 19 choices 2025-09-09T14:32:34.2624910Z >>time: 0.008ms for , to_beat: infms 2025-09-09T14:32:34.2625490Z AUTOTUNE mm(16x128, 128x256) 2025-09-09T14:32:34.2626066Z triton_mm_1239 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:32:34.2626980Z triton_mm_1240 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:32:34.2635916Z triton_mm_1244 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:34.2636849Z triton_mm_1245 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:34.2637748Z triton_mm_1246 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:34.2638658Z triton_mm_1247 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:32:34.2639565Z triton_mm_1249 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:32:34.2640474Z triton_mm_1250 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:34.2641378Z triton_mm_1251 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:32:34.2642279Z triton_mm_1252 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:34.2643104Z SingleProcess AUTOTUNE benchmarking takes 0.3518 seconds and 0.2359 seconds precompiling for 18 choices 2025-09-09T14:32:34.2643934Z >>time: 0.008ms for , to_beat: 0.008ms 2025-09-09T14:32:34.2644979Z >>time: 0.004ms for , to_beat: 0.008ms 2025-09-09T14:32:34.2645939Z >>time: 0.006ms for matmul, to_beat: 0.004ms 2025-09-09T14:32:34.2646874Z best_cls= 2025-09-09T14:32:34.2647310Z 2025-09-09T14:32:34.2647492Z PASSED 2025-09-09T14:32:34.2647984Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_23_cuda (m, k, n): (16, 256, 128) 2025-09-09T14:32:34.2648588Z activation_shapes: torch.Size([16, 256]), times_seen: 1 2025-09-09T14:32:34.2649062Z weight_shape: torch.Size([128, 256]), dtype: torch.float16, bias_shape: torch.Size([128]) 2025-09-09T14:32:34.2649507Z AUTOTUNE addmm(16x128, 16x256, 256x128) 2025-09-09T14:32:34.2650111Z triton_mm_1268 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:32:34.2651018Z triton_mm_1270 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:32:34.2651925Z triton_mm_1280 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:34.2652813Z triton_mm_1267 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:32:34.2653695Z triton_mm_1269 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:32:34.2654589Z triton_mm_1271 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:32:34.2655532Z triton_mm_1272 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:34.2656498Z triton_mm_1273 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:34.2657392Z triton_mm_1274 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:34.2658279Z triton_mm_1275 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:32:34.2659100Z SingleProcess AUTOTUNE benchmarking takes 0.3716 seconds and 0.3698 seconds precompiling for 19 choices 2025-09-09T14:32:34.2659842Z >>time: 0.005ms for , to_beat: infms 2025-09-09T14:32:34.2660347Z AUTOTUNE mm(16x256, 256x128) 2025-09-09T14:32:34.2660933Z triton_mm_1285 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:32:50.8038566Z triton_mm_1286 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:32:50.8039564Z triton_mm_1287 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:32:50.8040466Z triton_mm_1290 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:50.8041364Z triton_mm_1291 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:50.8043432Z triton_mm_1292 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:32:50.8044344Z triton_mm_1293 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:50.8045527Z triton_mm_1294 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:32:50.8046558Z triton_mm_1295 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:50.8047590Z triton_mm_1298 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:32:50.8048509Z SingleProcess AUTOTUNE benchmarking takes 0.3431 seconds and 0.3113 seconds precompiling for 18 choices 2025-09-09T14:32:50.8049445Z >>time: 0.007ms for , to_beat: 0.005ms 2025-09-09T14:32:50.8050490Z >>time: 0.005ms for , to_beat: 0.005ms 2025-09-09T14:32:50.8051560Z >>time: 0.006ms for matmul, to_beat: 0.005ms 2025-09-09T14:32:50.8052514Z best_cls= 2025-09-09T14:32:50.8053023Z 2025-09-09T14:32:50.8053313Z PASSED 2025-09-09T14:32:50.8053911Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_24_cuda (m, k, n): (64, 256, 128) 2025-09-09T14:32:50.8054569Z activation_shapes: torch.Size([64, 256]), times_seen: 1 2025-09-09T14:32:50.8055083Z weight_shape: torch.Size([128, 256]), dtype: torch.float16, bias_shape: torch.Size([128]) 2025-09-09T14:32:50.8055606Z AUTOTUNE addmm(64x128, 64x256, 256x128) 2025-09-09T14:32:50.8056337Z triton_mm_1315 0.0216 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:32:50.8057360Z triton_mm_1311 0.0225 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:32:50.8058340Z triton_mm_1312 0.0225 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:50.8059322Z triton_mm_1313 0.0225 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:32:50.8060299Z triton_mm_1314 0.0225 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:32:50.8061319Z triton_mm_1317 0.0225 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:50.8062295Z triton_mm_1318 0.0225 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:32:50.8063321Z triton_mm_1320 0.0225 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:50.8064328Z triton_mm_1321 0.0225 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:32:50.8065309Z triton_mm_1322 0.0225 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:50.8066217Z SingleProcess AUTOTUNE benchmarking takes 0.3656 seconds and 0.5470 seconds precompiling for 19 choices 2025-09-09T14:32:50.8067157Z >>time: 0.006ms for , to_beat: infms 2025-09-09T14:32:50.8067751Z AUTOTUNE mm(64x256, 256x128) 2025-09-09T14:32:50.8068339Z triton_mm_1328 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:32:50.8069482Z triton_mm_1329 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:50.8070494Z triton_mm_1331 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:32:50.8071478Z triton_mm_1332 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:32:50.8072482Z triton_mm_1335 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:32:50.8073454Z triton_mm_1333 0.0215 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:50.8074444Z triton_mm_1341 0.0215 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:50.8075426Z triton_mm_1334 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:50.8076440Z triton_mm_1337 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:32:50.8077397Z triton_mm_1338 0.0225 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:32:50.8078304Z SingleProcess AUTOTUNE benchmarking takes 0.3358 seconds and 0.4880 seconds precompiling for 18 choices 2025-09-09T14:32:50.8079222Z >>time: 0.013ms for , to_beat: 0.006ms 2025-09-09T14:32:50.8080200Z >>time: 0.009ms for matmul, to_beat: 0.006ms 2025-09-09T14:32:50.8081029Z best_cls= 2025-09-09T14:32:50.8081459Z 2025-09-09T14:32:50.8081611Z PASSED 2025-09-09T14:32:50.8082107Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_25_cuda (m, k, n): (16, 128, 128) 2025-09-09T14:32:50.8082673Z PASSED 2025-09-09T14:32:50.8083159Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_26_cuda (m, k, n): (64, 128, 128) 2025-09-09T14:32:50.8083760Z activation_shapes: torch.Size([64, 128]), times_seen: 1 2025-09-09T14:32:50.8084264Z weight_shape: torch.Size([128, 128]), dtype: torch.bfloat16, bias_shape: torch.Size([128]) 2025-09-09T14:32:50.8084716Z AUTOTUNE addmm(64x128, 64x128, 128x128) 2025-09-09T14:32:50.8085424Z triton_mm_1367 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:32:50.8086341Z triton_mm_1355 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:32:50.8087239Z triton_mm_1371 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:32:50.8088142Z triton_mm_1356 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:50.8089145Z triton_mm_1357 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:32:50.8090033Z triton_mm_1358 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:32:50.8091050Z triton_mm_1359 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:32:50.8091939Z triton_mm_1360 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:32:50.8092932Z triton_mm_1361 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:33:15.1752619Z triton_mm_1362 0.0215 ms 90.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=8 2025-09-09T14:33:15.1753484Z SingleProcess AUTOTUNE benchmarking takes 0.3819 seconds and 0.4434 seconds precompiling for 19 choices 2025-09-09T14:33:15.1754251Z >>time: 0.007ms for , to_beat: infms 2025-09-09T14:33:15.1754775Z AUTOTUNE mm(64x128, 128x128) 2025-09-09T14:33:15.1755359Z triton_mm_1384 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:33:15.1756274Z triton_mm_1372 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:33:15.1757186Z triton_mm_1373 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:33:15.1758103Z triton_mm_1374 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:33:15.1759019Z triton_mm_1375 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:33:15.1759901Z triton_mm_1377 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:33:15.1760788Z triton_mm_1378 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:33:15.1761675Z triton_mm_1381 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:33:15.1762571Z triton_mm_1382 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:33:15.1763469Z triton_mm_1383 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:33:15.1764299Z SingleProcess AUTOTUNE benchmarking takes 0.3534 seconds and 0.3921 seconds precompiling for 18 choices 2025-09-09T14:33:15.1765139Z >>time: 0.009ms for , to_beat: 0.007ms 2025-09-09T14:33:15.1766103Z >>time: 0.006ms for matmul, to_beat: 0.007ms 2025-09-09T14:33:15.1767053Z >>time: 0.009ms for , to_beat: 0.013ms 2025-09-09T14:33:15.1768077Z >>time: 0.009ms for interpolated, breakeven constant: 0.33 2025-09-09T14:33:15.1769277Z best_cls= 2025-09-09T14:33:15.1769620Z 2025-09-09T14:33:15.1769920Z PASSED 2025-09-09T14:33:15.1770433Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_27_cuda (m, k, n): (16, 128, 256) 2025-09-09T14:33:15.1771217Z activation_shapes: torch.Size([16, 128]), times_seen: 1 2025-09-09T14:33:15.1771713Z weight_shape: torch.Size([256, 128]), dtype: torch.bfloat16, bias_shape: torch.Size([256]) 2025-09-09T14:33:15.1772165Z AUTOTUNE addmm(16x256, 16x128, 128x256) 2025-09-09T14:33:15.1772764Z triton_mm_1416 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:33:15.1773666Z triton_mm_1409 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:33:15.1774569Z triton_mm_1410 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:33:15.1775466Z triton_mm_1411 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:33:15.1776432Z triton_mm_1412 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:33:15.1777327Z triton_mm_1413 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:33:15.1778266Z triton_mm_1414 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:33:15.1779159Z triton_mm_1417 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:33:15.1780053Z triton_mm_1419 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:33:15.1780963Z triton_mm_1420 0.0215 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:33:15.1781793Z SingleProcess AUTOTUNE benchmarking takes 0.3842 seconds and 0.2698 seconds precompiling for 19 choices 2025-09-09T14:33:15.1782532Z >>time: 0.005ms for , to_beat: infms 2025-09-09T14:33:15.1783045Z AUTOTUNE mm(16x128, 128x256) 2025-09-09T14:33:15.1783610Z triton_mm_1426 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:33:15.1784532Z triton_mm_1431 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:33:15.1785436Z triton_mm_1432 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:33:15.1786336Z triton_mm_1434 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:33:15.1787249Z triton_mm_1437 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:33:15.1788167Z triton_mm_1438 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:33:15.1789127Z triton_mm_1440 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:33:15.1790128Z triton_mm_1442 0.0195 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:33:15.1791135Z triton_mm_1427 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:33:15.1792038Z triton_mm_1428 0.0205 ms 95.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:33:15.1792869Z SingleProcess AUTOTUNE benchmarking takes 0.3494 seconds and 0.2449 seconds precompiling for 18 choices 2025-09-09T14:33:15.1793702Z >>time: 0.008ms for , to_beat: 0.005ms 2025-09-09T14:33:15.1794637Z >>time: 0.004ms for , to_beat: 0.005ms 2025-09-09T14:33:15.1795603Z >>time: 0.007ms for matmul, to_beat: 0.004ms 2025-09-09T14:33:15.1796467Z best_cls= 2025-09-09T14:33:15.1796898Z 2025-09-09T14:33:15.1797036Z PASSED 2025-09-09T14:33:15.1797700Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_28_cuda (m, k, n): (16, 256, 128) 2025-09-09T14:33:15.1798367Z activation_shapes: torch.Size([16, 256]), times_seen: 1 2025-09-09T14:33:15.1798852Z weight_shape: torch.Size([128, 256]), dtype: torch.bfloat16, bias_shape: torch.Size([128]) 2025-09-09T14:33:15.1799302Z AUTOTUNE addmm(16x128, 16x256, 256x128) 2025-09-09T14:33:15.1799910Z triton_mm_1464 0.0184 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:33:15.1800821Z triton_mm_1454 0.0195 ms 94.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:33:15.1801724Z triton_mm_1458 0.0195 ms 94.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:33:15.1802634Z triton_mm_1460 0.0195 ms 94.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:33:15.1803517Z triton_mm_1462 0.0195 ms 94.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:33:36.5816139Z triton_mm_1468 0.0195 ms 94.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:33:36.5818876Z triton_mm_1470 0.0195 ms 94.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:33:36.5819850Z triton_mm_1465 0.0195 ms 94.6% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:33:36.5820749Z triton_mm_1467 0.0195 ms 94.6% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:33:36.5821640Z triton_mm_1469 0.0195 ms 94.4% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=8 2025-09-09T14:33:36.5822473Z SingleProcess AUTOTUNE benchmarking takes 0.3796 seconds and 0.3307 seconds precompiling for 19 choices 2025-09-09T14:33:36.5823214Z >>time: 0.006ms for , to_beat: infms 2025-09-09T14:33:36.5823729Z AUTOTUNE mm(16x256, 256x128) 2025-09-09T14:33:36.5824715Z triton_mm_1474 0.0184 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=2 2025-09-09T14:33:36.5825628Z triton_mm_1477 0.0184 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:33:36.5826724Z triton_mm_1481 0.0184 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:33:36.5827619Z triton_mm_1484 0.0184 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:33:36.5828515Z triton_mm_1485 0.0184 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=4 2025-09-09T14:33:36.5829536Z triton_mm_1486 0.0184 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=8 2025-09-09T14:33:36.5830452Z triton_mm_1487 0.0184 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:33:36.5831362Z triton_mm_1480 0.0194 ms 94.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:33:36.5832349Z triton_mm_1471 0.0195 ms 94.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:33:36.5833346Z triton_mm_1472 0.0195 ms 94.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=2 2025-09-09T14:33:36.5834183Z SingleProcess AUTOTUNE benchmarking takes 0.3437 seconds and 0.3181 seconds precompiling for 18 choices 2025-09-09T14:33:36.5835199Z >>time: 0.006ms for , to_beat: 0.006ms 2025-09-09T14:33:36.5836224Z >>time: 0.005ms for , to_beat: 0.006ms 2025-09-09T14:33:36.5837184Z >>time: 0.006ms for matmul, to_beat: 0.005ms 2025-09-09T14:33:36.5838046Z best_cls= 2025-09-09T14:33:36.5838564Z 2025-09-09T14:33:36.5838858Z PASSED 2025-09-09T14:33:36.5839461Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_29_cuda (m, k, n): (64, 256, 128) 2025-09-09T14:33:36.5840071Z activation_shapes: torch.Size([64, 256]), times_seen: 1 2025-09-09T14:33:36.5840553Z weight_shape: torch.Size([128, 256]), dtype: torch.bfloat16, bias_shape: torch.Size([128]) 2025-09-09T14:33:36.5841078Z AUTOTUNE addmm(64x128, 64x256, 256x128) 2025-09-09T14:33:36.5841733Z triton_mm_1498 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=1, num_warps=2 2025-09-09T14:33:36.5842683Z triton_mm_1499 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:33:36.5843588Z triton_mm_1500 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:33:36.5844481Z triton_mm_1501 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:33:36.5845555Z triton_mm_1502 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:33:36.5846522Z triton_mm_1507 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:33:36.5847644Z triton_mm_1508 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:33:36.5848759Z triton_mm_1511 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:33:36.5849664Z triton_mm_1512 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:33:36.5850567Z triton_mm_1513 0.0215 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=8 2025-09-09T14:33:36.5851479Z SingleProcess AUTOTUNE benchmarking takes 0.3702 seconds and 0.5594 seconds precompiling for 19 choices 2025-09-09T14:33:36.5852313Z >>time: 0.011ms for , to_beat: infms 2025-09-09T14:33:36.5852831Z AUTOTUNE mm(64x256, 256x128) 2025-09-09T14:33:36.5853478Z triton_mm_1516 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:33:36.5854494Z triton_mm_1517 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:33:36.5855404Z triton_mm_1519 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=4 2025-09-09T14:33:36.5856431Z triton_mm_1521 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=4 2025-09-09T14:33:36.5857343Z triton_mm_1524 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:33:36.5858391Z triton_mm_1525 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:33:36.5859342Z triton_mm_1526 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=3, num_warps=4 2025-09-09T14:33:36.5860317Z triton_mm_1529 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, num_stages=4, num_warps=8 2025-09-09T14:33:36.5861311Z triton_mm_1530 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=2, num_warps=8 2025-09-09T14:33:36.5862213Z triton_mm_1531 0.0205 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, num_stages=5, num_warps=8 2025-09-09T14:33:36.5863046Z SingleProcess AUTOTUNE benchmarking takes 0.3359 seconds and 0.5112 seconds precompiling for 18 choices 2025-09-09T14:33:36.5863966Z >>time: 0.007ms for , to_beat: 0.011ms 2025-09-09T14:33:36.5865003Z >>time: 0.010ms for matmul, to_beat: 0.007ms 2025-09-09T14:33:36.5865857Z best_cls= 2025-09-09T14:33:36.5866365Z 2025-09-09T14:33:36.5866520Z PASSED 2025-09-09T14:33:36.5867043Z test/integration/test_integration.py::TestAOTI::test_aoti_00 SKIPPED 2025-09-09T14:33:36.5867719Z test/integration/test_integration.py::TestAOTI::test_aoti_01 SKIPPED 2025-09-09T14:33:36.5868351Z test/integration/test_integration.py::TestAOTI::test_aoti_02 SKIPPED 2025-09-09T14:33:36.5868991Z test/integration/test_integration.py::TestAOTI::test_aoti_03 SKIPPED 2025-09-09T14:33:36.5871248Z test/integration/test_integration.py::TestAOTI::test_aoti_04 SKIPPED 2025-09-09T14:33:36.5871901Z test/integration/test_integration.py::TestAOTI::test_aoti_05 SKIPPED 2025-09-09T14:33:36.5872646Z test/integration/test_integration.py::TestAOTI::test_aoti_06 SKIPPED 2025-09-09T14:33:36.5873273Z test/integration/test_integration.py::TestAOTI::test_aoti_07 SKIPPED 2025-09-09T14:33:36.5873911Z test/integration/test_integration.py::TestAOTI::test_aoti_08 SKIPPED 2025-09-09T14:33:42.1955975Z test/integration/test_integration.py::TestAOTI::test_aoti_09 SKIPPED 2025-09-09T14:33:42.1957001Z test/integration/test_integration.py::TestAOTI::test_aoti_10 SKIPPED 2025-09-09T14:33:42.1957645Z test/integration/test_integration.py::TestAOTI::test_aoti_11 SKIPPED 2025-09-09T14:33:42.1958281Z test/integration/test_integration.py::TestAOTI::test_aoti_12 SKIPPED 2025-09-09T14:33:42.1958919Z test/integration/test_integration.py::TestAOTI::test_aoti_13 SKIPPED 2025-09-09T14:33:42.1959574Z test/integration/test_integration.py::TestAOTI::test_aoti_14 SKIPPED 2025-09-09T14:33:42.1960215Z test/integration/test_integration.py::TestAOTI::test_aoti_15 SKIPPED 2025-09-09T14:33:42.1960860Z test/integration/test_integration.py::TestAOTI::test_aoti_16 SKIPPED 2025-09-09T14:33:42.1961493Z test/integration/test_integration.py::TestAOTI::test_aoti_17 SKIPPED 2025-09-09T14:33:42.1962147Z test/integration/test_integration.py::TestExport::test_export_00 PASSED 2025-09-09T14:33:42.1962810Z test/integration/test_integration.py::TestExport::test_export_01 PASSED 2025-09-09T14:33:42.1963464Z test/integration/test_integration.py::TestExport::test_export_02 PASSED 2025-09-09T14:33:42.1964111Z test/integration/test_integration.py::TestExport::test_export_03 PASSED 2025-09-09T14:33:42.1964767Z test/integration/test_integration.py::TestExport::test_export_04 PASSED 2025-09-09T14:33:42.1965419Z test/integration/test_integration.py::TestExport::test_export_05 PASSED 2025-09-09T14:33:42.1966071Z test/integration/test_integration.py::TestExport::test_export_06 PASSED 2025-09-09T14:33:42.1966762Z test/integration/test_integration.py::TestExport::test_export_07 PASSED 2025-09-09T14:33:42.1967433Z test/integration/test_integration.py::TestExport::test_export_08 PASSED 2025-09-09T14:33:42.1968090Z test/integration/test_integration.py::TestExport::test_export_09 PASSED 2025-09-09T14:33:42.1968741Z test/integration/test_integration.py::TestExport::test_export_10 PASSED 2025-09-09T14:33:42.1969402Z test/integration/test_integration.py::TestExport::test_export_11 PASSED 2025-09-09T14:33:42.1970056Z test/integration/test_integration.py::TestExport::test_export_12 PASSED 2025-09-09T14:33:42.1970720Z test/integration/test_integration.py::TestExport::test_export_13 PASSED 2025-09-09T14:33:42.1971383Z test/integration/test_integration.py::TestExport::test_export_14 PASSED 2025-09-09T14:33:42.1972037Z test/integration/test_integration.py::TestExport::test_export_15 PASSED 2025-09-09T14:33:42.1972694Z test/integration/test_integration.py::TestExport::test_export_16 PASSED 2025-09-09T14:33:42.1973344Z test/integration/test_integration.py::TestExport::test_export_17 PASSED 2025-09-09T14:33:42.1974003Z test/integration/test_integration.py::TestExport::test_export_18 PASSED 2025-09-09T14:33:42.1974655Z test/integration/test_integration.py::TestExport::test_export_19 PASSED 2025-09-09T14:33:42.1975312Z test/integration/test_integration.py::TestExport::test_export_20 PASSED 2025-09-09T14:33:42.1976078Z test/integration/test_integration.py::TestExport::test_export_21 PASSED 2025-09-09T14:33:42.1976728Z test/integration/test_integration.py::TestExport::test_export_22 PASSED 2025-09-09T14:33:42.1977427Z test/integration/test_integration.py::TestExport::test_export_23 PASSED 2025-09-09T14:33:42.1978477Z test/integration/test_integration.py::TestExport::test_export_float8 SKIPPED 2025-09-09T14:33:42.1979219Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_00 SKIPPED 2025-09-09T14:33:42.1980126Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_01 SKIPPED 2025-09-09T14:33:42.1980857Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_02 SKIPPED 2025-09-09T14:33:42.1981592Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_03 SKIPPED 2025-09-09T14:33:42.1982323Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_04 SKIPPED 2025-09-09T14:33:42.1983054Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_05 PASSED 2025-09-09T14:33:42.1983789Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_06 SKIPPED 2025-09-09T14:33:42.1984532Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_07 SKIPPED 2025-09-09T14:33:42.1985264Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_08 SKIPPED 2025-09-09T14:33:42.1985997Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_09 SKIPPED 2025-09-09T14:33:42.1986786Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_10 SKIPPED 2025-09-09T14:33:42.1987509Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_11 PASSED 2025-09-09T14:33:42.1988245Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_12 SKIPPED 2025-09-09T14:33:42.1988980Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_13 SKIPPED 2025-09-09T14:33:42.1989704Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_14 SKIPPED 2025-09-09T14:33:42.1990441Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_15 SKIPPED 2025-09-09T14:33:42.1991170Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_16 SKIPPED 2025-09-09T14:33:42.1991904Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_17 PASSED 2025-09-09T14:33:42.1992693Z test/integration/test_integration.py::TestBenchmarkModel::test_benchmark_model_cpu PASSED 2025-09-09T14:33:42.1993514Z test/integration/test_integration.py::TestBenchmarkModel::test_benchmark_model_cuda PASSED 2025-09-09T14:33:42.1994473Z test/integration/test_load_and_run_checkpoint.py::TestLoadAndRunCheckpoint::test_deprecated_hf_models_model_info0 SKIPPED 2025-09-09T14:33:42.1995978Z test/integration/test_load_and_run_checkpoint.py::TestLoadAndRunCheckpoint::test_deprecated_single_linear_model_name_torchao-testing/single-linear-Float8DynamicActivationFloat8WeightConfig-v1-0_13_dev SKIPPED 2025-09-09T14:33:42.1998209Z test/integration/test_load_and_run_checkpoint.py::TestLoadAndRunCheckpoint::test_single_linear_model_name_torchao-testing/single-linear-Float8DynamicActivationFloat8WeightConfig-v2-0_13_dev SKIPPED 2025-09-09T14:33:42.2000110Z test/integration/test_load_and_run_checkpoint.py::TestLoadAndRunCheckpoint::test_single_linear_model_name_torchao-testing/single-linear-Int4WeightOnlyConfig-preshuffled-v2-0_13_dev SKIPPED 2025-09-09T14:33:42.2001758Z test/integration/test_load_and_run_checkpoint.py::TestLoadAndRunCheckpoint::test_single_linear_model_name_torchao-testing/single-linear-Int4WeightOnlyConfig-v2-0_13_dev SKIPPED 2025-09-09T14:33:42.2002964Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_mm_0_cuda PASSED 2025-09-09T14:33:42.2003613Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_mm_1_cuda PASSED 2025-09-09T14:33:42.2004375Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_mm_float8_0_cuda SKIPPED 2025-09-09T14:33:42.2005118Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_mm_float8_1_cuda SKIPPED 2025-09-09T14:33:42.2005972Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_scaled_mm_0_cuda PASSED 2025-09-09T14:33:42.2006776Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_scaled_mm_1_cpu PASSED 2025-09-09T14:33:42.2007682Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_scaled_mm_2_cuda PASSED 2025-09-09T14:33:42.2008361Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_scaled_mm_3_cpu PASSED 2025-09-09T14:33:42.2009389Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x128[dtype0-2-512-128] SKIPPED 2025-09-09T14:33:42.2010469Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x128[dtype0-2-5120-1280] SKIPPED 2025-09-09T14:33:42.2011665Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x128[dtype0-3-2048-2048] SKIPPED 2025-09-09T14:33:42.2012750Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x128[dtype0-4-3584-640] SKIPPED 2025-09-09T14:33:42.2013940Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x128[dtype0-13-8704-8576] SKIPPED 2025-09-09T14:33:42.2015041Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x128[dtype0-26-18944-1664] SKIPPED 2025-09-09T14:33:42.2016207Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x128[dtype0-67-6656-1408] SKIPPED 2025-09-09T14:33:42.2017443Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x1[dtype0-2-512-128] SKIPPED 2025-09-09T14:33:42.2018495Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x1[dtype0-2-5120-1280] SKIPPED 2025-09-09T14:33:42.2019550Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x1[dtype0-3-2048-2048] SKIPPED 2025-09-09T14:33:42.2020596Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x1[dtype0-4-3584-640] SKIPPED 2025-09-09T14:33:42.2021654Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x1[dtype0-13-8704-8576] SKIPPED 2025-09-09T14:33:42.2022784Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x1[dtype0-26-18944-1664] SKIPPED 2025-09-09T14:33:42.2023906Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x1[dtype0-67-6656-1408] SKIPPED 2025-09-09T14:33:42.6362354Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_act_quant_lhs[128] SKIPPED 2025-09-09T14:33:42.6363811Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_act_quant_lhs[256] SKIPPED 2025-09-09T14:33:42.6364997Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_act_quant_rhs[128] SKIPPED 2025-09-09T14:33:42.6365996Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_act_quant_rhs[256] SKIPPED 2025-09-09T14:33:42.6367077Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_act_quant_transposed_lhs[4096-1024-128] SKIPPED 2025-09-09T14:33:42.6368226Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_act_quant_transposed_lhs[4096-1024-256] SKIPPED 2025-09-09T14:33:42.6369371Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_act_quant_transposed_lhs[4096-16384-128] SKIPPED 2025-09-09T14:33:42.6370522Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_act_quant_transposed_lhs[4096-16384-256] SKIPPED 2025-09-09T14:33:42.6372050Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_weight_quant_rhs[4096-1024-128] SKIPPED 2025-09-09T14:33:42.6373313Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_weight_quant_rhs[4096-1024-256] SKIPPED 2025-09-09T14:33:42.6374405Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_weight_quant_rhs[4096-16384-128] SKIPPED 2025-09-09T14:33:42.6375492Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_weight_quant_rhs[4096-16384-256] SKIPPED 2025-09-09T14:33:42.6376669Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_weight_quant_transposed_rhs[128] SKIPPED 2025-09-09T14:33:42.6377772Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_weight_quant_transposed_rhs[256] SKIPPED 2025-09-09T14:33:42.6378841Z test/prototype/inductor/test_int8_sdpa_fusion.py::SDPAPatternRewriterCpuTests::test_sdpa_int8_rewriter_cpu SKIPPED 2025-09-09T14:33:42.6379863Z test/prototype/module_swap_quantization/test_kmeans_codebook.py::TestKmeansCodebook::test_kmeans_codebook SKIPPED 2025-09-09T14:33:42.6380854Z test/prototype/module_swap_quantization/test_llm_ptq_data_getter.py::TestPTQDataGetter::test_data_getter SKIPPED 2025-09-09T14:33:42.6381812Z test/prototype/module_swap_quantization/test_module_swap.py::TestEmbeddingSwap::test_embedding_swap PASSED 2025-09-09T14:33:42.6382903Z test/prototype/module_swap_quantization/test_module_swap_quantization_utils.py::TestQuantizedModuleUtils::test_set_bit_widths_by_name PASSED 2025-09-09T14:33:42.6384020Z test/prototype/module_swap_quantization/test_quantized_modules.py::TestQuantizedLinear::test_quantize_dynamic PASSED 2025-09-09T14:33:42.6385103Z test/prototype/module_swap_quantization/test_quantized_modules.py::TestQuantizedLinear::test_quantize_dynamic_vectorized PASSED 2025-09-09T14:33:42.6386179Z test/prototype/module_swap_quantization/test_quantized_modules.py::TestQuantizedLinear::test_quantized_linear PASSED 2025-09-09T14:33:42.6387233Z test/prototype/module_swap_quantization/test_quantized_modules.py::TestQuantizedLinear::test_quantized_linear_init PASSED 2025-09-09T14:33:42.6388360Z test/prototype/module_swap_quantization/test_quantized_modules.py::TestQuantizedLinear::test_quantized_linear_passes_gradients PASSED 2025-09-09T14:33:42.6389603Z test/prototype/module_swap_quantization/test_quantized_modules.py::TestQuantizedLinear::test_quantized_linear_passes_gradients_to_activation_scale PASSED 2025-09-09T14:33:42.6390897Z test/prototype/module_swap_quantization/test_quantized_modules.py::TestQuantizedLinear::test_quantized_linear_passes_gradients_to_weight_scale PASSED 2025-09-09T14:33:42.6392160Z test/prototype/module_swap_quantization/test_quantized_modules.py::TestQuantizedLinear::test_set_weight_scale_to_min_max_test_all_options PASSED 2025-09-09T14:33:42.6393366Z test/prototype/module_swap_quantization/test_quantized_modules.py::TestQuantizedLinear::test_set_weight_scale_to_min_max_test_correct PASSED 2025-09-09T14:33:42.6394516Z test/prototype/module_swap_quantization/test_quantized_modules.py::TestQuantizedEmbedding::test_quantized_embedding PASSED 2025-09-09T14:33:42.6395522Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_get_qmin_qmax PASSED 2025-09-09T14:33:42.6396476Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_get_scale_from_min_max PASSED 2025-09-09T14:33:42.6397805Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_get_scale_from_min_max_vectorized PASSED 2025-09-09T14:33:42.6398855Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_get_scale_offset_asymmetric PASSED 2025-09-09T14:33:42.6400112Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_get_scale_offset_from_min_max PASSED 2025-09-09T14:33:42.6401323Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_get_scale_offset_from_min_max_tensorized PASSED 2025-09-09T14:33:42.6402393Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_get_scale_offset_symmetric PASSED 2025-09-09T14:33:42.6403412Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_get_scale_param_size PASSED 2025-09-09T14:33:42.6404376Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_quantize_forward PASSED 2025-09-09T14:33:42.6405402Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_quantize_forward_asymmetric_clipping PASSED 2025-09-09T14:33:42.6414794Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_quantize_forward_symmetric PASSED 2025-09-09T14:33:42.6415959Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_quantize_forward_symmetric_clipping PASSED 2025-09-09T14:33:42.6417100Z test/prototype/module_swap_quantization/test_quantizers.py::TestCodebookQuantizer::test_codebook_quantizer PASSED 2025-09-09T14:33:42.6418129Z test/prototype/module_swap_quantization/test_quantizers.py::TestCodebookQuantizer::test_vector_quantizer PASSED 2025-09-09T14:33:42.6419181Z test/prototype/module_swap_quantization/test_range_setting_methods.py::TestSetWeightMinMax::test_set_weight_min_max PASSED 2025-09-09T14:33:42.6420320Z test/prototype/module_swap_quantization/test_range_setting_methods.py::TestSetWeightMinMax::test_set_weight_min_max_grouped PASSED 2025-09-09T14:33:42.6421403Z test/prototype/module_swap_quantization/test_range_setting_methods.py::TestSetWeightMSE::test_set_weight_mse PASSED 2025-09-09T14:33:42.6422468Z test/prototype/module_swap_quantization/test_range_setting_methods.py::TestSetWeightMSE::test_set_weight_mse_grouped PASSED 2025-09-09T14:33:42.6423695Z test/prototype/module_swap_quantization/test_range_setting_methods.py::TestSetWeightRangeActivationLoss::test_set_weight_range_activation_loss PASSED 2025-09-09T14:33:42.6425078Z test/prototype/module_swap_quantization/test_range_setting_methods.py::TestSetWeightRangeActivationLoss::test_set_weight_range_activation_loss_progressive PASSED 2025-09-09T14:33:42.6426451Z test/prototype/module_swap_quantization/test_range_setting_methods.py::TestStaticActivationRangeSetting::test_static_activation_range_setting PASSED 2025-09-09T14:33:42.6427939Z test/prototype/module_swap_quantization/test_range_setting_methods.py::TestStaticActivationRangeSetting::test_static_activation_range_setting_no_input PASSED 2025-09-09T14:33:42.6429256Z test/prototype/module_swap_quantization/test_range_setting_methods.py::TestQuantizePerGroupScales::test_quantize_per_group_scales PASSED 2025-09-09T14:33:42.6430574Z test/prototype/module_swap_quantization/test_range_setting_methods.py::TestQuantizePerGroupScales::test_quantize_per_group_scales_dont_change_per_channel PASSED 2025-09-09T14:33:42.6431716Z test/prototype/safetensors/test_safetensors_support.py::TestSafeTensors::test_safetensors SKIPPED 2025-09-09T14:33:42.6432670Z test/prototype/safetensors/test_safetensors_utils.py::TestSafeTensorsUtils::test_metadata_torchao SKIPPED 2025-09-09T14:33:42.6433713Z test/prototype/safetensors/test_safetensors_utils.py::TestSafeTensorsUtils::test_not_metadata_torchao_metadata0 SKIPPED 2025-09-09T14:33:42.6434794Z test/prototype/safetensors/test_safetensors_utils.py::TestSafeTensorsUtils::test_not_metadata_torchao_metadata1 SKIPPED 2025-09-09T14:33:42.6436008Z test/prototype/safetensors/test_safetensors_utils.py::TestSafeTensorsUtils::test_not_metadata_torchao_metadata2 SKIPPED 2025-09-09T14:33:42.6437142Z test/prototype/safetensors/test_safetensors_utils.py::TestSafeTensorsUtils::test_not_metadata_torchao_metadata3 SKIPPED 2025-09-09T14:33:42.6438297Z test/prototype/safetensors/test_safetensors_utils.py::TestSafeTensorsUtils::test_not_metadata_torchao_metadata4 SKIPPED 2025-09-09T14:35:40.0698931Z test/prototype/safetensors/test_safetensors_utils.py::TestSafeTensorsUtils::test_not_metadata_torchao_metadata5 SKIPPED 2025-09-09T14:35:40.0701634Z test/prototype/safetensors/test_safetensors_utils.py::TestSafeTensorsUtils::test_not_metadata_torchao_metadata6 SKIPPED 2025-09-09T14:35:40.0703099Z test/prototype/safetensors/test_safetensors_utils.py::TestSafeTensorsUtils::test_not_metadata_torchao_metadata7 SKIPPED 2025-09-09T14:35:40.0704588Z test/prototype/safetensors/test_safetensors_utils.py::TestSafeTensorsUtils::test_not_metadata_torchao_metadata8 SKIPPED 2025-09-09T14:35:40.0705822Z test/prototype/safetensors/test_safetensors_utils.py::TestSafeTensorsUtils::test_not_metadata_torchao_metadata9 SKIPPED 2025-09-09T14:35:40.0706679Z test/prototype/test_awq.py::TestAWQ::test_awq_config SKIPPED (need t...) 2025-09-09T14:35:40.0707329Z test/prototype/test_awq.py::TestAWQ::test_awq_functionality SKIPPED 2025-09-09T14:35:40.0707953Z test/prototype/test_awq.py::TestAWQ::test_awq_loading SKIPPED (need ...) 2025-09-09T14:35:40.0708592Z test/prototype/test_awq.py::TestAWQ::test_awq_loading_vllm SKIPPED (...) 2025-09-09T14:35:40.0709265Z test/prototype/test_blockwise_triton.py::test_blockwise_quant_dequant[dtype0-2-512-128] Relative Error: 0.052687 2025-09-09T14:35:40.0709937Z PASSED 2025-09-09T14:35:40.0710504Z test/prototype/test_blockwise_triton.py::test_blockwise_quant_dequant[dtype0-3-2048-2048] Relative Error: 0.052613 2025-09-09T14:35:40.0711088Z PASSED 2025-09-09T14:35:40.0711594Z test/prototype/test_blockwise_triton.py::test_blockwise_quant_dequant[dtype0-4-3584-640] Relative Error: 0.052573 2025-09-09T14:35:40.0712227Z PASSED 2025-09-09T14:35:40.0712803Z test/prototype/test_blockwise_triton.py::test_blockwise_quant_dequant[dtype0-13-8704-8576] Relative Error: 0.052638 2025-09-09T14:35:40.0713399Z PASSED 2025-09-09T14:35:40.0713918Z test/prototype/test_blockwise_triton.py::test_blockwise_quant_dequant[dtype0-26-18944-1664] Relative Error: 0.052636 2025-09-09T14:35:40.0714597Z PASSED 2025-09-09T14:35:40.0715147Z test/prototype/test_blockwise_triton.py::test_blockwise_quant_dequant[dtype0-67-6656-1408] Relative Error: 0.052665 2025-09-09T14:35:40.0715727Z PASSED 2025-09-09T14:35:40.0716221Z test/prototype/test_blockwise_triton.py::test_blockwise_fp8_gemm[dtype0-2-512-128] Relative Error: 0.070320 2025-09-09T14:35:40.0716791Z PASSED 2025-09-09T14:35:40.0717319Z test/prototype/test_blockwise_triton.py::test_blockwise_fp8_gemm[dtype0-3-2048-2048] Relative Error: 0.074321 2025-09-09T14:35:40.0717878Z PASSED 2025-09-09T14:35:40.0718378Z test/prototype/test_blockwise_triton.py::test_blockwise_fp8_gemm[dtype0-4-3584-640] Relative Error: 0.074240 2025-09-09T14:35:40.0718960Z PASSED 2025-09-09T14:35:40.0719511Z test/prototype/test_blockwise_triton.py::test_blockwise_fp8_gemm[dtype0-13-8704-8576] Relative Error: 0.073402 2025-09-09T14:35:40.0720067Z PASSED 2025-09-09T14:35:40.0720580Z test/prototype/test_blockwise_triton.py::test_blockwise_fp8_gemm[dtype0-26-18944-1664] Relative Error: 0.073144 2025-09-09T14:35:40.0721169Z PASSED 2025-09-09T14:35:40.0721704Z test/prototype/test_blockwise_triton.py::test_blockwise_fp8_gemm[dtype0-67-6656-1408] Relative Error: 0.073484 2025-09-09T14:35:40.0722271Z PASSED 2025-09-09T14:35:40.0722915Z test/prototype/test_codebook_coreml.py::TestCodebookQuantization::test_choose_qparams_codebook SKIPPED 2025-09-09T14:35:40.0723998Z test/prototype/test_codebook_coreml.py::TestCodebookQuantization::test_choose_qparams_codebook_row_grouping SKIPPED 2025-09-09T14:35:40.0725603Z test/prototype/test_codebook_coreml.py::TestCodebookQuantization::test_codebook_quantized_tensor_from_float SKIPPED 2025-09-09T14:35:40.0726942Z test/prototype/test_codebook_coreml.py::TestCodebookQuantization::test_codebook_quantized_tensor_from_float2 SKIPPED 2025-09-09T14:35:40.0728137Z test/prototype/test_codebook_coreml.py::TestCodebookQuantization::test_codebook_quantized_tensor_from_float_row_grouping SKIPPED 2025-09-09T14:35:40.0729136Z test/prototype/test_codebook_coreml.py::TestCodebookQuantization::test_export SKIPPED 2025-09-09T14:35:40.0729977Z test/prototype/test_codebook_coreml.py::TestCodebookQuantization::test_quantize_api SKIPPED 2025-09-09T14:35:40.0730868Z test/prototype/test_codebook_quant.py::TestCodebookQuantization::test_choose_qparams_codebook PASSED 2025-09-09T14:35:40.0731850Z test/prototype/test_codebook_quant.py::TestCodebookQuantization::test_codebook_quantized_tensor_from_float PASSED 2025-09-09T14:35:40.0732884Z test/prototype/test_codebook_quant.py::TestCodebookQuantization::test_codebook_quantized_tensor_from_float2 PASSED 2025-09-09T14:35:40.0733810Z test/prototype/test_codebook_quant.py::TestCodebookQuantization::test_quantize_api PASSED 2025-09-09T14:35:40.0734767Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-1-granularity0-dtype0] SKIPPED 2025-09-09T14:35:40.0735729Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-1-granularity0-dtype1] SKIPPED 2025-09-09T14:35:40.0736764Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-1-granularity1-dtype0] SKIPPED 2025-09-09T14:35:40.0737712Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-1-granularity1-dtype1] SKIPPED 2025-09-09T14:35:40.0738653Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-2-granularity0-dtype0] SKIPPED 2025-09-09T14:35:40.0739607Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-2-granularity0-dtype1] SKIPPED 2025-09-09T14:35:40.0740558Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-2-granularity1-dtype0] SKIPPED 2025-09-09T14:35:40.0741533Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-2-granularity1-dtype1] SKIPPED 2025-09-09T14:35:40.0742487Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-3-granularity0-dtype0] SKIPPED 2025-09-09T14:35:40.0743440Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-3-granularity0-dtype1] SKIPPED 2025-09-09T14:35:40.0744398Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-3-granularity1-dtype0] SKIPPED 2025-09-09T14:35:40.0745350Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-3-granularity1-dtype1] SKIPPED 2025-09-09T14:35:40.0746316Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-4-granularity0-dtype0] SKIPPED 2025-09-09T14:35:40.0747284Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-4-granularity0-dtype1] SKIPPED 2025-09-09T14:35:40.0748236Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-4-granularity1-dtype0] SKIPPED 2025-09-09T14:35:40.0749197Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-4-granularity1-dtype1] SKIPPED 2025-09-09T14:35:40.0750146Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-1-granularity0-dtype0] SKIPPED 2025-09-09T14:35:40.0751100Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-1-granularity0-dtype1] SKIPPED 2025-09-09T14:35:40.0752054Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-1-granularity1-dtype0] SKIPPED 2025-09-09T14:35:40.0753117Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-1-granularity1-dtype1] SKIPPED 2025-09-09T14:35:40.0754157Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-2-granularity0-dtype0] SKIPPED 2025-09-09T14:35:40.0755108Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-2-granularity0-dtype1] SKIPPED 2025-09-09T14:35:40.0756062Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-2-granularity1-dtype0] SKIPPED 2025-09-09T14:35:40.0757018Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-2-granularity1-dtype1] SKIPPED 2025-09-09T14:35:40.0757965Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-3-granularity0-dtype0] SKIPPED 2025-09-09T14:35:40.0758917Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-3-granularity0-dtype1] SKIPPED 2025-09-09T14:35:40.0759956Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-3-granularity1-dtype0] SKIPPED 2025-09-09T14:35:40.0760995Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-3-granularity1-dtype1] SKIPPED 2025-09-09T14:35:40.0761964Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-4-granularity0-dtype0] SKIPPED 2025-09-09T14:35:40.0763022Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-4-granularity0-dtype1] SKIPPED 2025-09-09T14:35:40.0764108Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-4-granularity1-dtype0] SKIPPED 2025-09-09T14:35:40.0765200Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-4-granularity1-dtype1] SKIPPED 2025-09-09T14:35:40.0766243Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-1-granularity0-dtype0] SKIPPED 2025-09-09T14:35:40.0767246Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-1-granularity0-dtype1] SKIPPED 2025-09-09T14:35:40.0768209Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-1-granularity1-dtype0] SKIPPED 2025-09-09T14:35:40.0769169Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-1-granularity1-dtype1] SKIPPED 2025-09-09T14:35:40.0770130Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-2-granularity0-dtype0] SKIPPED 2025-09-09T14:35:47.1252541Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-2-granularity0-dtype1] SKIPPED 2025-09-09T14:35:47.1253467Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-2-granularity1-dtype0] SKIPPED 2025-09-09T14:35:47.1254343Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-2-granularity1-dtype1] SKIPPED 2025-09-09T14:35:47.1255256Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-3-granularity0-dtype0] SKIPPED 2025-09-09T14:35:47.1256226Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-3-granularity0-dtype1] SKIPPED 2025-09-09T14:35:47.1257126Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-3-granularity1-dtype0] SKIPPED 2025-09-09T14:35:47.1257982Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-3-granularity1-dtype1] SKIPPED 2025-09-09T14:35:47.1258847Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-4-granularity0-dtype0] SKIPPED 2025-09-09T14:35:47.1259705Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-4-granularity0-dtype1] SKIPPED 2025-09-09T14:35:47.1260579Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-4-granularity1-dtype0] SKIPPED 2025-09-09T14:35:47.1261830Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-4-granularity1-dtype1] SKIPPED 2025-09-09T14:35:47.1262703Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-1-granularity0-dtype0] SKIPPED 2025-09-09T14:35:47.1263734Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-1-granularity0-dtype1] SKIPPED 2025-09-09T14:35:47.1264599Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-1-granularity1-dtype0] SKIPPED 2025-09-09T14:35:47.1265475Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-1-granularity1-dtype1] SKIPPED 2025-09-09T14:35:47.1266339Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-2-granularity0-dtype0] SKIPPED 2025-09-09T14:35:47.1267198Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-2-granularity0-dtype1] SKIPPED 2025-09-09T14:35:47.1268067Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-2-granularity1-dtype0] SKIPPED 2025-09-09T14:35:47.1268932Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-2-granularity1-dtype1] SKIPPED 2025-09-09T14:35:47.1269797Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-3-granularity0-dtype0] SKIPPED 2025-09-09T14:35:47.1270668Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-3-granularity0-dtype1] SKIPPED 2025-09-09T14:35:47.1271529Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-3-granularity1-dtype0] SKIPPED 2025-09-09T14:35:47.1272390Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-3-granularity1-dtype1] SKIPPED 2025-09-09T14:35:47.1273250Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-4-granularity0-dtype0] SKIPPED 2025-09-09T14:35:47.1274116Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-4-granularity0-dtype1] SKIPPED 2025-09-09T14:35:47.1274990Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-4-granularity1-dtype0] SKIPPED 2025-09-09T14:35:47.1275850Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-4-granularity1-dtype1] SKIPPED 2025-09-09T14:35:47.1276690Z test/prototype/test_gguf_quant.py::TestGGUFQuantization::test_choose_qparams_gguf PASSED 2025-09-09T14:35:47.1277552Z test/prototype/test_gguf_quant.py::TestGGUFQuantization::test_gguf_quantized_tensor_from_float PASSED 2025-09-09T14:35:47.1278372Z test/prototype/test_gguf_quant.py::TestGGUFQuantization::test_quantize_api PASSED 2025-09-09T14:35:47.1279198Z test/prototype/test_mixed_precision.py::TestWeightOnlyQuantNaive::test_quantization_intNwo PASSED 2025-09-09T14:35:47.1280008Z test/prototype/test_parametrization.py::TestFakeSparsity::test_jit_trace PASSED 2025-09-09T14:35:47.1280775Z test/prototype/test_parametrization.py::TestFakeSparsity::test_masking_logic PASSED 2025-09-09T14:35:47.1281576Z test/prototype/test_parametrization.py::TestFakeSparsity::test_state_dict_preserved PASSED 2025-09-09T14:35:47.1282411Z test/prototype/test_parametrization.py::TestFakeSparsity::test_weights_parametrized PASSED 2025-09-09T14:35:47.1283164Z test/prototype/test_paretoq.py::TestParetoQ::test_quantize_functions PASSED 2025-09-09T14:35:47.1283840Z test/prototype/test_paretoq.py::TestParetoQ::test_quantized_linear PASSED 2025-09-09T14:35:47.1284775Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_0_unif_quant_False_hard_prox_False_per_group_quantizer_False PASSED 2025-09-09T14:35:47.1285916Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_0_unif_quant_False_hard_prox_False_per_group_quantizer_True PASSED 2025-09-09T14:35:47.1287106Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_0_unif_quant_False_hard_prox_True_per_group_quantizer_False PASSED 2025-09-09T14:35:47.1288387Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_0_unif_quant_False_hard_prox_True_per_group_quantizer_True PASSED 2025-09-09T14:35:47.1289523Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_0_unif_quant_True_hard_prox_False_per_group_quantizer_False PASSED 2025-09-09T14:35:47.1290733Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_0_unif_quant_True_hard_prox_False_per_group_quantizer_True PASSED 2025-09-09T14:35:47.1291860Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_0_unif_quant_True_hard_prox_True_per_group_quantizer_False PASSED 2025-09-09T14:35:47.1293008Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_0_unif_quant_True_hard_prox_True_per_group_quantizer_True PASSED 2025-09-09T14:35:47.1294141Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_1_unif_quant_False_hard_prox_False_per_group_quantizer_False PASSED 2025-09-09T14:35:47.1295302Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_1_unif_quant_False_hard_prox_False_per_group_quantizer_True PASSED 2025-09-09T14:35:47.1296514Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_1_unif_quant_False_hard_prox_True_per_group_quantizer_False PASSED 2025-09-09T14:35:47.1298073Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_1_unif_quant_False_hard_prox_True_per_group_quantizer_True PASSED 2025-09-09T14:35:47.1299217Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_1_unif_quant_True_hard_prox_False_per_group_quantizer_False PASSED 2025-09-09T14:35:47.1300349Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_1_unif_quant_True_hard_prox_False_per_group_quantizer_True PASSED 2025-09-09T14:35:47.1301468Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_1_unif_quant_True_hard_prox_True_per_group_quantizer_False PASSED 2025-09-09T14:35:47.1302603Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_1_unif_quant_True_hard_prox_True_per_group_quantizer_True PASSED 2025-09-09T14:35:47.1303732Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_2_unif_quant_False_hard_prox_False_per_group_quantizer_False PASSED 2025-09-09T14:35:47.1304879Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_2_unif_quant_False_hard_prox_False_per_group_quantizer_True PASSED 2025-09-09T14:35:47.1306021Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_2_unif_quant_False_hard_prox_True_per_group_quantizer_False PASSED 2025-09-09T14:35:47.1307192Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_2_unif_quant_False_hard_prox_True_per_group_quantizer_True PASSED 2025-09-09T14:35:47.1308338Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_2_unif_quant_True_hard_prox_False_per_group_quantizer_False PASSED 2025-09-09T14:35:47.1309471Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_2_unif_quant_True_hard_prox_False_per_group_quantizer_True PASSED 2025-09-09T14:35:47.1310595Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_2_unif_quant_True_hard_prox_True_per_group_quantizer_False PASSED 2025-09-09T14:35:47.1311730Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_2_unif_quant_True_hard_prox_True_per_group_quantizer_True PASSED 2025-09-09T14:35:47.1312860Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_4_unif_quant_False_hard_prox_False_per_group_quantizer_False PASSED 2025-09-09T14:35:47.1314002Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_4_unif_quant_False_hard_prox_False_per_group_quantizer_True PASSED 2025-09-09T14:35:47.1315273Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_4_unif_quant_False_hard_prox_True_per_group_quantizer_False PASSED 2025-09-09T14:35:47.1316408Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_4_unif_quant_False_hard_prox_True_per_group_quantizer_True PASSED 2025-09-09T14:36:19.0328150Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_4_unif_quant_True_hard_prox_False_per_group_quantizer_False PASSED 2025-09-09T14:36:19.0331069Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_4_unif_quant_True_hard_prox_False_per_group_quantizer_True PASSED 2025-09-09T14:36:19.0332260Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_4_unif_quant_True_hard_prox_True_per_group_quantizer_False PASSED 2025-09-09T14:36:19.0333401Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_4_unif_quant_True_hard_prox_True_per_group_quantizer_True PASSED 2025-09-09T14:36:19.0334468Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_int4_weight_only_e2e PASSED 2025-09-09T14:36:19.0335451Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_int4_weight_only_group_size_256 PASSED 2025-09-09T14:36:19.0336527Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_int4_weight_only_group_size_32 PASSED 2025-09-09T14:36:19.0337569Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_b_2_group_size_32 PASSED 2025-09-09T14:36:19.0338570Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_b_2_group_size_512 PASSED 2025-09-09T14:36:19.0339592Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_b_3_group_size_32 PASSED 2025-09-09T14:36:19.0340627Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_b_3_group_size_512 PASSED 2025-09-09T14:36:19.0341650Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_b_4_group_size_32 PASSED 2025-09-09T14:36:19.0342651Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_b_4_group_size_512 PASSED 2025-09-09T14:36:19.0343654Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_b_8_group_size_32 PASSED 2025-09-09T14:36:19.0344642Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_b_8_group_size_512 PASSED 2025-09-09T14:36:19.0345713Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_e2e_b_2 PASSED 2025-09-09T14:36:19.0346777Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_e2e_b_3 PASSED 2025-09-09T14:36:19.0347809Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_e2e_b_4 PASSED 2025-09-09T14:36:19.0348840Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_e2e_b_8 PASSED 2025-09-09T14:36:19.0349899Z test/prototype/test_parq.py::TestStretchedUnifTorchaoQuantizer::test_intx_weight_only PASSED 2025-09-09T14:36:19.0350996Z test/prototype/test_parq.py::TestStretchedUnifTorchaoQuantizer::test_intx_weight_only_e2e PASSED 2025-09-09T14:36:19.0352176Z test/prototype/test_parq.py::TestStretchedUnifTorchaoQuantizer::test_intx_weight_only_parq_equivalent PASSED 2025-09-09T14:36:19.0353620Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_2_float16_group_size_128 PASSED 2025-09-09T14:36:19.0355213Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_2_float16_group_size_32 PASSED 2025-09-09T14:36:19.0356823Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_2_float32_group_size_128 PASSED 2025-09-09T14:36:19.0360615Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_2_float32_group_size_32 PASSED 2025-09-09T14:36:19.0362244Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_3_float16_group_size_128 PASSED 2025-09-09T14:36:19.0364028Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_3_float16_group_size_32 PASSED 2025-09-09T14:36:19.0365627Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_3_float32_group_size_128 PASSED 2025-09-09T14:36:19.0367271Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_3_float32_group_size_32 PASSED 2025-09-09T14:36:19.0368872Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_4_float16_group_size_128 PASSED 2025-09-09T14:36:19.0370484Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_4_float16_group_size_32 PASSED 2025-09-09T14:36:19.0372297Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_4_float32_group_size_128 PASSED 2025-09-09T14:36:19.0374033Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_4_float32_group_size_32 PASSED 2025-09-09T14:36:19.0375639Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_8_float16_group_size_128 PASSED 2025-09-09T14:36:19.0377339Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_8_float16_group_size_32 PASSED 2025-09-09T14:36:19.0378944Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_8_float32_group_size_128 PASSED 2025-09-09T14:36:19.0380684Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_8_float32_group_size_32 PASSED 2025-09-09T14:36:19.0382208Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_bitnet_training_compile_False SKIPPED 2025-09-09T14:36:19.0383426Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_bitnet_training_compile_True SKIPPED 2025-09-09T14:36:19.0384828Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_False_config0_module_swap_False PASSED 2025-09-09T14:36:19.0386531Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_False_config0_module_swap_True PASSED 2025-09-09T14:36:19.0388172Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_False_config1_module_swap_False PASSED 2025-09-09T14:36:19.0389757Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_False_config1_module_swap_True PASSED 2025-09-09T14:36:19.0391582Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_False_config2_module_swap_False PASSED 2025-09-09T14:36:19.0393165Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_False_config2_module_swap_True PASSED 2025-09-09T14:36:19.0394852Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_False_config3_module_swap_False PASSED 2025-09-09T14:36:19.0396436Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_False_config3_module_swap_True PASSED 2025-09-09T14:36:19.0398600Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_True_config0_module_swap_False PASSED 2025-09-09T14:36:19.0400193Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_True_config0_module_swap_True PASSED 2025-09-09T14:36:19.0401904Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_True_config1_module_swap_False PASSED 2025-09-09T14:36:19.0403466Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_True_config1_module_swap_True PASSED 2025-09-09T14:36:19.0405042Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_True_config2_module_swap_False PASSED 2025-09-09T14:36:19.0406615Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_True_config2_module_swap_True PASSED 2025-09-09T14:36:19.0408179Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_True_config3_module_swap_False PASSED 2025-09-09T14:36:19.0409754Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_True_config3_module_swap_True PASSED 2025-09-09T14:36:19.0411073Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_stochastic_rounding_device_cpu PASSED 2025-09-09T14:36:19.0412082Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_stochastic_rounding_device_cuda PASSED 2025-09-09T14:36:19.0413199Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims0_bias_False_device_cpu PASSED 2025-09-09T14:38:30.6538726Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims0_bias_False_device_cuda PASSED 2025-09-09T14:38:30.6540041Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims0_bias_True_device_cpu PASSED 2025-09-09T14:38:30.6541697Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims0_bias_True_device_cuda PASSED 2025-09-09T14:38:30.6542990Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims1_bias_False_device_cpu PASSED 2025-09-09T14:38:30.6544359Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims1_bias_False_device_cuda PASSED 2025-09-09T14:38:30.6545640Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims1_bias_True_device_cpu PASSED 2025-09-09T14:38:30.6547002Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims1_bias_True_device_cuda PASSED 2025-09-09T14:38:30.6548275Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims2_bias_False_device_cpu PASSED 2025-09-09T14:38:30.6549694Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims2_bias_False_device_cuda PASSED 2025-09-09T14:38:30.6551002Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims2_bias_True_device_cpu PASSED 2025-09-09T14:38:30.6552306Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims2_bias_True_device_cuda PASSED 2025-09-09T14:38:30.6553701Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims0_bias_False_device_cpu PASSED 2025-09-09T14:38:30.6555455Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims0_bias_False_device_cuda PASSED 2025-09-09T14:38:30.6556796Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims0_bias_True_device_cpu PASSED 2025-09-09T14:38:30.6559712Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims0_bias_True_device_cuda PASSED 2025-09-09T14:38:30.6560985Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims1_bias_False_device_cpu PASSED 2025-09-09T14:38:30.6562314Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims1_bias_False_device_cuda PASSED 2025-09-09T14:38:30.6563630Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims1_bias_True_device_cpu PASSED 2025-09-09T14:38:30.6564861Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims1_bias_True_device_cuda PASSED 2025-09-09T14:38:30.6566369Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims2_bias_False_device_cpu PASSED 2025-09-09T14:38:30.6567824Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims2_bias_False_device_cuda PASSED 2025-09-09T14:38:30.6569099Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims2_bias_True_device_cpu PASSED 2025-09-09T14:38:30.6570338Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims2_bias_True_device_cuda PASSED 2025-09-09T14:38:30.6571519Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_training_compile_False_device_cpu PASSED 2025-09-09T14:38:30.6572649Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_training_compile_False_device_cuda PASSED 2025-09-09T14:38:30.6573784Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_training_compile_True_device_cpu PASSED 2025-09-09T14:38:30.6574982Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_training_compile_True_device_cuda PASSED 2025-09-09T14:38:30.6575942Z test/prototype/test_quantized_training.py::TestFSDP2::test_fsdp2_correctness dist init r=0, world=2 2025-09-09T14:38:30.6576452Z dist init r=1, world=2 2025-09-09T14:38:30.6576715Z PASSED 2025-09-09T14:38:30.6577189Z test/prototype/test_quantized_training.py::TestFSDP2::test_precompute_bitnet_scale dist init r=0, world=2 2025-09-09T14:38:30.6577704Z dist init r=1, world=2 2025-09-09T14:38:30.6577965Z PASSED 2025-09-09T14:38:30.6578422Z test/prototype/test_scheduler.py::TestScheduler::test_constructor PASSED 2025-09-09T14:38:30.6579175Z test/prototype/test_scheduler.py::TestScheduler::test_lambda_scheduler PASSED 2025-09-09T14:38:30.6579875Z test/prototype/test_scheduler.py::TestScheduler::test_order_of_steps PASSED 2025-09-09T14:38:30.6580528Z test/prototype/test_scheduler.py::TestScheduler::test_step PASSED 2025-09-09T14:38:30.6581203Z test/prototype/test_scheduler.py::TestCubicScheduler::test_constructor PASSED 2025-09-09T14:38:30.6581888Z test/prototype/test_scheduler.py::TestCubicScheduler::test_step PASSED 2025-09-09T14:38:30.6582792Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_dynamic_device_cpu_bfloat16 SKIPPED 2025-09-09T14:38:30.6583880Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_dynamic_device_cpu_float16 SKIPPED 2025-09-09T14:38:30.6585077Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_dynamic_device_cpu_float32 SKIPPED 2025-09-09T14:38:30.6586179Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_dynamic_device_cuda_bfloat16 SKIPPED 2025-09-09T14:38:30.6587346Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_dynamic_device_cuda_float16 SKIPPED 2025-09-09T14:38:30.6588647Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_dynamic_device_cuda_float32 SKIPPED 2025-09-09T14:38:30.6589973Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_static_device_cpu_bfloat16 SKIPPED 2025-09-09T14:38:30.6591056Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_static_device_cpu_float16 SKIPPED 2025-09-09T14:38:30.6592149Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_static_device_cpu_float32 SKIPPED 2025-09-09T14:38:30.6593236Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_static_device_cuda_bfloat16 SKIPPED 2025-09-09T14:38:30.6594337Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_static_device_cuda_float16 SKIPPED 2025-09-09T14:38:30.6595431Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_static_device_cuda_float32 SKIPPED 2025-09-09T14:38:30.6596526Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_dynamic_device_cpu_bfloat16 SKIPPED 2025-09-09T14:38:30.6597859Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_dynamic_device_cpu_float16 SKIPPED 2025-09-09T14:38:30.6598992Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_dynamic_device_cpu_float32 SKIPPED 2025-09-09T14:38:30.6600112Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_dynamic_device_cuda_bfloat16 SKIPPED 2025-09-09T14:38:30.6601219Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_dynamic_device_cuda_float16 SKIPPED 2025-09-09T14:38:30.6602305Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_dynamic_device_cuda_float32 SKIPPED 2025-09-09T14:38:30.6603450Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_static_device_cpu_bfloat16 SKIPPED 2025-09-09T14:38:30.6604547Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_static_device_cpu_float16 SKIPPED 2025-09-09T14:38:30.6605648Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_static_device_cpu_float32 SKIPPED 2025-09-09T14:38:30.6606747Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_static_device_cuda_bfloat16 SKIPPED 2025-09-09T14:38:30.6618357Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_static_device_cuda_float16 SKIPPED 2025-09-09T14:38:30.6619674Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_static_device_cuda_float32 SKIPPED 2025-09-09T14:38:30.6966414Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_dynamic_device_cpu_bfloat16 SKIPPED 2025-09-09T14:38:30.6967919Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_dynamic_device_cpu_float16 SKIPPED 2025-09-09T14:38:30.6969490Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_dynamic_device_cpu_float32 SKIPPED 2025-09-09T14:38:30.6970808Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_dynamic_device_cuda_bfloat16 SKIPPED 2025-09-09T14:38:30.6972050Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_dynamic_device_cuda_float16 SKIPPED 2025-09-09T14:38:30.6973146Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_dynamic_device_cuda_float32 SKIPPED 2025-09-09T14:38:30.6974259Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_static_device_cpu_bfloat16 SKIPPED 2025-09-09T14:38:30.6975350Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_static_device_cpu_float16 SKIPPED 2025-09-09T14:38:30.6976535Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_static_device_cpu_float32 SKIPPED 2025-09-09T14:38:30.6977628Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_static_device_cuda_bfloat16 SKIPPED 2025-09-09T14:38:30.6978732Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_static_device_cuda_float16 SKIPPED 2025-09-09T14:38:30.6979813Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_static_device_cuda_float32 SKIPPED 2025-09-09T14:38:30.6980968Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_dynamic_device_cpu_bfloat16 SKIPPED 2025-09-09T14:38:30.6982173Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_dynamic_device_cpu_float16 SKIPPED 2025-09-09T14:38:30.6983375Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_dynamic_device_cpu_float32 SKIPPED 2025-09-09T14:38:30.6984589Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_dynamic_device_cuda_bfloat16 SKIPPED 2025-09-09T14:38:30.6985802Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_dynamic_device_cuda_float16 SKIPPED 2025-09-09T14:38:30.6987025Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_dynamic_device_cuda_float32 SKIPPED 2025-09-09T14:38:30.6988233Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_static_device_cpu_bfloat16 SKIPPED 2025-09-09T14:38:30.6989425Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_static_device_cpu_float16 SKIPPED 2025-09-09T14:38:30.6990628Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_static_device_cpu_float32 SKIPPED 2025-09-09T14:38:30.6991827Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_static_device_cuda_bfloat16 SKIPPED 2025-09-09T14:38:30.6993045Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_static_device_cuda_float16 SKIPPED 2025-09-09T14:38:30.6994245Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_static_device_cuda_float32 SKIPPED 2025-09-09T14:38:30.6995455Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_dynamic_device_cpu_bfloat16 SKIPPED 2025-09-09T14:38:30.6996772Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_dynamic_device_cpu_float16 SKIPPED 2025-09-09T14:38:30.6998278Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_dynamic_device_cpu_float32 SKIPPED 2025-09-09T14:38:30.6999629Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_dynamic_device_cuda_bfloat16 SKIPPED 2025-09-09T14:38:30.7000853Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_dynamic_device_cuda_float16 SKIPPED 2025-09-09T14:38:30.7002068Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_dynamic_device_cuda_float32 SKIPPED 2025-09-09T14:38:30.7003269Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_static_device_cpu_bfloat16 SKIPPED 2025-09-09T14:38:30.7004476Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_static_device_cpu_float16 SKIPPED 2025-09-09T14:38:30.7005666Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_static_device_cpu_float32 SKIPPED 2025-09-09T14:38:30.7006877Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_static_device_cuda_bfloat16 SKIPPED 2025-09-09T14:38:30.7008091Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_static_device_cuda_float16 SKIPPED 2025-09-09T14:38:30.7009290Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_static_device_cuda_float32 SKIPPED 2025-09-09T14:38:30.7010503Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_dynamic_device_cpu_bfloat16 SKIPPED 2025-09-09T14:38:30.7011718Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_dynamic_device_cpu_float16 SKIPPED 2025-09-09T14:38:30.7012933Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_dynamic_device_cpu_float32 SKIPPED 2025-09-09T14:38:30.7014157Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_dynamic_device_cuda_bfloat16 SKIPPED 2025-09-09T14:38:30.7015388Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_dynamic_device_cuda_float16 SKIPPED 2025-09-09T14:38:30.7016696Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_dynamic_device_cuda_float32 SKIPPED 2025-09-09T14:38:30.7017919Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_static_device_cpu_bfloat16 SKIPPED 2025-09-09T14:38:30.7019180Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_static_device_cpu_float16 SKIPPED 2025-09-09T14:38:30.7020399Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_static_device_cpu_float32 SKIPPED 2025-09-09T14:38:30.7021616Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_static_device_cuda_bfloat16 SKIPPED 2025-09-09T14:38:30.7022947Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_static_device_cuda_float16 SKIPPED 2025-09-09T14:38:30.7024359Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_static_device_cuda_float32 SKIPPED 2025-09-09T14:38:30.7025816Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_dynamic_device_cpu_bfloat16 SKIPPED 2025-09-09T14:38:30.7027182Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_dynamic_device_cpu_float16 SKIPPED 2025-09-09T14:38:30.7028496Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_dynamic_device_cpu_float32 SKIPPED 2025-09-09T14:38:30.7029874Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_dynamic_device_cuda_bfloat16 SKIPPED 2025-09-09T14:38:30.7031240Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_dynamic_device_cuda_float16 SKIPPED 2025-09-09T14:38:30.7032536Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_dynamic_device_cuda_float32 SKIPPED 2025-09-09T14:40:16.0254282Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_static_device_cpu_bfloat16 SKIPPED 2025-09-09T14:40:16.0257171Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_static_device_cpu_float16 SKIPPED 2025-09-09T14:40:16.0258720Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_static_device_cpu_float32 SKIPPED 2025-09-09T14:40:16.0260228Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_static_device_cuda_bfloat16 SKIPPED 2025-09-09T14:40:16.0261736Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_static_device_cuda_float16 SKIPPED 2025-09-09T14:40:16.0263271Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_static_device_cuda_float32 SKIPPED 2025-09-09T14:40:16.0264810Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_dynamic_device_cpu_bfloat16 SKIPPED 2025-09-09T14:40:16.0266355Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_dynamic_device_cpu_float16 SKIPPED 2025-09-09T14:40:16.0267887Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_dynamic_device_cpu_float32 SKIPPED 2025-09-09T14:40:16.0269421Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_dynamic_device_cuda_bfloat16 SKIPPED 2025-09-09T14:40:16.0270973Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_dynamic_device_cuda_float16 SKIPPED 2025-09-09T14:40:16.0272527Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_dynamic_device_cuda_float32 SKIPPED 2025-09-09T14:40:16.0274065Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_static_device_cpu_bfloat16 SKIPPED 2025-09-09T14:40:16.0275595Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_static_device_cpu_float16 SKIPPED 2025-09-09T14:40:16.0277117Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_static_device_cpu_float32 SKIPPED 2025-09-09T14:40:16.0278639Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_static_device_cuda_bfloat16 SKIPPED 2025-09-09T14:40:16.0280645Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_static_device_cuda_float16 SKIPPED 2025-09-09T14:40:16.0282182Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_static_device_cuda_float32 SKIPPED 2025-09-09T14:40:16.0283892Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_dynamic_device_cpu_bfloat16 SKIPPED 2025-09-09T14:40:16.0285435Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_dynamic_device_cpu_float16 SKIPPED 2025-09-09T14:40:16.0286984Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_dynamic_device_cpu_float32 SKIPPED 2025-09-09T14:40:16.0288522Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_dynamic_device_cuda_bfloat16 SKIPPED 2025-09-09T14:40:16.0290075Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_dynamic_device_cuda_float16 SKIPPED 2025-09-09T14:40:16.0291616Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_dynamic_device_cuda_float32 SKIPPED 2025-09-09T14:40:16.0293183Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_static_device_cpu_bfloat16 SKIPPED 2025-09-09T14:40:16.0294699Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_static_device_cpu_float16 SKIPPED 2025-09-09T14:40:16.0296297Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_static_device_cpu_float32 SKIPPED 2025-09-09T14:40:16.0298046Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_static_device_cuda_bfloat16 SKIPPED 2025-09-09T14:40:16.0299590Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_static_device_cuda_float16 SKIPPED 2025-09-09T14:40:16.0301109Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_static_device_cuda_float32 SKIPPED 2025-09-09T14:40:16.0302294Z test/prototype/test_sparsifier.py::TestBaseSparsifier::test_constructor PASSED 2025-09-09T14:40:16.0303177Z test/prototype/test_sparsifier.py::TestBaseSparsifier::test_convert PASSED 2025-09-09T14:40:16.0304037Z test/prototype/test_sparsifier.py::TestBaseSparsifier::test_mask_squash PASSED 2025-09-09T14:40:16.0304940Z test/prototype/test_sparsifier.py::TestBaseSparsifier::test_mask_squash_with_params1 PASSED 2025-09-09T14:40:16.0305784Z test/prototype/test_sparsifier.py::TestBaseSparsifier::test_mask_squash_with_params2 PASSED 2025-09-09T14:40:16.0306622Z test/prototype/test_sparsifier.py::TestBaseSparsifier::test_mask_squash_with_params3 PASSED 2025-09-09T14:40:16.0307405Z test/prototype/test_sparsifier.py::TestBaseSparsifier::test_prepare_config PASSED 2025-09-09T14:40:16.0308135Z test/prototype/test_sparsifier.py::TestBaseSparsifier::test_state_dict PASSED 2025-09-09T14:40:16.0308825Z test/prototype/test_sparsifier.py::TestBaseSparsifier::test_step PASSED 2025-09-09T14:40:16.0309557Z test/prototype/test_sparsifier.py::TestWeightNormSparsifier::test_constructor PASSED 2025-09-09T14:40:16.0310352Z test/prototype/test_sparsifier.py::TestWeightNormSparsifier::test_mask_squash PASSED 2025-09-09T14:40:16.0311115Z test/prototype/test_sparsifier.py::TestWeightNormSparsifier::test_prepare PASSED 2025-09-09T14:40:16.0311905Z test/prototype/test_sparsifier.py::TestWeightNormSparsifier::test_sparsity_levels PASSED 2025-09-09T14:40:16.0312884Z test/prototype/test_sparsifier.py::TestWeightNormSparsifier::test_step PASSED 2025-09-09T14:40:16.0313644Z test/prototype/test_sparsifier.py::TestWeightNormSparsifier::test_step_2_of_4 PASSED 2025-09-09T14:40:16.0314573Z test/prototype/test_sparsifier.py::TestNearlyDiagonalSparsifier::test_constructor PASSED 2025-09-09T14:40:16.0315401Z test/prototype/test_sparsifier.py::TestNearlyDiagonalSparsifier::test_mask_squash PASSED 2025-09-09T14:40:16.0316214Z test/prototype/test_sparsifier.py::TestNearlyDiagonalSparsifier::test_prepare PASSED 2025-09-09T14:40:16.0317053Z test/prototype/test_sparsifier.py::TestNearlyDiagonalSparsifier::test_sparsity_levels PASSED 2025-09-09T14:40:16.0317862Z test/prototype/test_sparsifier.py::TestNearlyDiagonalSparsifier::test_step PASSED 2025-09-09T14:40:16.0318666Z test/prototype/test_sparsity_utils.py::TestSparsityUtilFunctions::test_fqn_to_module PASSED 2025-09-09T14:40:16.0319542Z test/prototype/test_sparsity_utils.py::TestSparsityUtilFunctions::test_fqn_to_module_fail PASSED 2025-09-09T14:40:16.0320455Z test/prototype/test_sparsity_utils.py::TestSparsityUtilFunctions::test_fqn_to_module_for_tensors PASSED 2025-09-09T14:40:16.0321410Z test/prototype/test_sparsity_utils.py::TestSparsityUtilFunctions::test_get_arg_info_from_tensor_fqn PASSED 2025-09-09T14:40:16.0322380Z test/prototype/test_sparsity_utils.py::TestSparsityUtilFunctions::test_get_arg_info_from_tensor_fqn_fail PASSED 2025-09-09T14:40:16.0323340Z test/prototype/test_sparsity_utils.py::TestSparsityUtilFunctions::test_module_to_fqn PASSED 2025-09-09T14:40:16.0324200Z test/prototype/test_sparsity_utils.py::TestSparsityUtilFunctions::test_module_to_fqn_fail PASSED 2025-09-09T14:40:16.0325081Z test/prototype/test_sparsity_utils.py::TestSparsityUtilFunctions::test_module_to_fqn_root PASSED 2025-09-09T14:40:16.0325866Z test/prototype/test_spinquant.py::test_spinquant_no_quantization[cpu] PASSED 2025-09-09T14:40:16.0326560Z test/prototype/test_spinquant.py::test_spinquant_no_quantization[cuda] PASSED 2025-09-09T14:40:16.0327410Z test/prototype/test_structured_sparsifier.py::TestSaliencyPruner::test_lstm_saliency_pruner_update_mask PASSED 2025-09-09T14:40:16.0328373Z test/prototype/test_structured_sparsifier.py::TestSaliencyPruner::test_saliency_pruner_update_mask PASSED 2025-09-09T14:40:16.0329324Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_complex_conv2d PASSED 2025-09-09T14:40:16.0330264Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_constructor PASSED 2025-09-09T14:40:16.0331203Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prepare_conv2d PASSED 2025-09-09T14:40:42.9440984Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prepare_linear PASSED 2025-09-09T14:40:42.9442154Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_conv2d_activation_conv2d PASSED 2025-09-09T14:40:42.9443233Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_conv2d_bias_conv2d PASSED 2025-09-09T14:40:42.9444273Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_conv2d_conv2d PASSED 2025-09-09T14:40:42.9445314Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_conv2d_padding_conv2d PASSED 2025-09-09T14:40:42.9446367Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_conv2d_pool_conv2d PASSED 2025-09-09T14:40:42.9447447Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_linear_activation_linear PASSED 2025-09-09T14:40:42.9448533Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_linear_bias_linear PASSED 2025-09-09T14:40:42.9450050Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_linear_linear PASSED 2025-09-09T14:40:42.9451162Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_lstm_layernorm_linear_multiple_layer PASSED 2025-09-09T14:40:42.9452679Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_lstm_layernorm_linear_single_layer PASSED 2025-09-09T14:40:42.9454095Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_lstm_linear_multiple_layer PASSED 2025-09-09T14:40:42.9455458Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_lstm_linear_single_layer PASSED 2025-09-09T14:40:42.9456859Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_step_conv2d PASSED 2025-09-09T14:40:42.9458022Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_step_linear PASSED 2025-09-09T14:40:42.9459120Z test/prototype/test_structured_sparsifier.py::TestFPGMPruner::test_compute_distance PASSED 2025-09-09T14:40:42.9460122Z test/prototype/test_structured_sparsifier.py::TestFPGMPruner::test_update_mask PASSED 2025-09-09T14:40:42.9461262Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_attention_block SKIPPED 2025-09-09T14:40:42.9462479Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_conv2d SKIPPED 2025-09-09T14:40:42.9463705Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_conv2d_binary SKIPPED 2025-09-09T14:40:42.9464955Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_conv2d_binary2 SKIPPED 2025-09-09T14:40:42.9466248Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_dynamic_quant_linear SKIPPED 2025-09-09T14:40:42.9467579Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_filter_linear_recipe SKIPPED 2025-09-09T14:40:42.9468822Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_linear SKIPPED 2025-09-09T14:40:42.9470028Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_linear_unary SKIPPED 2025-09-09T14:40:42.9471349Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_linear_unary_dynamic SKIPPED 2025-09-09T14:40:42.9472700Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_linear_unary_dynamic_qat SKIPPED 2025-09-09T14:40:42.9474019Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_linear_unary_qat SKIPPED 2025-09-09T14:40:42.9475264Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_qat_conv2d SKIPPED 2025-09-09T14:40:42.9476518Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_qat_conv2d_binary SKIPPED 2025-09-09T14:40:42.9477825Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_qat_conv2d_binary2 SKIPPED 2025-09-09T14:40:42.9479149Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_qat_dynamic_quant_linear SKIPPED 2025-09-09T14:40:42.9480582Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_set_module_name_and_module_type_case1 SKIPPED 2025-09-09T14:40:42.9482112Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_set_module_name_and_module_type_case2 SKIPPED 2025-09-09T14:40:42.9483782Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_set_module_name_and_module_type_with_mixed_configs SKIPPED 2025-09-09T14:40:42.9485267Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_set_module_name_qconfig SKIPPED 2025-09-09T14:40:42.9486801Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_set_module_name_qconfig_for_dynamic_quant SKIPPED 2025-09-09T14:40:42.9488320Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_set_module_name_qconfig_with_underscores SKIPPED 2025-09-09T14:40:42.9489800Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_set_module_name_with_mixed_configs SKIPPED 2025-09-09T14:40:42.9491114Z test/quantization/pt2e/test_duplicate_dq.py::TestDuplicateDQPass::test_avgpool_use_different_qconfig PASSED 2025-09-09T14:40:42.9492292Z test/quantization/pt2e/test_duplicate_dq.py::TestDuplicateDQPass::test_no_add_quant_duplicate_dq PASSED 2025-09-09T14:40:42.9493449Z test/quantization/pt2e/test_duplicate_dq.py::TestDuplicateDQPass::test_no_need_for_duplicate_dq PASSED 2025-09-09T14:40:42.9494572Z test/quantization/pt2e/test_duplicate_dq.py::TestDuplicateDQPass::test_simple_duplicate_dq PASSED 2025-09-09T14:40:42.9495625Z test/quantization/pt2e/test_graph_utils.py::TestGraphUtils::test_conv_bn_conv_relu PASSED 2025-09-09T14:40:42.9497042Z test/quantization/pt2e/test_graph_utils.py::TestGraphUtils::test_conv_bn_relu W0909 14:40:20.553927 937 site-packages/torch/fx/experimental/symbolic_shapes.py:2704] Failed to reduce inequalities: 1/2 2025-09-09T14:40:42.9498357Z PASSED 2025-09-09T14:40:42.9499438Z test/quantization/pt2e/test_graph_utils.py::TestGraphUtils::test_customized_equivalet_types_dict W0909 14:40:20.670188 937 site-packages/torch/fx/experimental/symbolic_shapes.py:2704] Failed to reduce inequalities: 1/2 2025-09-09T14:40:42.9500621Z PASSED 2025-09-09T14:40:42.9501431Z test/quantization/pt2e/test_metadata_porting.py::TestMetaDataPorting::test_metadata_porting_for_dq SKIPPED 2025-09-09T14:40:42.9502689Z test/quantization/pt2e/test_metadata_porting.py::TestMetaDataPorting::test_metadata_porting_for_dq_no_static_q PASSED 2025-09-09T14:40:42.9503960Z test/quantization/pt2e/test_metadata_porting.py::TestMetaDataPorting::test_metadata_porting_for_two_dq PASSED 2025-09-09T14:40:42.9505247Z test/quantization/pt2e/test_metadata_porting.py::TestMetaDataPorting::test_metadata_porting_with_no_quant_inbetween PASSED 2025-09-09T14:40:42.9506493Z test/quantization/pt2e/test_metadata_porting.py::TestMetaDataPorting::test_no_metadata_porting PASSED 2025-09-09T14:40:42.9507748Z test/quantization/pt2e/test_metadata_porting.py::TestMetaDataPorting::test_no_metadata_porting_through_unknown_ops PASSED 2025-09-09T14:40:42.9508997Z test/quantization/pt2e/test_metadata_porting.py::TestMetaDataPorting::test_simple_metadata_porting PASSED 2025-09-09T14:40:42.9510238Z test/quantization/pt2e/test_numeric_debugger.py::TestNumericDebuggerInfra::test_added_node_gets_unique_id SKIPPED 2025-09-09T14:40:42.9511464Z test/quantization/pt2e/test_numeric_debugger.py::TestNumericDebuggerInfra::test_control_flow SKIPPED 2025-09-09T14:40:42.9512690Z test/quantization/pt2e/test_numeric_debugger.py::TestNumericDebuggerInfra::test_copy_preserve_handle SKIPPED 2025-09-09T14:40:42.9513945Z test/quantization/pt2e/test_numeric_debugger.py::TestNumericDebuggerInfra::test_deepcopy_preserve_handle SKIPPED 2025-09-09T14:40:42.9515265Z test/quantization/pt2e/test_numeric_debugger.py::TestNumericDebuggerInfra::test_prepare_for_propagation_comparison SKIPPED 2025-09-09T14:40:42.9516587Z test/quantization/pt2e/test_numeric_debugger.py::TestNumericDebuggerInfra::test_re_export_preserve_handle SKIPPED 2025-09-09T14:40:42.9517965Z test/quantization/pt2e/test_numeric_debugger.py::TestNumericDebuggerInfra::test_run_decompositions_map_handle_to_new_nodes SKIPPED 2025-09-09T14:40:42.9519574Z test/quantization/pt2e/test_numeric_debugger.py::TestNumericDebuggerInfra::test_run_decompositions_same_handle_id SKIPPED 2025-09-09T14:40:42.9520937Z test/quantization/pt2e/test_numeric_debugger.py::TestNumericDebuggerInfra::test_simple SKIPPED 2025-09-09T14:40:42.9522191Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_allow_exported_model_train_eval PASSED 2025-09-09T14:40:42.9523238Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_allow_exported_model_train_eval_idempotent PASSED 2025-09-09T14:41:06.0235803Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_allow_implicit_sharing PASSED 2025-09-09T14:41:06.0236710Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_chunked_bn_fusion PASSED 2025-09-09T14:41:06.0240064Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_composable_quantizer_linear_conv [W909 14:40:43.452778035 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:41:06.0242670Z [W909 14:40:43.452813466 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:41:06.0244879Z [W909 14:40:43.452826556 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:41:06.0247039Z [W909 14:40:43.452842607 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:41:06.0249200Z [W909 14:40:43.452865547 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:41:06.0251360Z [W909 14:40:43.452877217 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:41:06.0253513Z [W909 14:40:43.452900778 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:41:06.0255668Z [W909 14:40:43.452919058 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:41:06.0258363Z [W909 14:40:43.452968529 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:41:06.0260586Z [W909 14:40:43.452989390 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:41:06.0262919Z [W909 14:40:43.452996590 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:41:06.0265084Z [W909 14:40:43.453012220 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:41:06.0267243Z [W909 14:40:43.453035951 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:41:06.0269451Z [W909 14:40:43.453048231 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:41:06.0271612Z [W909 14:40:43.453093272 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:41:06.0273774Z [W909 14:40:43.453112503 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:41:06.0275077Z PASSED 2025-09-09T14:41:06.0275693Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_composable_quantizer_throw PASSED 2025-09-09T14:41:06.0276714Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_composable_quantizer_transform_for_annotation PASSED 2025-09-09T14:41:06.0277735Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_constant_prop_preserve_metadata PASSED 2025-09-09T14:41:06.0278623Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_conv3d_bn_relu PASSED 2025-09-09T14:41:06.0279479Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_conv_padding_bn_relu PASSED 2025-09-09T14:41:06.0280376Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_conv_transpose3d_bn_relu PASSED 2025-09-09T14:41:06.0281272Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_conv_transpose_bn_relu PASSED 2025-09-09T14:41:06.0282123Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_derived_qspec PASSED 2025-09-09T14:41:06.0282990Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_derived_qspec_per_channel PASSED 2025-09-09T14:41:06.0283875Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_disallow_eval_train PASSED 2025-09-09T14:41:06.0284861Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_dont_fold_other_constant PASSED 2025-09-09T14:41:06.0285816Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_embedding_conv_linear_quantization PASSED 2025-09-09T14:41:06.0286813Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_embedding_quantizer PASSED 2025-09-09T14:41:06.0287742Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_fixed_qparams_qspec_observer_dedup PASSED 2025-09-09T14:41:06.0288689Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_fixed_qparams_qspec_ptq PASSED 2025-09-09T14:41:06.0289620Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_fixed_qparams_qspec_qat PASSED 2025-09-09T14:41:06.0290538Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_fold_all_ops_before_quantize PASSED 2025-09-09T14:41:06.0291409Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_fold_quantize PASSED 2025-09-09T14:41:06.0292280Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_fold_quantize_per_channel PASSED 2025-09-09T14:41:06.0293205Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_groupwise_per_channel_quant PASSED 2025-09-09T14:41:06.0294113Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_input_edge_sanity_check PASSED 2025-09-09T14:41:06.0295000Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_max_pool2d_quantizer PASSED 2025-09-09T14:41:06.0295933Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_model_is_exported PASSED 2025-09-09T14:41:29.6860921Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_move_exported_model_bn_device_cpu PASSED 2025-09-09T14:41:29.6861979Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_move_exported_model_bn_device_cuda PASSED 2025-09-09T14:41:29.6862976Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_move_exported_model_dropout PASSED 2025-09-09T14:41:29.6863954Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_move_exported_model_dropout_inplace PASSED 2025-09-09T14:41:29.6864941Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_multi_users_without_output_observer PASSED 2025-09-09T14:41:29.6865861Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_observer_callback PASSED 2025-09-09T14:41:29.6866791Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_prepare_obs_or_fq_callback PASSED 2025-09-09T14:41:29.6867700Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_preserve_nn_module_stack PASSED 2025-09-09T14:41:29.6868669Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_quantization_dtype_bfloat16_float8_e4m3fn PASSED 2025-09-09T14:41:29.6869706Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_quantization_dtype_bfloat16_float8_e5m2 PASSED 2025-09-09T14:41:29.6870708Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_quantization_dtype_bfloat16_int16 PASSED 2025-09-09T14:41:29.6871709Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_quantization_dtype_float32_float8_e4m3fn PASSED 2025-09-09T14:41:29.6872726Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_quantization_dtype_float32_float8_e5m2 PASSED 2025-09-09T14:41:29.6873703Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_quantization_dtype_float32_int16 PASSED 2025-09-09T14:41:29.6874552Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_quantize_in_place_ops input_act1 is a node 2025-09-09T14:41:29.6875131Z PASSED 2025-09-09T14:41:29.6876082Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_reentrant PASSED 2025-09-09T14:41:29.6876931Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_save_load PASSED 2025-09-09T14:41:29.6877890Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_shared_qspec PASSED 2025-09-09T14:41:29.6878755Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_shared_qspec_transitivity PASSED 2025-09-09T14:41:29.6879694Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_shared_qspec_transitivity_case_2 PASSED 2025-09-09T14:41:29.6880583Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_simple_quantizer PASSED 2025-09-09T14:41:29.6881371Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_speed PASSED 2025-09-09T14:41:29.6882197Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_transform_for_annotation PASSED 2025-09-09T14:41:29.6883139Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_wo_annotate_conv_output_quantizer PASSED 2025-09-09T14:41:29.6884127Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2EAffineQuantization::test_channel_group_quantization prepared model: GraphModule( 2025-09-09T14:41:29.6884818Z (linear): Module() 2025-09-09T14:41:29.6885146Z (activation_post_process_1): AffineQuantizedMinMaxObserver() 2025-09-09T14:41:29.6885605Z (activation_post_process_0): AffineQuantizedMinMaxObserver() 2025-09-09T14:41:29.6885962Z ) 2025-09-09T14:41:29.6886065Z 2025-09-09T14:41:29.6886070Z 2025-09-09T14:41:29.6886074Z 2025-09-09T14:41:29.6886170Z def forward(self, x): 2025-09-09T14:41:29.6886461Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:41:29.6886816Z linear_weight = self.linear.weight 2025-09-09T14:41:29.6887305Z activation_post_process_1 = self.activation_post_process_1(linear_weight); linear_weight = None 2025-09-09T14:41:29.6887804Z linear_bias = self.linear.bias 2025-09-09T14:41:29.6888198Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:41:29.6889103Z linear = torch.ops.aten.linear.default(activation_post_process_0, activation_post_process_1, linear_bias); activation_post_process_0 = activation_post_process_1 = linear_bias = None 2025-09-09T14:41:29.6889952Z return pytree.tree_unflatten((linear,), self._out_spec) 2025-09-09T14:41:29.6890276Z 2025-09-09T14:41:29.6890569Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:41:29.6890943Z quantized model GraphModule( 2025-09-09T14:41:29.6891204Z (linear): Module() 2025-09-09T14:41:29.6891415Z ) 2025-09-09T14:41:29.6891527Z 2025-09-09T14:41:29.6891531Z 2025-09-09T14:41:29.6891535Z 2025-09-09T14:41:29.6891624Z def forward(self, x): 2025-09-09T14:41:29.6891924Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:41:29.6892252Z _scale0 = self._scale0 2025-09-09T14:41:29.6892516Z _zero_point0 = self._zero_point0 2025-09-09T14:41:29.6892812Z quantize_affine = self._frozen_param0 2025-09-09T14:41:29.6893642Z dequantize_affine = torch.ops.torchao.dequantize_affine(quantize_affine, (1, 128), _scale0, _zero_point0, torch.uint8, 0, 255, output_dtype = torch.float32); quantize_affine = _scale0 = _zero_point0 = None 2025-09-09T14:41:29.6894471Z linear_bias = self.linear.bias 2025-09-09T14:41:29.6894750Z _scale1 = self._scale1 2025-09-09T14:41:29.6894999Z _zero_point1 = self._zero_point1 2025-09-09T14:41:29.6895526Z quantize_affine_1 = torch.ops.torchao.quantize_affine(x, (1, 128), _scale1, _zero_point1, torch.uint8, 0, 255); x = None 2025-09-09T14:41:29.6896916Z dequantize_affine_1 = torch.ops.torchao.dequantize_affine(quantize_affine_1, (1, 128), _scale1, _zero_point1, torch.uint8, 0, 255, output_dtype = torch.float32); quantize_affine_1 = _scale1 = _zero_point1 = None 2025-09-09T14:41:29.6898799Z linear = torch.ops.aten.linear.default(dequantize_affine_1, dequantize_affine, linear_bias); dequantize_affine_1 = dequantize_affine = linear_bias = None 2025-09-09T14:41:29.6899551Z return pytree.tree_unflatten((linear,), self._out_spec) 2025-09-09T14:41:29.6899988Z 2025-09-09T14:41:29.6900268Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:41:29.6900677Z PASSED 2025-09-09T14:41:29.6901416Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2EAffineQuantization::test_dynamic_affine_act_per_channel_weights PASSED 2025-09-09T14:41:29.6902553Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2EAffineQuantization::test_dynamic_per_tok_act_per_group_weights prepared model: GraphModule( 2025-09-09T14:41:29.6903272Z (linear): Module() 2025-09-09T14:41:29.6903599Z (activation_post_process_1): AffineQuantizedMinMaxObserver() 2025-09-09T14:41:29.6904089Z (activation_post_process_0): AffineQuantizedPlaceholderObserver() 2025-09-09T14:41:29.6904460Z ) 2025-09-09T14:41:29.6904559Z 2025-09-09T14:41:29.6904564Z 2025-09-09T14:41:29.6904574Z 2025-09-09T14:41:29.6904667Z def forward(self, x): 2025-09-09T14:41:29.6904956Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:41:29.6905313Z linear_weight = self.linear.weight 2025-09-09T14:41:29.6905867Z activation_post_process_1 = self.activation_post_process_1(linear_weight); linear_weight = None 2025-09-09T14:41:29.6906484Z linear_bias = self.linear.bias 2025-09-09T14:41:29.6906970Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:41:29.6907991Z linear = torch.ops.aten.linear.default(activation_post_process_0, activation_post_process_1, linear_bias); activation_post_process_0 = activation_post_process_1 = linear_bias = None 2025-09-09T14:41:29.6908847Z return pytree.tree_unflatten((linear,), self._out_spec) 2025-09-09T14:41:29.6909168Z 2025-09-09T14:41:29.6909453Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:41:29.6909828Z quantized model GraphModule( 2025-09-09T14:41:29.6910087Z (linear): Module() 2025-09-09T14:41:29.6910304Z ) 2025-09-09T14:41:29.6910405Z 2025-09-09T14:41:29.6910409Z 2025-09-09T14:41:29.6910418Z 2025-09-09T14:41:29.6910505Z def forward(self, x): 2025-09-09T14:41:29.6910798Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:41:29.6911124Z _scale0 = self._scale0 2025-09-09T14:41:29.6911377Z _zero_point0 = self._zero_point0 2025-09-09T14:41:29.6911664Z quantize_affine = self._frozen_param0 2025-09-09T14:41:29.6912888Z dequantize_affine = torch.ops.torchao.dequantize_affine(quantize_affine, (1, 128), _scale0, _zero_point0, torch.int8, -127, 127, output_dtype = torch.float32); quantize_affine = _scale0 = _zero_point0 = None 2025-09-09T14:41:29.6913909Z linear_bias = self.linear.bias 2025-09-09T14:41:29.6914535Z choose_qparams_affine = torch.ops.torchao.choose_qparams_affine(x, 'SYMMETRIC', (1, 128), torch.int8, -128, 127, None, None, None) 2025-09-09T14:41:29.6923456Z getitem = choose_qparams_affine[0] 2025-09-09T14:41:29.6923863Z getitem_1 = choose_qparams_affine[1]; choose_qparams_affine = None 2025-09-09T14:41:29.6924506Z quantize_affine_1 = torch.ops.torchao.quantize_affine(x, (1, 128), getitem, getitem_1, torch.int8, -128, 127); x = None 2025-09-09T14:41:29.6925604Z dequantize_affine_1 = torch.ops.torchao.dequantize_affine(quantize_affine_1, (1, 128), getitem, getitem_1, torch.int8, -128, 127, output_dtype = torch.float32); quantize_affine_1 = getitem = getitem_1 = None 2025-09-09T14:41:29.6926883Z linear = torch.ops.aten.linear.default(dequantize_affine_1, dequantize_affine, linear_bias); dequantize_affine_1 = dequantize_affine = linear_bias = None 2025-09-09T14:41:29.6927652Z return pytree.tree_unflatten((linear,), self._out_spec) 2025-09-09T14:41:29.6927986Z 2025-09-09T14:41:29.6928273Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:41:29.6928715Z PASSED 2025-09-09T14:41:29.6929501Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_fold_bn_erases_bn_node SKIPPED 2025-09-09T14:41:29.6930580Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_bn_bias_derived_qspec SKIPPED 2025-09-09T14:41:29.6931707Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_bn_fusion SKIPPED 2025-09-09T14:41:29.6932723Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_bn_fusion_cuda SKIPPED 2025-09-09T14:41:29.6933793Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_bn_fusion_literal_args SKIPPED 2025-09-09T14:41:29.6934882Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_bn_fusion_no_conv_bias SKIPPED 2025-09-09T14:42:25.6261630Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_bn_per_channel_weight_bias SKIPPED 2025-09-09T14:42:25.6263045Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_bn_relu_fusion SKIPPED 2025-09-09T14:42:25.6264376Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_bn_relu_fusion_cuda SKIPPED 2025-09-09T14:42:25.6265749Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_bn_relu_fusion_no_conv_bias SKIPPED 2025-09-09T14:42:25.6267060Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_no_bias SKIPPED 2025-09-09T14:42:25.6268321Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_transpose_bn SKIPPED 2025-09-09T14:42:25.6269628Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_transpose_bn_relu SKIPPED 2025-09-09T14:42:25.6270925Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_inplace_add_relu SKIPPED 2025-09-09T14:42:25.6272275Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_per_channel_weight_custom_dtype SKIPPED 2025-09-09T14:42:25.6273627Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_preserve_source_fn_stack SKIPPED 2025-09-09T14:42:25.6274937Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_update_shared_qspec SKIPPED 2025-09-09T14:42:25.6276207Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_fold_bn_erases_bn_node PASSED 2025-09-09T14:42:25.6277483Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_bn_bias_derived_qspec PASSED 2025-09-09T14:42:25.6278675Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_bn_fusion model pt2e: GraphModule( 2025-09-09T14:42:25.6279432Z (conv): Module() 2025-09-09T14:42:25.6279686Z (bn): Module() 2025-09-09T14:42:25.6280056Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:42:25.6281215Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:42:25.6282592Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:42:25.6283228Z ) 2025-09-09T14:42:25.6283560Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:42:25.6285175Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0017, 0.0019, 0.0024]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:42:25.6286797Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2148, -0.0992, -0.2048]), max_val=tensor([0.0771, 0.2459, 0.3011])) 2025-09-09T14:42:25.6287793Z ) 2025-09-09T14:42:25.6288132Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:42:25.6289283Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0111]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:42:25.6290657Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.4139524698257446, max_val=1.4139440059661865) 2025-09-09T14:42:25.6291294Z ) 2025-09-09T14:42:25.6291503Z ) 2025-09-09T14:42:25.6291624Z 2025-09-09T14:42:25.6291629Z 2025-09-09T14:42:25.6291634Z 2025-09-09T14:42:25.6291747Z def forward(self, x): 2025-09-09T14:42:25.6292095Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:42:25.6292510Z conv_weight = self.conv.weight 2025-09-09T14:42:25.6292848Z conv_bias = self.conv.bias 2025-09-09T14:42:25.6293169Z bn_weight = self.bn.weight 2025-09-09T14:42:25.6293474Z bn_bias = self.bn.bias 2025-09-09T14:42:25.6293791Z bn_running_mean = self.bn.running_mean 2025-09-09T14:42:25.6294156Z bn_running_var = self.bn.running_var 2025-09-09T14:42:25.6294564Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:42:25.6295105Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:42:25.6295914Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:42:25.6296571Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:42:25.6297046Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:42:25.6297767Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:42:25.6298306Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T14:42:25.6298926Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:42:25.6299630Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:42:25.6300382Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:42:25.6301601Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:42:25.6302696Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T14:42:25.6303349Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T14:42:25.6304091Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T14:42:25.6304789Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:42:25.6305862Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:42:25.6307015Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T14:42:25.6307756Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:42:25.6308244Z 2025-09-09T14:42:25.6308581Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:42:25.6309030Z model fx: GraphModule( 2025-09-09T14:42:25.6309414Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:42:25.6310767Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:42:25.6312158Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:42:25.6312910Z ) 2025-09-09T14:42:25.6313142Z (conv): ConvBn1d( 2025-09-09T14:42:25.6313416Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T14:42:25.6313919Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:42:25.6314536Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:42:25.6315735Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0017, 0.0019, 0.0024]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:42:25.6317228Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2148, -0.0992, -0.2048]), max_val=tensor([0.0771, 0.2459, 0.3011])) 2025-09-09T14:42:25.6317871Z ) 2025-09-09T14:42:25.6318052Z ) 2025-09-09T14:42:25.6318335Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:42:25.6319276Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0111]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:42:25.6320372Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.4139524698257446, max_val=1.4139440059661865) 2025-09-09T14:42:25.6320881Z ) 2025-09-09T14:42:25.6321056Z ) 2025-09-09T14:42:25.6321155Z 2025-09-09T14:42:25.6321160Z 2025-09-09T14:42:25.6321164Z 2025-09-09T14:42:25.6321250Z def forward(self, x): 2025-09-09T14:42:25.6321607Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:42:25.6322140Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:42:25.6322690Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:42:25.6323130Z return activation_post_process_1 2025-09-09T14:42:25.6323403Z 2025-09-09T14:42:25.6323689Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:42:25.6324055Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:42:25.6324319Z [0., 0., 0.], 2025-09-09T14:42:25.6324588Z [0., 0., 0.]]], grad_fn=) 2025-09-09T14:42:25.6324904Z converted model pt2e: GraphModule( 2025-09-09T14:42:25.6325177Z (conv): Module() 2025-09-09T14:42:25.6325382Z (bn): Module() 2025-09-09T14:42:25.6325585Z ) 2025-09-09T14:42:25.6325685Z 2025-09-09T14:42:25.6325689Z 2025-09-09T14:42:25.6325693Z 2025-09-09T14:42:25.6325780Z def forward(self, x): 2025-09-09T14:42:25.6326072Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:42:25.6326406Z conv_bias = self.conv.bias 2025-09-09T14:42:25.6326723Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:42:25.6327426Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:42:25.6328659Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:42:25.6329700Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:42:25.6330165Z _scale_0 = self._scale_0 2025-09-09T14:42:25.6330429Z _zero_point_0 = self._zero_point_0 2025-09-09T14:42:25.6330736Z quantize_per_channel = self._frozen_param0 2025-09-09T14:42:41.8827499Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T14:42:41.8829232Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T14:42:41.8830915Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_2, 0.011089790612459183, 0, -128, 127, torch.int8); conv1d_2 = None 2025-09-09T14:42:41.8832532Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.011089790612459183, 0, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:42:41.8833789Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T14:42:41.8834302Z 2025-09-09T14:42:41.8834640Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:42:41.8835100Z onverted model fx: GraphModule( 2025-09-09T14:42:41.8835558Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T14:42:41.8836013Z ) 2025-09-09T14:42:41.8836142Z 2025-09-09T14:42:41.8836148Z 2025-09-09T14:42:41.8836153Z 2025-09-09T14:42:41.8836261Z def forward(self, x): 2025-09-09T14:42:41.8837013Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:42:41.8838552Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:42:41.8839850Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:42:41.8840906Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.011089790612459183, 0, -128, 127, torch.int8); conv = None 2025-09-09T14:42:41.8842494Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.011089790612459183, 0, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:42:41.8843597Z return dequantize_per_tensor_default_1 2025-09-09T14:42:41.8843936Z 2025-09-09T14:42:41.8844265Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:42:41.8844715Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:42:41.8845003Z [0., 0., 0.], 2025-09-09T14:42:41.8845256Z [0., 0., 0.]]]) 2025-09-09T14:42:41.8845540Z model pt2e: GraphModule( 2025-09-09T14:42:41.8845815Z (conv): Module() 2025-09-09T14:42:41.8846087Z (bn): Module() 2025-09-09T14:42:41.8846457Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:42:41.8847624Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:42:41.8849003Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:42:41.8849668Z ) 2025-09-09T14:42:41.8850028Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:42:41.8851199Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:42:41.8852575Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.214811772108078, max_val=0.30109599232673645) 2025-09-09T14:42:41.8853213Z ) 2025-09-09T14:42:41.8853534Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:42:41.8854782Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0111]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:42:41.8856252Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.4139524698257446, max_val=1.4138529300689697) 2025-09-09T14:42:41.8856977Z ) 2025-09-09T14:42:41.8857183Z ) 2025-09-09T14:42:41.8857299Z 2025-09-09T14:42:41.8857304Z 2025-09-09T14:42:41.8857309Z 2025-09-09T14:42:41.8857412Z def forward(self, x): 2025-09-09T14:42:41.8857756Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:42:41.8858167Z conv_weight = self.conv.weight 2025-09-09T14:42:41.8858493Z conv_bias = self.conv.bias 2025-09-09T14:42:41.8858802Z bn_weight = self.bn.weight 2025-09-09T14:42:41.8859102Z bn_bias = self.bn.bias 2025-09-09T14:42:41.8859416Z bn_running_mean = self.bn.running_mean 2025-09-09T14:42:41.8859777Z bn_running_var = self.bn.running_var 2025-09-09T14:42:41.8860176Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:42:41.8860712Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:42:41.8861446Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:42:41.8862107Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:42:41.8862580Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:42:41.8863071Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:42:41.8863604Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T14:42:41.8864206Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:42:41.8864901Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:42:41.8865650Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:42:41.8866867Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:42:41.8867963Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T14:42:41.8868607Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T14:42:41.8869339Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T14:42:41.8870030Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:42:41.8871099Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:42:41.8872254Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T14:42:41.8872990Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:42:41.8873472Z 2025-09-09T14:42:41.8873807Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:42:41.8874256Z model fx: GraphModule( 2025-09-09T14:42:41.8874634Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:42:41.8875798Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:42:41.8877164Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:42:41.8877790Z ) 2025-09-09T14:42:41.8878009Z (conv): ConvBn1d( 2025-09-09T14:42:41.8878272Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T14:42:41.8878765Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:42:41.8879445Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:42:41.8880605Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:42:41.8882073Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.214811772108078, max_val=0.30109599232673645) 2025-09-09T14:42:41.8882711Z ) 2025-09-09T14:42:41.8882946Z ) 2025-09-09T14:42:41.8883300Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:42:41.8884454Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0111]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:42:41.8885556Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.4139524698257446, max_val=1.4138529300689697) 2025-09-09T14:42:41.8886064Z ) 2025-09-09T14:42:41.8886230Z ) 2025-09-09T14:42:41.8886326Z 2025-09-09T14:42:41.8886330Z 2025-09-09T14:42:41.8886341Z 2025-09-09T14:42:41.8886424Z def forward(self, x): 2025-09-09T14:42:41.8886787Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:42:41.8887317Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:42:41.8887851Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:42:41.8888279Z return activation_post_process_1 2025-09-09T14:42:41.8888532Z 2025-09-09T14:42:41.8888819Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:42:41.8889188Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:42:41.8889427Z [0., 0., 0.], 2025-09-09T14:42:41.8889664Z [0., 0., 0.]]], grad_fn=) 2025-09-09T14:42:41.8889967Z converted model pt2e: GraphModule( 2025-09-09T14:42:41.8890234Z (conv): Module() 2025-09-09T14:42:41.8890431Z (bn): Module() 2025-09-09T14:42:41.8890628Z ) 2025-09-09T14:42:41.8890727Z 2025-09-09T14:42:41.8890737Z 2025-09-09T14:42:41.8890740Z 2025-09-09T14:42:41.8890825Z def forward(self, x): 2025-09-09T14:42:41.8891111Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:42:41.8891440Z conv_bias = self.conv.bias 2025-09-09T14:42:41.8891746Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:42:41.8892466Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:42:41.8893699Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:42:41.8894754Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:42:41.8895245Z quantize_per_tensor = self._frozen_param0 2025-09-09T14:43:05.4822883Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.002370834583416581, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T14:43:05.4825791Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T14:43:05.4827212Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_2, 0.01108943298459053, 0, -128, 127, torch.int8); conv1d_2 = None 2025-09-09T14:43:05.4828510Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.01108943298459053, 0, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:43:05.4829758Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:43:05.4830183Z 2025-09-09T14:43:05.4830466Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:43:05.4830987Z onverted model fx: GraphModule( 2025-09-09T14:43:05.4831361Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T14:43:05.4831735Z ) 2025-09-09T14:43:05.4831836Z 2025-09-09T14:43:05.4831841Z 2025-09-09T14:43:05.4831844Z 2025-09-09T14:43:05.4831939Z def forward(self, x): 2025-09-09T14:43:05.4832552Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:43:05.4833792Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:43:05.4834802Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:43:05.4835642Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.01108943298459053, 0, -128, 127, torch.int8); conv = None 2025-09-09T14:43:05.4836909Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.01108943298459053, 0, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:43:05.4837775Z return dequantize_per_tensor_default_1 2025-09-09T14:43:05.4838052Z 2025-09-09T14:43:05.4838337Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:43:05.4838701Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:43:05.4838939Z [0., 0., 0.], 2025-09-09T14:43:05.4839152Z [0., 0., 0.]]]) 2025-09-09T14:43:05.4839580Z PASSED 2025-09-09T14:43:05.4840179Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_bn_fusion_cuda model pt2e: GraphModule( 2025-09-09T14:43:05.4840808Z (conv): Module() 2025-09-09T14:43:05.4841013Z (bn): Module() 2025-09-09T14:43:05.4841330Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:43:05.4842450Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0104], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:43:05.4843727Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:43:05.4844241Z ) 2025-09-09T14:43:05.4844521Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:43:05.4845698Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0022, 0.0020, 0.0022], device='cuda:0'), zero_point=tensor([0, 0, 0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:43:05.4847258Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2799, -0.2557, -0.2618], device='cuda:0'), max_val=tensor([0.1970, 0.2308, 0.2775], device='cuda:0')) 2025-09-09T14:43:05.4848000Z ) 2025-09-09T14:43:05.4848286Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:43:05.4849403Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0110], device='cuda:0'), zero_point=tensor([-1], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:43:05.4850675Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3961666822433472, max_val=1.4123179912567139) 2025-09-09T14:43:05.4851194Z ) 2025-09-09T14:43:05.4851451Z ) 2025-09-09T14:43:05.4851558Z 2025-09-09T14:43:05.4851563Z 2025-09-09T14:43:05.4851567Z 2025-09-09T14:43:05.4851654Z def forward(self, x): 2025-09-09T14:43:05.4852031Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:43:05.4852372Z conv_weight = self.conv.weight 2025-09-09T14:43:05.4852659Z conv_bias = self.conv.bias 2025-09-09T14:43:05.4852918Z bn_weight = self.bn.weight 2025-09-09T14:43:05.4853176Z bn_bias = self.bn.bias 2025-09-09T14:43:05.4853440Z bn_running_mean = self.bn.running_mean 2025-09-09T14:43:05.4853756Z bn_running_var = self.bn.running_var 2025-09-09T14:43:05.4854093Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:43:05.4854539Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:43:05.4855132Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:43:05.4855664Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:43:05.4856145Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:43:05.4856558Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:43:05.4857019Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T14:43:05.4857562Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:43:05.4858128Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:43:05.4858746Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:43:05.4859714Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:43:05.4860597Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T14:43:05.4861133Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T14:43:05.4861707Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T14:43:05.4862267Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:43:05.4863132Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:43:05.4864059Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T14:43:05.4864656Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:43:05.4865046Z 2025-09-09T14:43:05.4865334Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:43:05.4865703Z model fx: GraphModule( 2025-09-09T14:43:05.4866034Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:43:05.4867145Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0104], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:43:05.4868418Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:43:05.4868929Z ) 2025-09-09T14:43:05.4869109Z (conv): ConvBn1d( 2025-09-09T14:43:05.4869339Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T14:43:05.4869743Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:43:05.4870213Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:43:05.4871433Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0022, 0.0020, 0.0022], device='cuda:0'), zero_point=tensor([0, 0, 0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:43:05.4873066Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2799, -0.2557, -0.2618], device='cuda:0'), max_val=tensor([0.1970, 0.2308, 0.2775], device='cuda:0')) 2025-09-09T14:43:05.4873802Z ) 2025-09-09T14:43:05.4873976Z ) 2025-09-09T14:43:05.4874259Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:43:05.4875367Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0110], device='cuda:0'), zero_point=tensor([-1], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:43:05.4876647Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3961666822433472, max_val=1.4123179912567139) 2025-09-09T14:43:05.4877169Z ) 2025-09-09T14:43:05.4877339Z ) 2025-09-09T14:43:05.4877444Z 2025-09-09T14:43:05.4877448Z 2025-09-09T14:43:05.4877457Z 2025-09-09T14:43:05.4877544Z def forward(self, x): 2025-09-09T14:43:05.4877895Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:43:05.4878431Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:43:05.4878986Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:43:05.4879415Z return activation_post_process_1 2025-09-09T14:43:05.4879685Z 2025-09-09T14:43:05.4879963Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:43:05.4880337Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:43:05.4880572Z [0., 0., 0.], 2025-09-09T14:43:05.4880848Z [0., 0., 0.]]], device='cuda:0', grad_fn=) 2025-09-09T14:43:05.4881188Z converted model pt2e: GraphModule( 2025-09-09T14:43:05.4881459Z (conv): Module() 2025-09-09T14:43:05.4881677Z (bn): Module() 2025-09-09T14:43:05.4881875Z ) 2025-09-09T14:43:05.4881981Z 2025-09-09T14:43:05.4881985Z 2025-09-09T14:43:05.4881989Z 2025-09-09T14:43:05.4882084Z def forward(self, x): 2025-09-09T14:43:08.1918981Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:43:08.1919349Z conv_bias = self.conv.bias 2025-09-09T14:43:08.1919720Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:43:08.1920628Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933931648731, 0, -128, 127, torch.int8); x = None 2025-09-09T14:43:08.1921998Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933931648731, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:43:08.1923055Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:43:08.1923523Z _scale_0 = self._scale_0 2025-09-09T14:43:08.1923790Z _zero_point_0 = self._zero_point_0 2025-09-09T14:43:08.1924106Z quantize_per_channel = self._frozen_param0 2025-09-09T14:43:08.1924986Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T14:43:08.1926336Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T14:43:08.1927533Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_2, 0.011013665236532688, -1, -128, 127, torch.int8); conv1d_2 = None 2025-09-09T14:43:08.1930234Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.011013665236532688, -1, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:43:08.1931365Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T14:43:08.1931773Z 2025-09-09T14:43:08.1932073Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:43:08.1932451Z onverted model fx: GraphModule( 2025-09-09T14:43:08.1932832Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T14:43:08.1933205Z ) 2025-09-09T14:43:08.1933306Z 2025-09-09T14:43:08.1933310Z 2025-09-09T14:43:08.1933314Z 2025-09-09T14:43:08.1933400Z def forward(self, x): 2025-09-09T14:43:08.1934017Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933931648731, 0, -128, 127, torch.int8); x = None 2025-09-09T14:43:08.1943082Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933931648731, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:43:08.1944114Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:43:08.1944995Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.011013665236532688, -1, -128, 127, torch.int8); conv = None 2025-09-09T14:43:08.1946280Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.011013665236532688, -1, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:43:08.1947165Z return dequantize_per_tensor_default_1 2025-09-09T14:43:08.1947449Z 2025-09-09T14:43:08.1947736Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:43:08.1948110Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:43:08.1948347Z [0., 0., 0.], 2025-09-09T14:43:08.1948569Z [0., 0., 0.]]], device='cuda:0') 2025-09-09T14:43:08.1948845Z model pt2e: GraphModule( 2025-09-09T14:43:08.1949073Z (conv): Module() 2025-09-09T14:43:08.1949281Z (bn): Module() 2025-09-09T14:43:08.1949577Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:43:08.1950689Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0104], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:43:08.1951965Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:43:08.1952464Z ) 2025-09-09T14:43:08.1952743Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:43:08.1953865Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0022], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:43:08.1955141Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.2799264192581177, max_val=0.27745386958122253) 2025-09-09T14:43:08.1955653Z ) 2025-09-09T14:43:08.1955924Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:43:08.1957059Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0110], device='cuda:0'), zero_point=tensor([-1], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:43:08.1958326Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3980178833007812, max_val=1.4123179912567139) 2025-09-09T14:43:08.1958831Z ) 2025-09-09T14:43:08.1959004Z ) 2025-09-09T14:43:08.1959217Z 2025-09-09T14:43:08.1959222Z 2025-09-09T14:43:08.1959226Z 2025-09-09T14:43:08.1959320Z def forward(self, x): 2025-09-09T14:43:08.1959602Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:43:08.1960022Z conv_weight = self.conv.weight 2025-09-09T14:43:08.1960293Z conv_bias = self.conv.bias 2025-09-09T14:43:08.1960552Z bn_weight = self.bn.weight 2025-09-09T14:43:08.1960795Z bn_bias = self.bn.bias 2025-09-09T14:43:08.1961057Z bn_running_mean = self.bn.running_mean 2025-09-09T14:43:08.1961355Z bn_running_var = self.bn.running_var 2025-09-09T14:43:08.1961686Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:43:08.1962122Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:43:08.1962699Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:43:08.1963222Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:43:08.1963613Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:43:08.1964021Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:43:08.1964459Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T14:43:08.1964955Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:43:08.1965517Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:43:08.1966130Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:43:08.1967101Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:43:08.1967979Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T14:43:08.1968523Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T14:43:08.1969100Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T14:43:08.1969650Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:43:08.1970524Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:43:08.1971450Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T14:43:08.1972047Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:43:08.1972437Z 2025-09-09T14:43:08.1972717Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:43:08.1973083Z model fx: GraphModule( 2025-09-09T14:43:08.1973402Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:43:08.1974525Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0104], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:43:08.1975800Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:43:08.1976363Z ) 2025-09-09T14:43:08.1976549Z (conv): ConvBn1d( 2025-09-09T14:43:08.1976769Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T14:43:08.1977181Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:43:08.1977645Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:43:08.1978845Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0022], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:43:08.1980210Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.2799264192581177, max_val=0.27745386958122253) 2025-09-09T14:43:08.1980720Z ) 2025-09-09T14:43:08.1980901Z ) 2025-09-09T14:43:08.1981173Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:43:08.1982283Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0110], device='cuda:0'), zero_point=tensor([-1], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:43:08.1983554Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3980178833007812, max_val=1.4123179912567139) 2025-09-09T14:43:08.1984060Z ) 2025-09-09T14:43:08.1984231Z ) 2025-09-09T14:43:08.1984327Z 2025-09-09T14:43:08.1984338Z 2025-09-09T14:43:08.1984342Z 2025-09-09T14:43:08.1984427Z def forward(self, x): 2025-09-09T14:43:08.1984779Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:43:31.8170139Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:43:31.8171017Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:43:31.8171558Z return activation_post_process_1 2025-09-09T14:43:31.8171881Z 2025-09-09T14:43:31.8172218Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:43:31.8172675Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:43:31.8172963Z [0., 0., 0.], 2025-09-09T14:43:31.8173286Z [0., 0., 0.]]], device='cuda:0', grad_fn=) 2025-09-09T14:43:31.8173692Z converted model pt2e: GraphModule( 2025-09-09T14:43:31.8174025Z (conv): Module() 2025-09-09T14:43:31.8174272Z (bn): Module() 2025-09-09T14:43:31.8174512Z ) 2025-09-09T14:43:31.8174652Z 2025-09-09T14:43:31.8174658Z 2025-09-09T14:43:31.8174662Z 2025-09-09T14:43:31.8174777Z def forward(self, x): 2025-09-09T14:43:31.8175133Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:43:31.8175550Z conv_bias = self.conv.bias 2025-09-09T14:43:31.8175999Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:43:31.8176887Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933931648731, 0, -128, 127, torch.int8); x = None 2025-09-09T14:43:31.8178425Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933931648731, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:43:31.8179722Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:43:31.8180332Z quantize_per_tensor = self._frozen_param0 2025-09-09T14:43:31.8181316Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.002204145072028041, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T14:43:31.8182889Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T14:43:31.8184376Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_2, 0.01102092582732439, -1, -128, 127, torch.int8); conv1d_2 = None 2025-09-09T14:43:31.8185982Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.01102092582732439, -1, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:43:31.8187241Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:43:31.8187747Z 2025-09-09T14:43:31.8188363Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:43:31.8188843Z onverted model fx: GraphModule( 2025-09-09T14:43:31.8189470Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T14:43:31.8189936Z ) 2025-09-09T14:43:31.8190060Z 2025-09-09T14:43:31.8190065Z 2025-09-09T14:43:31.8190070Z 2025-09-09T14:43:31.8190181Z def forward(self, x): 2025-09-09T14:43:31.8190951Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933931648731, 0, -128, 127, torch.int8); x = None 2025-09-09T14:43:31.8192506Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933931648731, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:43:31.8193836Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:43:31.8194859Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.01102092582732439, -1, -128, 127, torch.int8); conv = None 2025-09-09T14:43:31.8196130Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.01102092582732439, -1, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:43:31.8197019Z return dequantize_per_tensor_default_1 2025-09-09T14:43:31.8197556Z 2025-09-09T14:43:31.8197870Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:43:31.8198291Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:43:31.8198545Z [0., 0., 0.], 2025-09-09T14:43:31.8198792Z [0., 0., 0.]]], device='cuda:0') 2025-09-09T14:43:31.8199302Z PASSED 2025-09-09T14:43:31.8199940Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_bn_fusion_literal_args model pt2e: GraphModule( 2025-09-09T14:43:31.8200607Z (conv): Module() 2025-09-09T14:43:31.8200821Z (bn): Module() 2025-09-09T14:43:31.8201139Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:43:31.8202089Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0087]), zero_point=tensor([57], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:43:31.8203200Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.611703872680664, max_val=0.6104744076728821) 2025-09-09T14:43:31.8203716Z ) 2025-09-09T14:43:31.8203997Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:43:31.8205007Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024, 0.0022, 0.0021]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:43:31.8206304Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.3119, -0.2799, -0.2618]), max_val=tensor([0.1970, 0.1855, 0.2308])) 2025-09-09T14:43:31.8206961Z ) 2025-09-09T14:43:31.8207248Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:43:31.8208202Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0119]), zero_point=tensor([-46], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:43:31.8209308Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.9784814715385437, max_val=2.047511577606201) 2025-09-09T14:43:31.8209817Z ) 2025-09-09T14:43:31.8209998Z ) 2025-09-09T14:43:31.8210099Z 2025-09-09T14:43:31.8210103Z 2025-09-09T14:43:31.8210107Z 2025-09-09T14:43:31.8210201Z def forward(self, x): 2025-09-09T14:43:31.8210490Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:43:31.8210841Z conv_weight = self.conv.weight 2025-09-09T14:43:31.8211269Z conv_bias = self.conv.bias 2025-09-09T14:43:31.8211538Z bn_weight = self.bn.weight 2025-09-09T14:43:31.8211792Z bn_bias = self.bn.bias 2025-09-09T14:43:31.8212163Z bn_running_mean = self.bn.running_mean 2025-09-09T14:43:31.8212466Z bn_running_var = self.bn.running_var 2025-09-09T14:43:31.8212807Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:43:31.8213257Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:43:31.8213846Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:43:31.8214383Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:43:31.8214780Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:43:31.8215200Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:43:31.8215642Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T14:43:31.8216210Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:43:31.8216787Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:43:31.8217409Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:43:31.8218398Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like, [2], [4]); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:43:31.8219306Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T14:43:31.8219851Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T14:43:31.8220430Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T14:43:31.8220982Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:43:31.8221862Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:43:31.8222797Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T14:43:31.8223404Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:43:31.8223806Z 2025-09-09T14:43:31.8224098Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:43:31.8224471Z model fx: GraphModule( 2025-09-09T14:43:31.8224798Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:43:31.8225802Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0087]), zero_point=tensor([57], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:43:31.8226908Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.611703872680664, max_val=0.6104744076728821) 2025-09-09T14:43:31.8227423Z ) 2025-09-09T14:43:31.8227621Z (conv): ConvBn1d( 2025-09-09T14:43:31.8227882Z 3, 3, kernel_size=(3,), stride=(2,), padding=(4,) 2025-09-09T14:43:31.8228331Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:43:31.8228802Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:43:31.8229776Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024, 0.0022, 0.0021]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:43:31.8231078Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.3119, -0.2799, -0.2618]), max_val=tensor([0.1970, 0.1855, 0.2308])) 2025-09-09T14:43:31.8231727Z ) 2025-09-09T14:43:31.8232025Z ) 2025-09-09T14:43:31.8232312Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:43:50.1485348Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0119]), zero_point=tensor([-46], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:43:50.1487133Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.9784814715385437, max_val=2.047511577606201) 2025-09-09T14:43:50.1487786Z ) 2025-09-09T14:43:50.1487997Z ) 2025-09-09T14:43:50.1488117Z 2025-09-09T14:43:50.1488122Z 2025-09-09T14:43:50.1488127Z 2025-09-09T14:43:50.1488235Z def forward(self, x): 2025-09-09T14:43:50.1488673Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:43:50.1489338Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:43:50.1490030Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:43:50.1490561Z return activation_post_process_1 2025-09-09T14:43:50.1490878Z 2025-09-09T14:43:50.1491227Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:43:50.1491687Z diff: tensor([[[0., 0., 0., 0., 0., 0.], 2025-09-09T14:43:50.1492018Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:43:50.1492362Z [0., 0., 0., 0., 0., 0.]]], grad_fn=) 2025-09-09T14:43:50.1492753Z converted model pt2e: GraphModule( 2025-09-09T14:43:50.1493075Z (conv): Module() 2025-09-09T14:43:50.1493318Z (bn): Module() 2025-09-09T14:43:50.1493555Z ) 2025-09-09T14:43:50.1493675Z 2025-09-09T14:43:50.1493680Z 2025-09-09T14:43:50.1493685Z 2025-09-09T14:43:50.1493788Z def forward(self, x): 2025-09-09T14:43:50.1494134Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:43:50.1494538Z conv_bias = self.conv.bias 2025-09-09T14:43:50.1494905Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:43:50.1495874Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.008714424446225166, 57, -128, 127, torch.int8); x = None 2025-09-09T14:43:50.1497620Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.008714424446225166, 57, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:43:50.1498939Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:43:50.1499513Z _scale_0 = self._scale_0 2025-09-09T14:43:50.1499831Z _zero_point_0 = self._zero_point_0 2025-09-09T14:43:50.1500201Z quantize_per_channel = self._frozen_param0 2025-09-09T14:43:50.1501313Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T14:43:50.1503016Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias, [2], [4]); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T14:43:50.1504544Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_2, 0.011866639368236065, -46, -128, 127, torch.int8); conv1d_2 = None 2025-09-09T14:43:50.1506176Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.011866639368236065, -46, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:43:50.1507452Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T14:43:50.1507959Z 2025-09-09T14:43:50.1508305Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:43:50.1508763Z onverted model fx: GraphModule( 2025-09-09T14:43:50.1509468Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(2,), padding=(4,)) 2025-09-09T14:43:50.1510008Z ) 2025-09-09T14:43:50.1510132Z 2025-09-09T14:43:50.1510138Z 2025-09-09T14:43:50.1510262Z 2025-09-09T14:43:50.1510368Z def forward(self, x): 2025-09-09T14:43:50.1511145Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.008714424446225166, 57, -128, 127, torch.int8); x = None 2025-09-09T14:43:50.1512754Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.008714424446225166, 57, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:43:50.1514022Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:43:50.1515097Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.011866639368236065, -46, -128, 127, torch.int8); conv = None 2025-09-09T14:43:50.1516817Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.011866639368236065, -46, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:43:50.1517757Z return dequantize_per_tensor_default_1 2025-09-09T14:43:50.1518035Z 2025-09-09T14:43:50.1518311Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:43:50.1518686Z diff: tensor([[[0., 0., 0., 0., 0., 0.], 2025-09-09T14:43:50.1518951Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:43:50.1519200Z [0., 0., 0., 0., 0., 0.]]]) 2025-09-09T14:43:50.1519468Z model pt2e: GraphModule( 2025-09-09T14:43:50.1519702Z (conv): Module() 2025-09-09T14:43:50.1519916Z (bn): Module() 2025-09-09T14:43:50.1520214Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:43:50.1521175Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0087]), zero_point=tensor([57], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:43:50.1522266Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.611703872680664, max_val=0.6104744076728821) 2025-09-09T14:43:50.1522784Z ) 2025-09-09T14:43:50.1523066Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:43:50.1524006Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:43:50.1525123Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.31192728877067566, max_val=0.23078496754169464) 2025-09-09T14:43:50.1525635Z ) 2025-09-09T14:43:50.1525915Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:43:50.1526853Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0119]), zero_point=tensor([-44], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:43:50.1527940Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.9980934262275696, max_val=2.047511577606201) 2025-09-09T14:43:50.1528453Z ) 2025-09-09T14:43:50.1528619Z ) 2025-09-09T14:43:50.1528721Z 2025-09-09T14:43:50.1528726Z 2025-09-09T14:43:50.1528730Z 2025-09-09T14:43:50.1528815Z def forward(self, x): 2025-09-09T14:43:50.1529108Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:43:50.1529441Z conv_weight = self.conv.weight 2025-09-09T14:43:50.1529717Z conv_bias = self.conv.bias 2025-09-09T14:43:50.1529969Z bn_weight = self.bn.weight 2025-09-09T14:43:50.1530220Z bn_bias = self.bn.bias 2025-09-09T14:43:50.1530474Z bn_running_mean = self.bn.running_mean 2025-09-09T14:43:50.1530779Z bn_running_var = self.bn.running_var 2025-09-09T14:43:50.1531108Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:43:50.1531647Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:43:50.1532238Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:43:50.1532840Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:43:50.1533236Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:43:50.1533646Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:43:50.1534092Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T14:43:50.1534588Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:43:50.1535154Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:43:50.1535821Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:43:50.1536809Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like, [2], [4]); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:43:50.1537704Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T14:43:50.1538231Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T14:43:50.1538805Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T14:43:50.1539355Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:43:50.1540218Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:43:50.1541154Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T14:43:50.1541800Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:43:50.1542191Z 2025-09-09T14:43:50.1542484Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:43:50.1542848Z model fx: GraphModule( 2025-09-09T14:43:50.1543173Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:43:50.1544108Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0087]), zero_point=tensor([57], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:43:50.1545205Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.611703872680664, max_val=0.6104744076728821) 2025-09-09T14:43:50.1545715Z ) 2025-09-09T14:43:50.1545895Z (conv): ConvBn1d( 2025-09-09T14:43:50.1546152Z 3, 3, kernel_size=(3,), stride=(2,), padding=(4,) 2025-09-09T14:43:50.1546586Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:43:50.1547057Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:43:50.1547970Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:44:11.7866492Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.31192728877067566, max_val=0.23078496754169464) 2025-09-09T14:44:11.7868634Z ) 2025-09-09T14:44:11.7868905Z ) 2025-09-09T14:44:11.7869253Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:44:11.7870447Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0119]), zero_point=tensor([-44], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:44:11.7872127Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.9980934262275696, max_val=2.047511577606201) 2025-09-09T14:44:11.7872780Z ) 2025-09-09T14:44:11.7873001Z ) 2025-09-09T14:44:11.7873126Z 2025-09-09T14:44:11.7873291Z 2025-09-09T14:44:11.7873296Z 2025-09-09T14:44:11.7873407Z def forward(self, x): 2025-09-09T14:44:11.7873849Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:44:11.7874513Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:44:11.7875184Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:44:11.7875717Z return activation_post_process_1 2025-09-09T14:44:11.7876037Z 2025-09-09T14:44:11.7876384Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:44:11.7876866Z diff: tensor([[[0., 0., 0., 0., 0., 0.], 2025-09-09T14:44:11.7877194Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:44:11.7877543Z [0., 0., 0., 0., 0., 0.]]], grad_fn=) 2025-09-09T14:44:11.7877941Z converted model pt2e: GraphModule( 2025-09-09T14:44:11.7878298Z (conv): Module() 2025-09-09T14:44:11.7878564Z (bn): Module() 2025-09-09T14:44:11.7878815Z ) 2025-09-09T14:44:11.7878936Z 2025-09-09T14:44:11.7878941Z 2025-09-09T14:44:11.7878946Z 2025-09-09T14:44:11.7879058Z def forward(self, x): 2025-09-09T14:44:11.7879399Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:44:11.7879811Z conv_bias = self.conv.bias 2025-09-09T14:44:11.7880176Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:44:11.7881066Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.008714424446225166, 57, -128, 127, torch.int8); x = None 2025-09-09T14:44:11.7882615Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.008714424446225166, 57, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:44:11.7883922Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:44:11.7884531Z quantize_per_tensor = self._frozen_param0 2025-09-09T14:44:11.7885522Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.0024561204481869936, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T14:44:11.7887104Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias, [2], [4]); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T14:44:11.7888660Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_2, 0.011943548917770386, -44, -128, 127, torch.int8); conv1d_2 = None 2025-09-09T14:44:11.7890299Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.011943548917770386, -44, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:44:11.7891691Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:44:11.7892128Z 2025-09-09T14:44:11.7892417Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:44:11.7892797Z onverted model fx: GraphModule( 2025-09-09T14:44:11.7893210Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(2,), padding=(4,)) 2025-09-09T14:44:11.7893627Z ) 2025-09-09T14:44:11.7893729Z 2025-09-09T14:44:11.7893733Z 2025-09-09T14:44:11.7893737Z 2025-09-09T14:44:11.7893823Z def forward(self, x): 2025-09-09T14:44:11.7894451Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.008714424446225166, 57, -128, 127, torch.int8); x = None 2025-09-09T14:44:11.7895863Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.008714424446225166, 57, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:44:11.7896874Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:44:11.7898015Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.011943548917770386, -44, -128, 127, torch.int8); conv = None 2025-09-09T14:44:11.7899280Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.011943548917770386, -44, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:44:11.7900164Z return dequantize_per_tensor_default_1 2025-09-09T14:44:11.7900450Z 2025-09-09T14:44:11.7900730Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:44:11.7901109Z diff: tensor([[[0., 0., 0., 0., 0., 0.], 2025-09-09T14:44:11.7901378Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:44:11.7901635Z [0., 0., 0., 0., 0., 0.]]]) 2025-09-09T14:44:11.7902098Z PASSED 2025-09-09T14:44:11.7902719Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_bn_fusion_no_conv_bias model pt2e: GraphModule( 2025-09-09T14:44:11.7903372Z (conv): Module() 2025-09-09T14:44:11.7903583Z (bn): Module() 2025-09-09T14:44:11.7903896Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:44:11.7904831Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0150]), zero_point=tensor([14], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:44:11.7905939Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1280412673950195, max_val=1.6863642930984497) 2025-09-09T14:44:11.7906451Z ) 2025-09-09T14:44:11.7906740Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:44:11.7907741Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0022, 0.0020, 0.0022]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:44:11.7909036Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2799, -0.2557, -0.2618]), max_val=tensor([0.1970, 0.2308, 0.2775])) 2025-09-09T14:44:11.7909693Z ) 2025-09-09T14:44:11.7909971Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:44:11.7910906Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0146]), zero_point=tensor([8], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:44:11.7911993Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.9912875890731812, max_val=1.733071208000183) 2025-09-09T14:44:11.7912497Z ) 2025-09-09T14:44:11.7912672Z ) 2025-09-09T14:44:11.7912772Z 2025-09-09T14:44:11.7912776Z 2025-09-09T14:44:11.7912787Z 2025-09-09T14:44:11.7912879Z def forward(self, x): 2025-09-09T14:44:11.7913162Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:44:11.7913514Z conv_weight = self.conv.weight 2025-09-09T14:44:11.7913790Z bn_weight = self.bn.weight 2025-09-09T14:44:11.7914048Z bn_bias = self.bn.bias 2025-09-09T14:44:11.7914306Z bn_running_mean = self.bn.running_mean 2025-09-09T14:44:11.7914616Z bn_running_var = self.bn.running_var 2025-09-09T14:44:11.7914969Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:44:11.7923301Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:44:11.7923920Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:44:11.7924459Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:44:11.7924860Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:44:11.7925463Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:44:11.7925918Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T14:44:11.7926523Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:44:11.7927095Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:44:11.7927938Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T14:44:11.7928759Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T14:44:11.7929295Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T14:44:11.7930189Z batch_norm_1 = torch.ops.aten.batch_norm.default(div_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); div_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:44:11.7931118Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T14:44:11.7931724Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:44:11.7932110Z 2025-09-09T14:44:11.7932398Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:44:11.7932760Z model fx: GraphModule( 2025-09-09T14:44:11.7933093Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:44:11.7934037Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0150]), zero_point=tensor([14], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:44:11.7935148Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1280412673950195, max_val=1.6863642930984497) 2025-09-09T14:44:11.7935664Z ) 2025-09-09T14:44:11.7935909Z (conv): ConvBn1d( 2025-09-09T14:44:11.7936162Z 3, 3, kernel_size=(3,), stride=(1,), bias=False 2025-09-09T14:44:11.7936585Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:44:11.7937057Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:44:11.7938018Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0022, 0.0020, 0.0022]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:44:30.2795053Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2799, -0.2557, -0.2618]), max_val=tensor([0.1970, 0.2308, 0.2775])) 2025-09-09T14:44:30.2795991Z ) 2025-09-09T14:44:30.2796220Z ) 2025-09-09T14:44:30.2796573Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:44:30.2797975Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0146]), zero_point=tensor([8], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:44:30.2799401Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.9912875890731812, max_val=1.733071208000183) 2025-09-09T14:44:30.2800049Z ) 2025-09-09T14:44:30.2800260Z ) 2025-09-09T14:44:30.2800388Z 2025-09-09T14:44:30.2800393Z 2025-09-09T14:44:30.2800398Z 2025-09-09T14:44:30.2800504Z def forward(self, x): 2025-09-09T14:44:30.2800943Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:44:30.2801605Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:44:30.2802291Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:44:30.2802822Z return activation_post_process_1 2025-09-09T14:44:30.2803152Z 2025-09-09T14:44:30.2803887Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:44:30.2804410Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:44:30.2804704Z [0., 0., 0.], 2025-09-09T14:44:30.2804962Z [0., 0., 0.]], 2025-09-09T14:44:30.2805299Z 2025-09-09T14:44:30.2805402Z [[0., 0., 0.], 2025-09-09T14:44:30.2805650Z [0., 0., 0.], 2025-09-09T14:44:30.2805905Z [0., 0., 0.]], 2025-09-09T14:44:30.2806067Z 2025-09-09T14:44:30.2806162Z [[0., 0., 0.], 2025-09-09T14:44:30.2806417Z [0., 0., 0.], 2025-09-09T14:44:30.2806701Z [0., 0., 0.]]], grad_fn=) 2025-09-09T14:44:30.2807078Z converted model pt2e: GraphModule( 2025-09-09T14:44:30.2807403Z (conv): Module() 2025-09-09T14:44:30.2807649Z (bn): Module() 2025-09-09T14:44:30.2807889Z ) 2025-09-09T14:44:30.2808009Z 2025-09-09T14:44:30.2808014Z 2025-09-09T14:44:30.2808019Z 2025-09-09T14:44:30.2808124Z def forward(self, x): 2025-09-09T14:44:30.2808472Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:44:30.2808933Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:44:30.2809833Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.014958452433347702, 14, -128, 127, torch.int8); x = None 2025-09-09T14:44:30.2811389Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.014958452433347702, 14, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:44:30.2812677Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:44:30.2813256Z _scale_0 = self._scale_0 2025-09-09T14:44:30.2813568Z _zero_point_0 = self._zero_point_0 2025-09-09T14:44:30.2813944Z quantize_per_channel = self._frozen_param0 2025-09-09T14:44:30.2815109Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T14:44:30.2816365Z conv_weight_bias = self.conv.weight_bias 2025-09-09T14:44:30.2817414Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_weight_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_weight_bias = None 2025-09-09T14:44:30.2818989Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_2, 0.014605328440666199, 8, -128, 127, torch.int8); conv1d_2 = None 2025-09-09T14:44:30.2820615Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.014605328440666199, 8, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:44:30.2821881Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T14:44:30.2822387Z 2025-09-09T14:44:30.2822736Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:44:30.2823196Z onverted model fx: GraphModule( 2025-09-09T14:44:30.2823699Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T14:44:30.2824203Z ) 2025-09-09T14:44:30.2824329Z 2025-09-09T14:44:30.2824334Z 2025-09-09T14:44:30.2824340Z 2025-09-09T14:44:30.2824450Z def forward(self, x): 2025-09-09T14:44:30.2825106Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.014958452433347702, 14, -128, 127, torch.int8); x = None 2025-09-09T14:44:30.2826338Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.014958452433347702, 14, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:44:30.2827349Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:44:30.2829250Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.014605328440666199, 8, -128, 127, torch.int8); conv = None 2025-09-09T14:44:30.2830532Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.014605328440666199, 8, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:44:30.2831510Z return dequantize_per_tensor_default_1 2025-09-09T14:44:30.2831794Z 2025-09-09T14:44:30.2832091Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:44:30.2832474Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:44:30.2832713Z [0., 0., 0.], 2025-09-09T14:44:30.2832932Z [0., 0., 0.]], 2025-09-09T14:44:30.2833068Z 2025-09-09T14:44:30.2833145Z [[0., 0., 0.], 2025-09-09T14:44:30.2833362Z [0., 0., 0.], 2025-09-09T14:44:30.2833569Z [0., 0., 0.]], 2025-09-09T14:44:30.2833711Z 2025-09-09T14:44:30.2833790Z [[0., 0., 0.], 2025-09-09T14:44:30.2833999Z [0., 0., 0.], 2025-09-09T14:44:30.2834223Z [0., 0., 0.]]]) 2025-09-09T14:44:30.2834464Z model pt2e: GraphModule( 2025-09-09T14:44:30.2834697Z (conv): Module() 2025-09-09T14:44:30.2834920Z (bn): Module() 2025-09-09T14:44:30.2835228Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:44:30.2836177Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0150]), zero_point=tensor([14], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:44:30.2837288Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1280412673950195, max_val=1.6863642930984497) 2025-09-09T14:44:30.2837811Z ) 2025-09-09T14:44:30.2838097Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:44:30.2839055Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0022]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:44:30.2840173Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.2799264192581177, max_val=0.27745386958122253) 2025-09-09T14:44:30.2840694Z ) 2025-09-09T14:44:30.2840980Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:44:30.2841919Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0146]), zero_point=tensor([8], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:44:30.2843012Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.9912875890731812, max_val=1.733071208000183) 2025-09-09T14:44:30.2843527Z ) 2025-09-09T14:44:30.2843698Z ) 2025-09-09T14:44:30.2843806Z 2025-09-09T14:44:30.2843810Z 2025-09-09T14:44:30.2843814Z 2025-09-09T14:44:30.2843899Z def forward(self, x): 2025-09-09T14:44:30.2844208Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:44:30.2844595Z conv_weight = self.conv.weight 2025-09-09T14:44:30.2844880Z bn_weight = self.bn.weight 2025-09-09T14:44:30.2845134Z bn_bias = self.bn.bias 2025-09-09T14:44:30.2845410Z bn_running_mean = self.bn.running_mean 2025-09-09T14:44:30.2845714Z bn_running_var = self.bn.running_var 2025-09-09T14:44:30.2846056Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:44:30.2846500Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:44:30.2847096Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:44:30.2847633Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:44:30.2848030Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:44:30.2848453Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:44:30.2848898Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T14:44:30.2849510Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:44:30.2850080Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:44:30.2851010Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T14:44:30.2851842Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T14:44:30.2852381Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T14:44:30.2853288Z batch_norm_1 = torch.ops.aten.batch_norm.default(div_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); div_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:44:30.2854236Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T14:44:30.2854900Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:44:30.2855296Z 2025-09-09T14:44:30.2855582Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:44:30.2856032Z model fx: GraphModule( 2025-09-09T14:44:30.2856367Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:44:30.2857321Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0150]), zero_point=tensor([14], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:44:30.2858440Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1280412673950195, max_val=1.6863642930984497) 2025-09-09T14:44:30.2858956Z ) 2025-09-09T14:44:30.2859145Z (conv): ConvBn1d( 2025-09-09T14:44:30.2859394Z 3, 3, kernel_size=(3,), stride=(1,), bias=False 2025-09-09T14:44:30.2859838Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:44:30.2860309Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:44:48.8046625Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0022]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:44:48.8048125Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.2799264192581177, max_val=0.27745386958122253) 2025-09-09T14:44:48.8048782Z ) 2025-09-09T14:44:48.8048992Z ) 2025-09-09T14:44:48.8049341Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:44:48.8050521Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0146]), zero_point=tensor([8], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:44:48.8051919Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.9912875890731812, max_val=1.733071208000183) 2025-09-09T14:44:48.8052572Z ) 2025-09-09T14:44:48.8052785Z ) 2025-09-09T14:44:48.8052918Z 2025-09-09T14:44:48.8052929Z 2025-09-09T14:44:48.8052934Z 2025-09-09T14:44:48.8053042Z def forward(self, x): 2025-09-09T14:44:48.8053482Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:44:48.8054147Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:44:48.8054838Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:44:48.8055373Z return activation_post_process_1 2025-09-09T14:44:48.8055702Z 2025-09-09T14:44:48.8056132Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:44:48.8056590Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:44:48.8056890Z [0., 0., 0.], 2025-09-09T14:44:48.8057145Z [0., 0., 0.]], 2025-09-09T14:44:48.8057310Z 2025-09-09T14:44:48.8057753Z [[0., 0., 0.], 2025-09-09T14:44:48.8058014Z [0., 0., 0.], 2025-09-09T14:44:48.8058279Z [0., 0., 0.]], 2025-09-09T14:44:48.8058442Z 2025-09-09T14:44:48.8058719Z [[0., 0., 0.], 2025-09-09T14:44:48.8058988Z [0., 0., 0.], 2025-09-09T14:44:48.8059274Z [0., 0., 0.]]], grad_fn=) 2025-09-09T14:44:48.8059655Z converted model pt2e: GraphModule( 2025-09-09T14:44:48.8059975Z (conv): Module() 2025-09-09T14:44:48.8060281Z (bn): Module() 2025-09-09T14:44:48.8060522Z ) 2025-09-09T14:44:48.8060643Z 2025-09-09T14:44:48.8060648Z 2025-09-09T14:44:48.8060653Z 2025-09-09T14:44:48.8060760Z def forward(self, x): 2025-09-09T14:44:48.8061110Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:44:48.8061565Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:44:48.8062473Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.014958452433347702, 14, -128, 127, torch.int8); x = None 2025-09-09T14:44:48.8064051Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.014958452433347702, 14, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:44:48.8065362Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:44:48.8065996Z quantize_per_tensor = self._frozen_param0 2025-09-09T14:44:48.8066996Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.002204145072028041, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T14:44:48.8067996Z conv_weight_bias = self.conv.weight_bias 2025-09-09T14:44:48.8069034Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_weight_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_weight_bias = None 2025-09-09T14:44:48.8070643Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_2, 0.014605328440666199, 8, -128, 127, torch.int8); conv1d_2 = None 2025-09-09T14:44:48.8072321Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.014605328440666199, 8, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:44:48.8073497Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:44:48.8073908Z 2025-09-09T14:44:48.8074201Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:44:48.8074583Z onverted model fx: GraphModule( 2025-09-09T14:44:48.8074957Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T14:44:48.8075332Z ) 2025-09-09T14:44:48.8075433Z 2025-09-09T14:44:48.8075437Z 2025-09-09T14:44:48.8075442Z 2025-09-09T14:44:48.8075529Z def forward(self, x): 2025-09-09T14:44:48.8076160Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.014958452433347702, 14, -128, 127, torch.int8); x = None 2025-09-09T14:44:48.8077405Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.014958452433347702, 14, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:44:48.8078409Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:44:48.8079266Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.014605328440666199, 8, -128, 127, torch.int8); conv = None 2025-09-09T14:44:48.8080572Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.014605328440666199, 8, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:44:48.8081458Z return dequantize_per_tensor_default_1 2025-09-09T14:44:48.8081839Z 2025-09-09T14:44:48.8082122Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:44:48.8082500Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:44:48.8082823Z [0., 0., 0.], 2025-09-09T14:44:48.8083037Z [0., 0., 0.]], 2025-09-09T14:44:48.8083175Z 2025-09-09T14:44:48.8083256Z [[0., 0., 0.], 2025-09-09T14:44:48.8083467Z [0., 0., 0.], 2025-09-09T14:44:48.8083672Z [0., 0., 0.]], 2025-09-09T14:44:48.8083812Z 2025-09-09T14:44:48.8083890Z [[0., 0., 0.], 2025-09-09T14:44:48.8084099Z [0., 0., 0.], 2025-09-09T14:44:48.8084304Z [0., 0., 0.]]]) 2025-09-09T14:44:48.8084544Z model pt2e: GraphModule( 2025-09-09T14:44:48.8084775Z (conv1): Module() 2025-09-09T14:44:48.8084985Z (bn1): Module() 2025-09-09T14:44:48.8085185Z (conv2): Module() 2025-09-09T14:44:48.8085395Z (bn2): Module() 2025-09-09T14:44:48.8085702Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:44:48.8086692Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0150]), zero_point=tensor([14], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:44:48.8087798Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1280412673950195, max_val=1.6863642930984497) 2025-09-09T14:44:48.8088307Z ) 2025-09-09T14:44:48.8088591Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:44:48.8089575Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0023, 0.0026, 0.0025]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:44:48.8090889Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2878, -0.2584, -0.3162]), max_val=tensor([0.2745, 0.3315, 0.3105])) 2025-09-09T14:44:48.8091549Z ) 2025-09-09T14:44:48.8091837Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:44:48.8092825Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0025, 0.0024, 0.0023]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:44:48.8094106Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.3233, -0.3086, -0.2979]), max_val=tensor([0.3026, 0.1712, 0.2405])) 2025-09-09T14:44:48.8094752Z ) 2025-09-09T14:44:48.8095033Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:44:48.8096066Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0161]), zero_point=tensor([-16], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:44:48.8097187Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.805302619934082, max_val=2.310788631439209) 2025-09-09T14:44:48.8097958Z ) 2025-09-09T14:44:48.8098245Z (activation_post_process_4): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:44:48.8099186Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0102]), zero_point=tensor([-11], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:44:48.8100277Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.1982380151748657, max_val=1.4133442640304565) 2025-09-09T14:44:48.8100795Z ) 2025-09-09T14:44:48.8100968Z ) 2025-09-09T14:44:48.8101075Z 2025-09-09T14:44:48.8101079Z 2025-09-09T14:44:48.8101083Z 2025-09-09T14:44:48.8101170Z def forward(self, x): 2025-09-09T14:44:48.8101462Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:44:48.8101805Z conv1_weight = self.conv1.weight 2025-09-09T14:44:48.8102097Z bn1_weight = self.bn1.weight 2025-09-09T14:44:48.8102499Z bn1_bias = self.bn1.bias 2025-09-09T14:44:48.8102768Z conv2_weight = self.conv2.weight 2025-09-09T14:44:48.8103049Z conv2_bias = self.conv2.bias 2025-09-09T14:44:48.8103424Z bn2_weight = self.bn2.weight 2025-09-09T14:44:48.8103680Z bn2_bias = self.bn2.bias 2025-09-09T14:44:48.8103957Z bn1_running_mean = self.bn1.running_mean 2025-09-09T14:44:48.8104279Z bn1_running_var = self.bn1.running_var 2025-09-09T14:44:48.8104621Z bn1_num_batches_tracked = self.bn1.num_batches_tracked 2025-09-09T14:44:48.8104977Z bn2_running_mean = self.bn2.running_mean 2025-09-09T14:44:48.8105284Z bn2_running_var = self.bn2.running_var 2025-09-09T14:44:48.8105622Z bn2_num_batches_tracked = self.bn2.num_batches_tracked 2025-09-09T14:44:48.8106064Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:44:48.8106664Z add_ = torch.ops.aten.add_.Tensor(bn1_num_batches_tracked, 1); bn1_num_batches_tracked = add_ = None 2025-09-09T14:44:48.8107340Z add__1 = torch.ops.aten.add_.Tensor(bn2_num_batches_tracked, 1); bn2_num_batches_tracked = add__1 = None 2025-09-09T14:44:48.8107879Z add = torch.ops.aten.add.Tensor(bn2_running_var, 1e-05) 2025-09-09T14:44:48.8108284Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:44:48.8108697Z div = torch.ops.aten.div.Tensor(bn2_weight, sqrt); sqrt = None 2025-09-09T14:44:48.8109146Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T14:44:48.8109651Z mul = torch.ops.aten.mul.Tensor(conv2_weight, reshape); conv2_weight = reshape = None 2025-09-09T14:44:48.8110269Z activation_post_process_3 = self.activation_post_process_3(mul); mul = None 2025-09-09T14:44:48.8110908Z zeros_like = torch.ops.aten.zeros_like.default(conv2_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:45:02.5062674Z add_2 = torch.ops.aten.add.Tensor(bn1_running_var, 1e-05) 2025-09-09T14:45:02.5063155Z sqrt_1 = torch.ops.aten.sqrt.default(add_2); add_2 = None 2025-09-09T14:45:02.5063627Z div_2 = torch.ops.aten.div.Tensor(bn1_weight, sqrt_1); sqrt_1 = None 2025-09-09T14:45:02.5064096Z reshape_3 = torch.ops.aten.reshape.default(div_2, [-1, 1, 1]) 2025-09-09T14:45:02.5064641Z mul_1 = torch.ops.aten.mul.Tensor(conv1_weight, reshape_3); conv1_weight = reshape_3 = None 2025-09-09T14:45:02.5065235Z activation_post_process_1 = self.activation_post_process_1(mul_1); mul_1 = None 2025-09-09T14:45:02.5066082Z conv1d_3 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T14:45:02.5066921Z reshape_4 = torch.ops.aten.reshape.default(div_2, [1, -1, 1]); div_2 = None 2025-09-09T14:45:02.5067473Z div_3 = torch.ops.aten.div.Tensor(conv1d_3, reshape_4); conv1d_3 = reshape_4 = None 2025-09-09T14:45:02.5068395Z batch_norm_3 = torch.ops.aten.batch_norm.default(div_3, bn1_weight, bn1_bias, bn1_running_mean, bn1_running_var, True, 0.1, 1e-05, True); div_3 = bn1_weight = bn1_bias = bn1_running_mean = bn1_running_var = None 2025-09-09T14:45:02.5069355Z activation_post_process_2 = self.activation_post_process_2(batch_norm_3); batch_norm_3 = None 2025-09-09T14:45:02.5070358Z conv1d_2 = torch.ops.aten.conv1d.default(activation_post_process_2, activation_post_process_3, zeros_like); activation_post_process_2 = activation_post_process_3 = zeros_like = None 2025-09-09T14:45:02.5071237Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T14:45:02.5071779Z div_1 = torch.ops.aten.div.Tensor(conv1d_2, reshape_1); conv1d_2 = reshape_1 = None 2025-09-09T14:45:02.5072358Z reshape_2 = torch.ops.aten.reshape.default(conv2_bias, [1, -1, 1]); conv2_bias = None 2025-09-09T14:45:02.5072923Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:45:02.5074057Z batch_norm_2 = torch.ops.aten.batch_norm.default(add_1, bn2_weight, bn2_bias, bn2_running_mean, bn2_running_var, True, 0.1, 1e-05, True); add_1 = bn2_weight = bn2_bias = bn2_running_mean = bn2_running_var = None 2025-09-09T14:45:02.5075019Z activation_post_process_4 = self.activation_post_process_4(batch_norm_2); batch_norm_2 = None 2025-09-09T14:45:02.5075776Z return pytree.tree_unflatten((activation_post_process_4,), self._out_spec) 2025-09-09T14:45:02.5076167Z 2025-09-09T14:45:02.5076456Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:45:02.5076820Z model fx: GraphModule( 2025-09-09T14:45:02.5077160Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:45:02.5078109Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0150]), zero_point=tensor([14], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:45:02.5079219Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1280412673950195, max_val=1.6863642930984497) 2025-09-09T14:45:02.5079738Z ) 2025-09-09T14:45:02.5079924Z (conv1): ConvBn1d( 2025-09-09T14:45:02.5080189Z 3, 3, kernel_size=(3,), stride=(1,), bias=False 2025-09-09T14:45:02.5080627Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:45:02.5081090Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:45:02.5082054Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0025, 0.0024, 0.0023]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:45:02.5083340Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.3233, -0.3086, -0.2979]), max_val=tensor([0.3026, 0.1712, 0.2405])) 2025-09-09T14:45:02.5083989Z ) 2025-09-09T14:45:02.5084174Z ) 2025-09-09T14:45:02.5084456Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:45:02.5085398Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0161]), zero_point=tensor([-16], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:45:02.5086493Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.805302619934082, max_val=2.310788631439209) 2025-09-09T14:45:02.5087003Z ) 2025-09-09T14:45:02.5087183Z (conv2): ConvBn1d( 2025-09-09T14:45:02.5087423Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T14:45:02.5087829Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:45:02.5088292Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:45:02.5089260Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0023, 0.0026, 0.0025]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:45:02.5090542Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2878, -0.2584, -0.3162]), max_val=tensor([0.2745, 0.3315, 0.3105])) 2025-09-09T14:45:02.5091197Z ) 2025-09-09T14:45:02.5091377Z ) 2025-09-09T14:45:02.5091657Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:45:02.5092597Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0102]), zero_point=tensor([-11], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:45:02.5093695Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.1982380151748657, max_val=1.4133442640304565) 2025-09-09T14:45:02.5094212Z ) 2025-09-09T14:45:02.5094386Z ) 2025-09-09T14:45:02.5094493Z 2025-09-09T14:45:02.5094497Z 2025-09-09T14:45:02.5094501Z 2025-09-09T14:45:02.5094590Z def forward(self, x): 2025-09-09T14:45:02.5095079Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:45:02.5095632Z conv1 = self.conv1(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:45:02.5096367Z activation_post_process_1 = self.activation_post_process_1(conv1); conv1 = None 2025-09-09T14:45:02.5096928Z conv2 = self.conv2(activation_post_process_1); activation_post_process_1 = None 2025-09-09T14:45:02.5097757Z activation_post_process_2 = self.activation_post_process_2(conv2); conv2 = None 2025-09-09T14:45:02.5098342Z return activation_post_process_2 2025-09-09T14:45:02.5098676Z 2025-09-09T14:45:02.5099021Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:45:02.5099380Z diff: tensor([[[0.], 2025-09-09T14:45:02.5099596Z [0.], 2025-09-09T14:45:02.5099786Z [0.]], 2025-09-09T14:45:02.5099910Z 2025-09-09T14:45:02.5099986Z [[0.], 2025-09-09T14:45:02.5100173Z [0.], 2025-09-09T14:45:02.5100369Z [0.]], 2025-09-09T14:45:02.5100497Z 2025-09-09T14:45:02.5100581Z [[0.], 2025-09-09T14:45:02.5100767Z [0.], 2025-09-09T14:45:02.5100985Z [0.]]], grad_fn=) 2025-09-09T14:45:02.5101282Z converted model pt2e: GraphModule( 2025-09-09T14:45:02.5101550Z (conv1): Module() 2025-09-09T14:45:02.5101752Z (bn1): Module() 2025-09-09T14:45:02.5101958Z (conv2): Module() 2025-09-09T14:45:02.5102164Z (bn2): Module() 2025-09-09T14:45:02.5102359Z ) 2025-09-09T14:45:02.5102460Z 2025-09-09T14:45:02.5102464Z 2025-09-09T14:45:02.5102468Z 2025-09-09T14:45:02.5102554Z def forward(self, x): 2025-09-09T14:45:02.5102845Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:45:02.5103188Z conv2_bias = self.conv2.bias 2025-09-09T14:45:02.5103504Z bn1_num_batches_tracked = self.bn1.num_batches_tracked 2025-09-09T14:45:02.5103889Z bn2_num_batches_tracked = self.bn2.num_batches_tracked 2025-09-09T14:45:02.5104634Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.014958452433347702, 14, -128, 127, torch.int8); x = None 2025-09-09T14:45:02.5105899Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.014958452433347702, 14, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:45:02.5106945Z add_ = torch.ops.aten.add_.Tensor(bn1_num_batches_tracked, 1); bn1_num_batches_tracked = add_ = None 2025-09-09T14:45:02.5107603Z add__1 = torch.ops.aten.add_.Tensor(bn2_num_batches_tracked, 1); bn2_num_batches_tracked = add__1 = None 2025-09-09T14:45:02.5108083Z _scale_0 = self._scale_0 2025-09-09T14:45:02.5108341Z _zero_point_0 = self._zero_point_0 2025-09-09T14:45:02.5108617Z _scale_1 = self._scale_1 2025-09-09T14:45:02.5108864Z _zero_point_1 = self._zero_point_1 2025-09-09T14:45:02.5109168Z quantize_per_channel_1 = self._frozen_param0 2025-09-09T14:45:02.5110078Z dequantize_per_channel_1 = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel_1, _scale_1, _zero_point_1, 0, -127, 127, torch.int8); quantize_per_channel_1 = _scale_1 = _zero_point_1 = None 2025-09-09T14:45:02.5110973Z conv1_weight_bias = self.conv1.weight_bias 2025-09-09T14:45:02.5111833Z conv1d_5 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_channel_1, conv1_weight_bias); dequantize_per_tensor_default = dequantize_per_channel_1 = conv1_weight_bias = None 2025-09-09T14:45:02.5113097Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_5, 0.016141533851623535, -16, -128, 127, torch.int8); conv1d_5 = None 2025-09-09T14:45:02.5114388Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.016141533851623535, -16, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:45:02.5115291Z quantize_per_channel = self._frozen_param1 2025-09-09T14:45:02.5116323Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T14:45:02.5117795Z conv1d_4 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default_1, dequantize_per_channel, conv2_bias); dequantize_per_tensor_default_1 = dequantize_per_channel = conv2_bias = None 2025-09-09T14:45:02.5119011Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_4, 0.010241499170660973, -11, -128, 127, torch.int8); conv1d_4 = None 2025-09-09T14:45:05.2314017Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.010241499170660973, -11, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:45:05.2315798Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:45:05.2327417Z 2025-09-09T14:45:05.2327957Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:45:05.2328609Z onverted model fx: GraphModule( 2025-09-09T14:45:05.2329255Z (conv1): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T14:45:05.2330153Z (conv2): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T14:45:05.2330811Z ) 2025-09-09T14:45:05.2330966Z 2025-09-09T14:45:05.2330986Z 2025-09-09T14:45:05.2330991Z 2025-09-09T14:45:05.2331120Z def forward(self, x): 2025-09-09T14:45:05.2332136Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.014958452433347702, 14, -128, 127, torch.int8); x = None 2025-09-09T14:45:05.2334313Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.014958452433347702, 14, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:45:05.2336218Z conv1 = self.conv1(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:45:05.2337670Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1, 0.016141533851623535, -16, -128, 127, torch.int8); conv1 = None 2025-09-09T14:45:05.2339704Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.016141533851623535, -16, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:45:05.2341378Z conv2 = self.conv2(dequantize_per_tensor_default_1); dequantize_per_tensor_default_1 = None 2025-09-09T14:45:05.2342811Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2, 0.010241499170660973, -11, -128, 127, torch.int8); conv2 = None 2025-09-09T14:45:05.2344981Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.010241499170660973, -11, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:45:05.2346387Z return dequantize_per_tensor_default_2 2025-09-09T14:45:05.2346845Z 2025-09-09T14:45:05.2347307Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:45:05.2347895Z diff: tensor([[[0.], 2025-09-09T14:45:05.2348220Z [0.], 2025-09-09T14:45:05.2348522Z [0.]], 2025-09-09T14:45:05.2348709Z 2025-09-09T14:45:05.2348819Z [[0.], 2025-09-09T14:45:05.2349101Z [0.], 2025-09-09T14:45:05.2349366Z [0.]], 2025-09-09T14:45:05.2349536Z 2025-09-09T14:45:05.2349653Z [[0.], 2025-09-09T14:45:05.2349928Z [0.], 2025-09-09T14:45:05.2350204Z [0.]]]) 2025-09-09T14:45:05.2350531Z model pt2e: GraphModule( 2025-09-09T14:45:05.2350876Z (conv1): Module() 2025-09-09T14:45:05.2351179Z (bn1): Module() 2025-09-09T14:45:05.2351461Z (conv2): Module() 2025-09-09T14:45:05.2351746Z (bn2): Module() 2025-09-09T14:45:05.2352624Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:45:05.2354299Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0150]), zero_point=tensor([14], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:45:05.2356266Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1280412673950195, max_val=1.6863642930984497) 2025-09-09T14:45:05.2357089Z ) 2025-09-09T14:45:05.2357484Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:45:05.2358954Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0026]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:45:05.2360740Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3161814510822296, max_val=0.33154603838920593) 2025-09-09T14:45:05.2361634Z ) 2025-09-09T14:45:05.2362112Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:45:05.2363499Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0025]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:45:05.2365219Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3232710361480713, max_val=0.30256387591362) 2025-09-09T14:45:05.2366029Z ) 2025-09-09T14:45:05.2366458Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:45:05.2367922Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0161]), zero_point=tensor([-16], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:45:05.2369686Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.807204008102417, max_val=2.3096539974212646) 2025-09-09T14:45:05.2370494Z ) 2025-09-09T14:45:05.2370918Z (activation_post_process_4): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:45:05.2372427Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0102]), zero_point=tensor([-11], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:45:05.2374251Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.2001533508300781, max_val=1.4126498699188232) 2025-09-09T14:45:05.2375058Z ) 2025-09-09T14:45:05.2375319Z ) 2025-09-09T14:45:05.2375463Z 2025-09-09T14:45:05.2375469Z 2025-09-09T14:45:05.2375475Z 2025-09-09T14:45:05.2375617Z def forward(self, x): 2025-09-09T14:45:05.2376143Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:45:05.2376661Z conv1_weight = self.conv1.weight 2025-09-09T14:45:05.2377106Z bn1_weight = self.bn1.weight 2025-09-09T14:45:05.2377529Z bn1_bias = self.bn1.bias 2025-09-09T14:45:05.2377916Z conv2_weight = self.conv2.weight 2025-09-09T14:45:05.2378354Z conv2_bias = self.conv2.bias 2025-09-09T14:45:05.2378766Z bn2_weight = self.bn2.weight 2025-09-09T14:45:05.2379185Z bn2_bias = self.bn2.bias 2025-09-09T14:45:05.2379629Z bn1_running_mean = self.bn1.running_mean 2025-09-09T14:45:05.2380070Z bn1_running_var = self.bn1.running_var 2025-09-09T14:45:05.2380593Z bn1_num_batches_tracked = self.bn1.num_batches_tracked 2025-09-09T14:45:05.2381156Z bn2_running_mean = self.bn2.running_mean 2025-09-09T14:45:05.2381644Z bn2_running_var = self.bn2.running_var 2025-09-09T14:45:05.2382160Z bn2_num_batches_tracked = self.bn2.num_batches_tracked 2025-09-09T14:45:05.2382885Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:45:05.2383773Z add_ = torch.ops.aten.add_.Tensor(bn1_num_batches_tracked, 1); bn1_num_batches_tracked = add_ = None 2025-09-09T14:45:05.2385041Z add__1 = torch.ops.aten.add_.Tensor(bn2_num_batches_tracked, 1); bn2_num_batches_tracked = add__1 = None 2025-09-09T14:45:05.2385941Z add = torch.ops.aten.add.Tensor(bn2_running_var, 1e-05) 2025-09-09T14:45:05.2388280Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:45:05.2388909Z div = torch.ops.aten.div.Tensor(bn2_weight, sqrt); sqrt = None 2025-09-09T14:45:05.2389587Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T14:45:05.2390436Z mul = torch.ops.aten.mul.Tensor(conv2_weight, reshape); conv2_weight = reshape = None 2025-09-09T14:45:05.2391399Z activation_post_process_3 = self.activation_post_process_3(mul); mul = None 2025-09-09T14:45:05.2392415Z zeros_like = torch.ops.aten.zeros_like.default(conv2_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:45:05.2393340Z add_2 = torch.ops.aten.add.Tensor(bn1_running_var, 1e-05) 2025-09-09T14:45:05.2394009Z sqrt_1 = torch.ops.aten.sqrt.default(add_2); add_2 = None 2025-09-09T14:45:05.2394747Z div_2 = torch.ops.aten.div.Tensor(bn1_weight, sqrt_1); sqrt_1 = None 2025-09-09T14:45:05.2395487Z reshape_3 = torch.ops.aten.reshape.default(div_2, [-1, 1, 1]) 2025-09-09T14:45:05.2396381Z mul_1 = torch.ops.aten.mul.Tensor(conv1_weight, reshape_3); conv1_weight = reshape_3 = None 2025-09-09T14:45:05.2397804Z activation_post_process_1 = self.activation_post_process_1(mul_1); mul_1 = None 2025-09-09T14:45:05.2399297Z conv1d_3 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T14:45:05.2400716Z reshape_4 = torch.ops.aten.reshape.default(div_2, [1, -1, 1]); div_2 = None 2025-09-09T14:45:05.2401598Z div_3 = torch.ops.aten.div.Tensor(conv1d_3, reshape_4); conv1d_3 = reshape_4 = None 2025-09-09T14:45:05.2403104Z batch_norm_3 = torch.ops.aten.batch_norm.default(div_3, bn1_weight, bn1_bias, bn1_running_mean, bn1_running_var, True, 0.1, 1e-05, True); div_3 = bn1_weight = bn1_bias = bn1_running_mean = bn1_running_var = None 2025-09-09T14:45:05.2404585Z activation_post_process_2 = self.activation_post_process_2(batch_norm_3); batch_norm_3 = None 2025-09-09T14:45:05.2409659Z conv1d_2 = torch.ops.aten.conv1d.default(activation_post_process_2, activation_post_process_3, zeros_like); activation_post_process_2 = activation_post_process_3 = zeros_like = None 2025-09-09T14:45:05.2411184Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T14:45:05.2412077Z div_1 = torch.ops.aten.div.Tensor(conv1d_2, reshape_1); conv1d_2 = reshape_1 = None 2025-09-09T14:45:05.2413051Z reshape_2 = torch.ops.aten.reshape.default(conv2_bias, [1, -1, 1]); conv2_bias = None 2025-09-09T14:45:05.2413989Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:45:05.2415462Z batch_norm_2 = torch.ops.aten.batch_norm.default(add_1, bn2_weight, bn2_bias, bn2_running_mean, bn2_running_var, True, 0.1, 1e-05, True); add_1 = bn2_weight = bn2_bias = bn2_running_mean = bn2_running_var = None 2025-09-09T14:45:05.2417093Z activation_post_process_4 = self.activation_post_process_4(batch_norm_2); batch_norm_2 = None 2025-09-09T14:45:05.2418089Z return pytree.tree_unflatten((activation_post_process_4,), self._out_spec) 2025-09-09T14:45:05.2418738Z 2025-09-09T14:45:05.2419215Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:45:05.2419792Z model fx: GraphModule( 2025-09-09T14:45:05.2420281Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:45:44.2116342Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0150]), zero_point=tensor([14], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:45:44.2117792Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1280412673950195, max_val=1.6863642930984497) 2025-09-09T14:45:44.2118920Z ) 2025-09-09T14:45:44.2119161Z (conv1): ConvBn1d( 2025-09-09T14:45:44.2119464Z 3, 3, kernel_size=(3,), stride=(1,), bias=False 2025-09-09T14:45:44.2120181Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:45:44.2120771Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:45:44.2121919Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0025]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:45:44.2123309Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3232710361480713, max_val=0.30256387591362) 2025-09-09T14:45:44.2123945Z ) 2025-09-09T14:45:44.2124162Z ) 2025-09-09T14:45:44.2124501Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:45:44.2125684Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0161]), zero_point=tensor([-16], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:45:44.2127069Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.807204008102417, max_val=2.3096539974212646) 2025-09-09T14:45:44.2127698Z ) 2025-09-09T14:45:44.2127920Z (conv2): ConvBn1d( 2025-09-09T14:45:44.2128197Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T14:45:44.2128743Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:45:44.2129316Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:45:44.2130448Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0026]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:45:44.2131843Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3161814510822296, max_val=0.33154603838920593) 2025-09-09T14:45:44.2132484Z ) 2025-09-09T14:45:44.2132702Z ) 2025-09-09T14:45:44.2133040Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:45:44.2134215Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0102]), zero_point=tensor([-11], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:45:44.2135599Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.2001533508300781, max_val=1.4126498699188232) 2025-09-09T14:45:44.2136337Z ) 2025-09-09T14:45:44.2136550Z ) 2025-09-09T14:45:44.2136669Z 2025-09-09T14:45:44.2136674Z 2025-09-09T14:45:44.2136679Z 2025-09-09T14:45:44.2136785Z def forward(self, x): 2025-09-09T14:45:44.2137218Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:45:44.2137892Z conv1 = self.conv1(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:45:44.2138627Z activation_post_process_1 = self.activation_post_process_1(conv1); conv1 = None 2025-09-09T14:45:44.2139324Z conv2 = self.conv2(activation_post_process_1); activation_post_process_1 = None 2025-09-09T14:45:44.2140004Z activation_post_process_2 = self.activation_post_process_2(conv2); conv2 = None 2025-09-09T14:45:44.2140544Z return activation_post_process_2 2025-09-09T14:45:44.2140862Z 2025-09-09T14:45:44.2141210Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:45:44.2141660Z diff: tensor([[[0.], 2025-09-09T14:45:44.2141918Z [0.], 2025-09-09T14:45:44.2142153Z [0.]], 2025-09-09T14:45:44.2142300Z 2025-09-09T14:45:44.2142394Z [[0.], 2025-09-09T14:45:44.2142628Z [0.], 2025-09-09T14:45:44.2142856Z [0.]], 2025-09-09T14:45:44.2143006Z 2025-09-09T14:45:44.2143101Z [[0.], 2025-09-09T14:45:44.2143326Z [0.], 2025-09-09T14:45:44.2143693Z [0.]]], grad_fn=) 2025-09-09T14:45:44.2144055Z converted model pt2e: GraphModule( 2025-09-09T14:45:44.2144386Z (conv1): Module() 2025-09-09T14:45:44.2144729Z (bn1): Module() 2025-09-09T14:45:44.2144975Z (conv2): Module() 2025-09-09T14:45:44.2145225Z (bn2): Module() 2025-09-09T14:45:44.2145461Z ) 2025-09-09T14:45:44.2145581Z 2025-09-09T14:45:44.2145592Z 2025-09-09T14:45:44.2145597Z 2025-09-09T14:45:44.2145702Z def forward(self, x): 2025-09-09T14:45:44.2146047Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:45:44.2146469Z conv2_bias = self.conv2.bias 2025-09-09T14:45:44.2146849Z bn1_num_batches_tracked = self.bn1.num_batches_tracked 2025-09-09T14:45:44.2147322Z bn2_num_batches_tracked = self.bn2.num_batches_tracked 2025-09-09T14:45:44.2148271Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.014958452433347702, 14, -128, 127, torch.int8); x = None 2025-09-09T14:45:44.2149825Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.014958452433347702, 14, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:45:44.2151142Z add_ = torch.ops.aten.add_.Tensor(bn1_num_batches_tracked, 1); bn1_num_batches_tracked = add_ = None 2025-09-09T14:45:44.2151972Z add__1 = torch.ops.aten.add_.Tensor(bn2_num_batches_tracked, 1); bn2_num_batches_tracked = add__1 = None 2025-09-09T14:45:44.2152587Z quantize_per_tensor_1 = self._frozen_param0 2025-09-09T14:45:44.2153654Z dequantize_per_tensor_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_1, 0.0025454412680119276, 0, -127, 127, torch.int8); quantize_per_tensor_1 = None 2025-09-09T14:45:44.2154759Z conv1_weight_bias = self.conv1.weight_bias 2025-09-09T14:45:44.2155641Z conv1d_5 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_tensor_1, conv1_weight_bias); dequantize_per_tensor_default = dequantize_per_tensor_1 = conv1_weight_bias = None 2025-09-09T14:45:44.2156911Z quantize_per_tensor_default_3 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_5, 0.016144542023539543, -16, -128, 127, torch.int8); conv1d_5 = None 2025-09-09T14:45:44.2158214Z dequantize_per_tensor_default_3 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_3, 0.016144542023539543, -16, -128, 127, torch.int8); quantize_per_tensor_default_3 = None 2025-09-09T14:45:44.2159172Z quantize_per_tensor = self._frozen_param1 2025-09-09T14:45:44.2159973Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.0026105986908078194, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T14:45:44.2161244Z conv1d_4 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default_3, dequantize_per_tensor, conv2_bias); dequantize_per_tensor_default_3 = dequantize_per_tensor = conv2_bias = None 2025-09-09T14:45:44.2162464Z quantize_per_tensor_default_4 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_4, 0.01024628710001707, -11, -128, 127, torch.int8); conv1d_4 = None 2025-09-09T14:45:44.2163772Z dequantize_per_tensor_default_4 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_4, 0.01024628710001707, -11, -128, 127, torch.int8); quantize_per_tensor_default_4 = None 2025-09-09T14:45:44.2164782Z return pytree.tree_unflatten((dequantize_per_tensor_default_4,), self._out_spec) 2025-09-09T14:45:44.2165204Z 2025-09-09T14:45:44.2165493Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:45:44.2165876Z onverted model fx: GraphModule( 2025-09-09T14:45:44.2166265Z (conv1): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T14:45:44.2166764Z (conv2): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T14:45:44.2167156Z ) 2025-09-09T14:45:44.2167258Z 2025-09-09T14:45:44.2167369Z 2025-09-09T14:45:44.2167374Z 2025-09-09T14:45:44.2167476Z def forward(self, x): 2025-09-09T14:45:44.2168101Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.014958452433347702, 14, -128, 127, torch.int8); x = None 2025-09-09T14:45:44.2169486Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.014958452433347702, 14, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:45:44.2170522Z conv1 = self.conv1(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:45:44.2171390Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1, 0.016144542023539543, -16, -128, 127, torch.int8); conv1 = None 2025-09-09T14:45:44.2172676Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.016144542023539543, -16, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:45:44.2173721Z conv2 = self.conv2(dequantize_per_tensor_default_1); dequantize_per_tensor_default_1 = None 2025-09-09T14:45:44.2174602Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2, 0.01024628710001707, -11, -128, 127, torch.int8); conv2 = None 2025-09-09T14:45:44.2175935Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.01024628710001707, -11, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:45:44.2176813Z return dequantize_per_tensor_default_2 2025-09-09T14:45:44.2177097Z 2025-09-09T14:45:44.2177380Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:45:44.2177749Z diff: tensor([[[0.], 2025-09-09T14:45:44.2177990Z [0.], 2025-09-09T14:45:44.2178211Z [0.]], 2025-09-09T14:45:44.2178333Z 2025-09-09T14:45:44.2178424Z [[0.], 2025-09-09T14:45:44.2178613Z [0.], 2025-09-09T14:45:44.2178809Z [0.]], 2025-09-09T14:45:44.2178934Z 2025-09-09T14:45:44.2179008Z [[0.], 2025-09-09T14:45:44.2179200Z [0.], 2025-09-09T14:45:44.2179386Z [0.]]]) 2025-09-09T14:45:44.2179810Z PASSED 2025-09-09T14:45:44.2180556Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_bn_per_channel_weight_bias PASSED 2025-09-09T14:45:44.2181554Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_bn_relu_fusion model pt2e: GraphModule( 2025-09-09T14:45:44.2182176Z (conv): Module() 2025-09-09T14:45:44.2182387Z (bn): Module() 2025-09-09T14:45:44.2182700Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:00.2766255Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:46:00.2767691Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:46:00.2768351Z ) 2025-09-09T14:46:00.2768698Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:00.2769943Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024, 0.0026, 0.0025]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:46:00.2771575Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2845, -0.3289, -0.3229]), max_val=tensor([0.2989, 0.2870, 0.2939])) 2025-09-09T14:46:00.2772391Z ) 2025-09-09T14:46:00.2772731Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:00.2774242Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0055]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:46:00.2775821Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.413901925086975) 2025-09-09T14:46:00.2776511Z ) 2025-09-09T14:46:00.2776721Z ) 2025-09-09T14:46:00.2776841Z 2025-09-09T14:46:00.2776846Z 2025-09-09T14:46:00.2776851Z 2025-09-09T14:46:00.2776956Z def forward(self, x): 2025-09-09T14:46:00.2777305Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:46:00.2777720Z conv_weight = self.conv.weight 2025-09-09T14:46:00.2778051Z conv_bias = self.conv.bias 2025-09-09T14:46:00.2778372Z bn_weight = self.bn.weight 2025-09-09T14:46:00.2778672Z bn_bias = self.bn.bias 2025-09-09T14:46:00.2778990Z bn_running_mean = self.bn.running_mean 2025-09-09T14:46:00.2779353Z bn_running_var = self.bn.running_var 2025-09-09T14:46:00.2779768Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:46:00.2780300Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:46:00.2781030Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:46:00.2781681Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:46:00.2782155Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:46:00.2782654Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:46:00.2783232Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T14:46:00.2783847Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:46:00.2784532Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:46:00.2785293Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:46:00.2786498Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:46:00.2787589Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T14:46:00.2788238Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T14:46:00.2788933Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T14:46:00.2789605Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:46:00.2790672Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:46:00.2791711Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T14:46:00.2792356Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T14:46:00.2793018Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:46:00.2793529Z 2025-09-09T14:46:00.2793898Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:46:00.2794340Z model fx: GraphModule( 2025-09-09T14:46:00.2794734Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:00.2795891Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:46:00.2797258Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:46:00.2798108Z ) 2025-09-09T14:46:00.2798333Z (conv): ConvBnReLU1d( 2025-09-09T14:46:00.2798769Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T14:46:00.2799270Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:46:00.2799993Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:00.2801185Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024, 0.0026, 0.0025]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:46:00.2802820Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2845, -0.3289, -0.3229]), max_val=tensor([0.2989, 0.2870, 0.2939])) 2025-09-09T14:46:00.2803645Z ) 2025-09-09T14:46:00.2803891Z ) 2025-09-09T14:46:00.2804241Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:00.2805433Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0055]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:46:00.2806767Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.413901925086975) 2025-09-09T14:46:00.2807359Z ) 2025-09-09T14:46:00.2807564Z ) 2025-09-09T14:46:00.2807684Z 2025-09-09T14:46:00.2807689Z 2025-09-09T14:46:00.2807694Z 2025-09-09T14:46:00.2807810Z def forward(self, x): 2025-09-09T14:46:00.2808235Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:46:00.2808889Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:46:00.2809555Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:46:00.2810084Z return activation_post_process_1 2025-09-09T14:46:00.2810408Z 2025-09-09T14:46:00.2810743Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:46:00.2811201Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:46:00.2811485Z [0., 0., 0.], 2025-09-09T14:46:00.2811776Z [0., 0., 0.]]], grad_fn=) 2025-09-09T14:46:00.2812148Z converted model pt2e: GraphModule( 2025-09-09T14:46:00.2812473Z (conv): Module() 2025-09-09T14:46:00.2812716Z (bn): Module() 2025-09-09T14:46:00.2812975Z ) 2025-09-09T14:46:00.2813118Z 2025-09-09T14:46:00.2813124Z 2025-09-09T14:46:00.2813131Z 2025-09-09T14:46:00.2813258Z def forward(self, x): 2025-09-09T14:46:00.2813596Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:46:00.2814006Z conv_bias = self.conv.bias 2025-09-09T14:46:00.2814368Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:46:00.2815241Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:46:00.2816851Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:46:00.2818161Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:46:00.2818750Z _scale_0 = self._scale_0 2025-09-09T14:46:00.2819063Z _zero_point_0 = self._zero_point_0 2025-09-09T14:46:00.2819438Z quantize_per_channel = self._frozen_param0 2025-09-09T14:46:00.2820629Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T14:46:00.2822121Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T14:46:00.2822976Z relu = torch.ops.aten.relu.default(conv1d_2); conv1d_2 = None 2025-09-09T14:46:00.2823906Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.005544713232666254, -128, -128, 127, torch.int8); relu = None 2025-09-09T14:46:00.2825266Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.005544713232666254, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:46:00.2826283Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T14:46:00.2826693Z 2025-09-09T14:46:00.2826985Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:46:00.2827361Z onverted model fx: GraphModule( 2025-09-09T14:46:00.2827625Z (conv): ConvReLU1d( 2025-09-09T14:46:00.2827958Z (0): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T14:46:00.2828324Z (1): ReLU() 2025-09-09T14:46:00.2828519Z ) 2025-09-09T14:46:00.2828699Z ) 2025-09-09T14:46:00.2828799Z 2025-09-09T14:46:00.2828809Z 2025-09-09T14:46:00.2828813Z 2025-09-09T14:46:00.2828908Z def forward(self, x): 2025-09-09T14:46:00.2829519Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:46:00.2830769Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:46:00.2831782Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:46:00.2832642Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.005544713232666254, -128, -128, 127, torch.int8); conv = None 2025-09-09T14:46:00.2833986Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.005544713232666254, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:46:16.7101682Z return dequantize_per_tensor_default_1 2025-09-09T14:46:16.7102279Z 2025-09-09T14:46:16.7102630Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:46:16.7103079Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:46:16.7103373Z [0., 0., 0.], 2025-09-09T14:46:16.7103624Z [0., 0., 0.]]]) 2025-09-09T14:46:16.7103911Z model pt2e: GraphModule( 2025-09-09T14:46:16.7104196Z (conv): Module() 2025-09-09T14:46:16.7104442Z (bn): Module() 2025-09-09T14:46:16.7104813Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:16.7105978Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:46:16.7107369Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:46:16.7108009Z ) 2025-09-09T14:46:16.7108345Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:16.7109519Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0026]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:46:16.7110897Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3289433717727661, max_val=0.29890719056129456) 2025-09-09T14:46:16.7111541Z ) 2025-09-09T14:46:16.7111880Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:16.7113041Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0055]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:46:16.7114736Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.413901925086975) 2025-09-09T14:46:16.7115332Z ) 2025-09-09T14:46:16.7115703Z ) 2025-09-09T14:46:16.7115824Z 2025-09-09T14:46:16.7115829Z 2025-09-09T14:46:16.7115842Z 2025-09-09T14:46:16.7115947Z def forward(self, x): 2025-09-09T14:46:16.7116290Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:46:16.7116706Z conv_weight = self.conv.weight 2025-09-09T14:46:16.7117040Z conv_bias = self.conv.bias 2025-09-09T14:46:16.7117356Z bn_weight = self.bn.weight 2025-09-09T14:46:16.7117662Z bn_bias = self.bn.bias 2025-09-09T14:46:16.7117970Z bn_running_mean = self.bn.running_mean 2025-09-09T14:46:16.7118388Z bn_running_var = self.bn.running_var 2025-09-09T14:46:16.7118803Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:46:16.7119348Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:46:16.7120074Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:46:16.7120723Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:46:16.7121207Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:46:16.7121702Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:46:16.7122237Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T14:46:16.7122841Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:46:16.7123536Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:46:16.7124288Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:46:16.7125500Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:46:16.7126597Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T14:46:16.7127243Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T14:46:16.7127963Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T14:46:16.7128640Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:46:16.7129778Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:46:16.7130844Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T14:46:16.7140217Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T14:46:16.7140902Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:46:16.7141388Z 2025-09-09T14:46:16.7141739Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:46:16.7142185Z model fx: GraphModule( 2025-09-09T14:46:16.7142594Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:16.7143765Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:46:16.7145152Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:46:16.7145791Z ) 2025-09-09T14:46:16.7146009Z (conv): ConvBnReLU1d( 2025-09-09T14:46:16.7146299Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T14:46:16.7146793Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:46:16.7147360Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:16.7148682Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0026]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:46:16.7150166Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3289433717727661, max_val=0.29890719056129456) 2025-09-09T14:46:16.7150812Z ) 2025-09-09T14:46:16.7151022Z ) 2025-09-09T14:46:16.7151358Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:16.7152537Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0055]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:46:16.7153861Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.413901925086975) 2025-09-09T14:46:16.7154437Z ) 2025-09-09T14:46:16.7154643Z ) 2025-09-09T14:46:16.7154768Z 2025-09-09T14:46:16.7154773Z 2025-09-09T14:46:16.7154778Z 2025-09-09T14:46:16.7154880Z def forward(self, x): 2025-09-09T14:46:16.7155328Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:46:16.7155979Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:46:16.7156673Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:46:16.7157246Z return activation_post_process_1 2025-09-09T14:46:16.7157588Z 2025-09-09T14:46:16.7157947Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:46:16.7158325Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:46:16.7158561Z [0., 0., 0.], 2025-09-09T14:46:16.7158833Z [0., 0., 0.]]], grad_fn=) 2025-09-09T14:46:16.7159166Z converted model pt2e: GraphModule( 2025-09-09T14:46:16.7159426Z (conv): Module() 2025-09-09T14:46:16.7159635Z (bn): Module() 2025-09-09T14:46:16.7159830Z ) 2025-09-09T14:46:16.7159927Z 2025-09-09T14:46:16.7159932Z 2025-09-09T14:46:16.7159942Z 2025-09-09T14:46:16.7160029Z def forward(self, x): 2025-09-09T14:46:16.7160315Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:46:16.7160650Z conv_bias = self.conv.bias 2025-09-09T14:46:16.7160947Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:46:16.7161657Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:46:16.7162900Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:46:16.7163944Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:46:16.7164442Z quantize_per_tensor = self._frozen_param0 2025-09-09T14:46:16.7165232Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.002590105403214693, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T14:46:16.7166485Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T14:46:16.7167332Z relu = torch.ops.aten.relu.default(conv1d_2); conv1d_2 = None 2025-09-09T14:46:16.7168115Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.005544713232666254, -128, -128, 127, torch.int8); relu = None 2025-09-09T14:46:16.7169412Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.005544713232666254, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:46:16.7170515Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:46:16.7170923Z 2025-09-09T14:46:16.7171207Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:46:16.7171651Z onverted model fx: GraphModule( 2025-09-09T14:46:16.7171910Z (conv): ConvReLU1d( 2025-09-09T14:46:16.7172233Z (0): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T14:46:16.7172597Z (1): ReLU() 2025-09-09T14:46:16.7172791Z ) 2025-09-09T14:46:16.7172956Z ) 2025-09-09T14:46:16.7173053Z 2025-09-09T14:46:16.7173058Z 2025-09-09T14:46:16.7173062Z 2025-09-09T14:46:16.7173153Z def forward(self, x): 2025-09-09T14:46:16.7173765Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:46:16.7174994Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:46:16.7176066Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:46:16.7176927Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.005544713232666254, -128, -128, 127, torch.int8); conv = None 2025-09-09T14:46:40.8422347Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.005544713232666254, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:46:40.8423745Z return dequantize_per_tensor_default_1 2025-09-09T14:46:40.8424164Z 2025-09-09T14:46:40.8424545Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:46:40.8425093Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:46:40.8425430Z [0., 0., 0.], 2025-09-09T14:46:40.8425698Z [0., 0., 0.]]]) 2025-09-09T14:46:40.8426150Z PASSED 2025-09-09T14:46:40.8426797Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_bn_relu_fusion_cuda model pt2e: GraphModule( 2025-09-09T14:46:40.8427715Z (conv): Module() 2025-09-09T14:46:40.8427998Z (bn): Module() 2025-09-09T14:46:40.8428404Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:40.8429608Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0104], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:46:40.8430884Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:46:40.8431398Z ) 2025-09-09T14:46:40.8431688Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:40.8432857Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0022, 0.0020, 0.0022], device='cuda:0'), zero_point=tensor([0, 0, 0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:46:40.8434411Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2799, -0.2557, -0.2618], device='cuda:0'), max_val=tensor([0.1970, 0.2308, 0.2775], device='cuda:0')) 2025-09-09T14:46:40.8435144Z ) 2025-09-09T14:46:40.8435422Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:40.8436551Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0055], device='cuda:0'), zero_point=tensor([-128], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:46:40.8438059Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.4123179912567139) 2025-09-09T14:46:40.8438535Z ) 2025-09-09T14:46:40.8438850Z ) 2025-09-09T14:46:40.8438951Z 2025-09-09T14:46:40.8438956Z 2025-09-09T14:46:40.8438959Z 2025-09-09T14:46:40.8439049Z def forward(self, x): 2025-09-09T14:46:40.8439345Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:46:40.8439684Z conv_weight = self.conv.weight 2025-09-09T14:46:40.8439975Z conv_bias = self.conv.bias 2025-09-09T14:46:40.8440241Z bn_weight = self.bn.weight 2025-09-09T14:46:40.8440491Z bn_bias = self.bn.bias 2025-09-09T14:46:40.8440759Z bn_running_mean = self.bn.running_mean 2025-09-09T14:46:40.8441061Z bn_running_var = self.bn.running_var 2025-09-09T14:46:40.8441399Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:46:40.8441838Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:46:40.8442435Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:46:40.8442974Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:46:40.8443374Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:46:40.8443789Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:46:40.8444225Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T14:46:40.8444731Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:46:40.8445292Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:46:40.8445907Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:46:40.8446932Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:46:40.8447811Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T14:46:40.8448351Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T14:46:40.8448924Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T14:46:40.8449476Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:46:40.8450343Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:46:40.8451182Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T14:46:40.8451712Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T14:46:40.8452271Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:46:40.8452662Z 2025-09-09T14:46:40.8452954Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:46:40.8453323Z model fx: GraphModule( 2025-09-09T14:46:40.8453659Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:40.8454768Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0104], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:46:40.8456140Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:46:40.8456659Z ) 2025-09-09T14:46:40.8456847Z (conv): ConvBnReLU1d( 2025-09-09T14:46:40.8457097Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T14:46:40.8457500Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:46:40.8458062Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:40.8459201Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0022, 0.0020, 0.0022], device='cuda:0'), zero_point=tensor([0, 0, 0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:46:40.8460843Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2799, -0.2557, -0.2618], device='cuda:0'), max_val=tensor([0.1970, 0.2308, 0.2775], device='cuda:0')) 2025-09-09T14:46:40.8461589Z ) 2025-09-09T14:46:40.8461764Z ) 2025-09-09T14:46:40.8462051Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:40.8463174Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0055], device='cuda:0'), zero_point=tensor([-128], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:46:40.8464401Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.4123179912567139) 2025-09-09T14:46:40.8464879Z ) 2025-09-09T14:46:40.8465047Z ) 2025-09-09T14:46:40.8465152Z 2025-09-09T14:46:40.8465157Z 2025-09-09T14:46:40.8465161Z 2025-09-09T14:46:40.8465248Z def forward(self, x): 2025-09-09T14:46:40.8465609Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:46:40.8466140Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:46:40.8466691Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:46:40.8467116Z return activation_post_process_1 2025-09-09T14:46:40.8467384Z 2025-09-09T14:46:40.8467668Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:46:40.8468048Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:46:40.8468286Z [0., 0., 0.], 2025-09-09T14:46:40.8468557Z [0., 0., 0.]]], device='cuda:0', grad_fn=) 2025-09-09T14:46:40.8468901Z converted model pt2e: GraphModule( 2025-09-09T14:46:40.8469166Z (conv): Module() 2025-09-09T14:46:40.8469382Z (bn): Module() 2025-09-09T14:46:40.8469577Z ) 2025-09-09T14:46:40.8469678Z 2025-09-09T14:46:40.8469689Z 2025-09-09T14:46:40.8469693Z 2025-09-09T14:46:40.8469782Z def forward(self, x): 2025-09-09T14:46:40.8470064Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:46:40.8470406Z conv_bias = self.conv.bias 2025-09-09T14:46:40.8470711Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:46:40.8471417Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933931648731, 0, -128, 127, torch.int8); x = None 2025-09-09T14:46:40.8472650Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933931648731, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:46:40.8473700Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:46:40.8474171Z _scale_0 = self._scale_0 2025-09-09T14:46:40.8474434Z _zero_point_0 = self._zero_point_0 2025-09-09T14:46:40.8474740Z quantize_per_channel = self._frozen_param0 2025-09-09T14:46:40.8475621Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T14:46:40.8477013Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T14:46:40.8477863Z relu = torch.ops.aten.relu.default(conv1d_2); conv1d_2 = None 2025-09-09T14:46:40.8478725Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.005538502242416143, -128, -128, 127, torch.int8); relu = None 2025-09-09T14:46:40.8480069Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.005538502242416143, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:46:57.2374569Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T14:46:57.2375215Z 2025-09-09T14:46:57.2375601Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:46:57.2376222Z onverted model fx: GraphModule( 2025-09-09T14:46:57.2376585Z (conv): ConvReLU1d( 2025-09-09T14:46:57.2376969Z (0): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T14:46:57.2377425Z (1): ReLU() 2025-09-09T14:46:57.2377626Z ) 2025-09-09T14:46:57.2377799Z ) 2025-09-09T14:46:57.2377948Z 2025-09-09T14:46:57.2377955Z 2025-09-09T14:46:57.2377961Z 2025-09-09T14:46:57.2378083Z def forward(self, x): 2025-09-09T14:46:57.2378739Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933931648731, 0, -128, 127, torch.int8); x = None 2025-09-09T14:46:57.2380173Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933931648731, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:46:57.2381273Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:46:57.2382227Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.005538502242416143, -128, -128, 127, torch.int8); conv = None 2025-09-09T14:46:57.2383707Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.005538502242416143, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:46:57.2384856Z return dequantize_per_tensor_default_1 2025-09-09T14:46:57.2385143Z 2025-09-09T14:46:57.2385426Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:46:57.2385805Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:46:57.2386041Z [0., 0., 0.], 2025-09-09T14:46:57.2386272Z [0., 0., 0.]]], device='cuda:0') 2025-09-09T14:46:57.2386557Z model pt2e: GraphModule( 2025-09-09T14:46:57.2386789Z (conv): Module() 2025-09-09T14:46:57.2387009Z (bn): Module() 2025-09-09T14:46:57.2387315Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:57.2388454Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0104], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:46:57.2389915Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:46:57.2390434Z ) 2025-09-09T14:46:57.2390719Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:57.2391834Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0022], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:46:57.2393119Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.2799264192581177, max_val=0.27745386958122253) 2025-09-09T14:46:57.2393639Z ) 2025-09-09T14:46:57.2393915Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:57.2395286Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0055], device='cuda:0'), zero_point=tensor([-128], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:46:57.2398356Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.4123179912567139) 2025-09-09T14:46:57.2398828Z ) 2025-09-09T14:46:57.2398995Z ) 2025-09-09T14:46:57.2399100Z 2025-09-09T14:46:57.2399105Z 2025-09-09T14:46:57.2399109Z 2025-09-09T14:46:57.2399195Z def forward(self, x): 2025-09-09T14:46:57.2399486Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:46:57.2399823Z conv_weight = self.conv.weight 2025-09-09T14:46:57.2400105Z conv_bias = self.conv.bias 2025-09-09T14:46:57.2400358Z bn_weight = self.bn.weight 2025-09-09T14:46:57.2400614Z bn_bias = self.bn.bias 2025-09-09T14:46:57.2400871Z bn_running_mean = self.bn.running_mean 2025-09-09T14:46:57.2401177Z bn_running_var = self.bn.running_var 2025-09-09T14:46:57.2401522Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:46:57.2401963Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:46:57.2402559Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:46:57.2403081Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:46:57.2403479Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:46:57.2403889Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:46:57.2404331Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T14:46:57.2404834Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:46:57.2405396Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:46:57.2406025Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:46:57.2407001Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:46:57.2407895Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T14:46:57.2408434Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T14:46:57.2409005Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T14:46:57.2409563Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:46:57.2410432Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:46:57.2411290Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T14:46:57.2411817Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T14:46:57.2412407Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:46:57.2412806Z 2025-09-09T14:46:57.2413090Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:46:57.2413470Z model fx: GraphModule( 2025-09-09T14:46:57.2413794Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:57.2414929Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0104], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:46:57.2416275Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:46:57.2416784Z ) 2025-09-09T14:46:57.2417127Z (conv): ConvBnReLU1d( 2025-09-09T14:46:57.2417370Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T14:46:57.2417780Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:46:57.2418351Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:57.2419448Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0022], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:46:57.2420736Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.2799264192581177, max_val=0.27745386958122253) 2025-09-09T14:46:57.2421250Z ) 2025-09-09T14:46:57.2421438Z ) 2025-09-09T14:46:57.2421717Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:46:57.2422845Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0055], device='cuda:0'), zero_point=tensor([-128], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:46:57.2424087Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.4123179912567139) 2025-09-09T14:46:57.2424554Z ) 2025-09-09T14:46:57.2424728Z ) 2025-09-09T14:46:57.2424827Z 2025-09-09T14:46:57.2424832Z 2025-09-09T14:46:57.2424836Z 2025-09-09T14:46:57.2424930Z def forward(self, x): 2025-09-09T14:46:57.2425280Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:46:57.2425820Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:46:57.2426368Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:46:57.2426800Z return activation_post_process_1 2025-09-09T14:46:57.2427069Z 2025-09-09T14:46:57.2427355Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:46:57.2427725Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:46:57.2427974Z [0., 0., 0.], 2025-09-09T14:46:57.2428244Z [0., 0., 0.]]], device='cuda:0', grad_fn=) 2025-09-09T14:46:57.2428576Z converted model pt2e: GraphModule( 2025-09-09T14:46:57.2428846Z (conv): Module() 2025-09-09T14:46:57.2429052Z (bn): Module() 2025-09-09T14:46:57.2429252Z ) 2025-09-09T14:46:57.2429353Z 2025-09-09T14:46:57.2429357Z 2025-09-09T14:46:57.2429361Z 2025-09-09T14:46:57.2429449Z def forward(self, x): 2025-09-09T14:46:57.2429738Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:46:57.2430074Z conv_bias = self.conv.bias 2025-09-09T14:46:57.2430384Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:46:57.2431098Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933931648731, 0, -128, 127, torch.int8); x = None 2025-09-09T14:46:57.2432380Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933931648731, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:46:57.2433424Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:46:57.2433929Z quantize_per_tensor = self._frozen_param0 2025-09-09T14:46:57.2434715Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.002204145072028041, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T14:47:21.4195369Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T14:47:21.4197632Z relu = torch.ops.aten.relu.default(conv1d_2); conv1d_2 = None 2025-09-09T14:47:21.4199216Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.005538502242416143, -128, -128, 127, torch.int8); relu = None 2025-09-09T14:47:21.4201026Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.005538502242416143, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:47:21.4202416Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:47:21.4202992Z 2025-09-09T14:47:21.4203301Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:47:21.4203691Z onverted model fx: GraphModule( 2025-09-09T14:47:21.4203956Z (conv): ConvReLU1d( 2025-09-09T14:47:21.4204300Z (0): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T14:47:21.4204666Z (1): ReLU() 2025-09-09T14:47:21.4204872Z ) 2025-09-09T14:47:21.4205051Z ) 2025-09-09T14:47:21.4205168Z 2025-09-09T14:47:21.4205173Z 2025-09-09T14:47:21.4205177Z 2025-09-09T14:47:21.4205266Z def forward(self, x): 2025-09-09T14:47:21.4205894Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933931648731, 0, -128, 127, torch.int8); x = None 2025-09-09T14:47:21.4207137Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933931648731, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:47:21.4208169Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:47:21.4209035Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.005538502242416143, -128, -128, 127, torch.int8); conv = None 2025-09-09T14:47:21.4210375Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.005538502242416143, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:47:21.4211271Z return dequantize_per_tensor_default_1 2025-09-09T14:47:21.4211553Z 2025-09-09T14:47:21.4211845Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:47:21.4212212Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:47:21.4212458Z [0., 0., 0.], 2025-09-09T14:47:21.4212691Z [0., 0., 0.]]], device='cuda:0') 2025-09-09T14:47:21.4213160Z PASSED 2025-09-09T14:47:21.4213797Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_bn_relu_fusion_no_conv_bias model pt2e: GraphModule( 2025-09-09T14:47:21.4214458Z (conv): Module() 2025-09-09T14:47:21.4214675Z (bn): Module() 2025-09-09T14:47:21.4214980Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:21.4215996Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:47:21.4217101Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:47:21.4217614Z ) 2025-09-09T14:47:21.4217908Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:21.4218896Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0026, 0.0026, 0.0025]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:47:21.4220237Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.3276, -0.3045, -0.2418]), max_val=tensor([0.2760, 0.3298, 0.3101])) 2025-09-09T14:47:21.4220888Z ) 2025-09-09T14:47:21.4221167Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:21.4222217Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0054]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:47:21.4223345Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.3778926134109497) 2025-09-09T14:47:21.4223818Z ) 2025-09-09T14:47:21.4223998Z ) 2025-09-09T14:47:21.4224098Z 2025-09-09T14:47:21.4224102Z 2025-09-09T14:47:21.4224107Z 2025-09-09T14:47:21.4224196Z def forward(self, x): 2025-09-09T14:47:21.4224492Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:47:21.4224836Z conv_weight = self.conv.weight 2025-09-09T14:47:21.4225124Z bn_weight = self.bn.weight 2025-09-09T14:47:21.4225375Z bn_bias = self.bn.bias 2025-09-09T14:47:21.4225640Z bn_running_mean = self.bn.running_mean 2025-09-09T14:47:21.4225941Z bn_running_var = self.bn.running_var 2025-09-09T14:47:21.4226287Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:47:21.4226728Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:47:21.4227321Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:47:21.4227852Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:47:21.4228245Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:47:21.4228664Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:47:21.4229105Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T14:47:21.4229611Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:47:21.4230176Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:47:21.4231015Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T14:47:21.4231837Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T14:47:21.4232384Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T14:47:21.4233274Z batch_norm_1 = torch.ops.aten.batch_norm.default(div_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); div_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:47:21.4234124Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T14:47:21.4234654Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T14:47:21.4235197Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:47:21.4235593Z 2025-09-09T14:47:21.4235881Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:47:21.4236259Z model fx: GraphModule( 2025-09-09T14:47:21.4236590Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:21.4237539Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:47:21.4238638Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:47:21.4239149Z ) 2025-09-09T14:47:21.4239347Z (conv): ConvBnReLU1d( 2025-09-09T14:47:21.4239637Z 3, 3, kernel_size=(3,), stride=(1,), bias=False 2025-09-09T14:47:21.4240085Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:47:21.4240561Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:21.4241602Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0026, 0.0026, 0.0025]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:47:21.4242898Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.3276, -0.3045, -0.2418]), max_val=tensor([0.2760, 0.3298, 0.3101])) 2025-09-09T14:47:21.4243632Z ) 2025-09-09T14:47:21.4243812Z ) 2025-09-09T14:47:21.4244097Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:21.4245046Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0054]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:47:21.4246102Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.3778926134109497) 2025-09-09T14:47:21.4253248Z ) 2025-09-09T14:47:21.4253449Z ) 2025-09-09T14:47:21.4253559Z 2025-09-09T14:47:21.4253563Z 2025-09-09T14:47:21.4253567Z 2025-09-09T14:47:21.4253664Z def forward(self, x): 2025-09-09T14:47:21.4254026Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:47:21.4254570Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:47:21.4255121Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:47:21.4255550Z return activation_post_process_1 2025-09-09T14:47:21.4255808Z 2025-09-09T14:47:21.4256144Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:47:21.4256507Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:47:21.4256740Z [0., 0., 0.], 2025-09-09T14:47:21.4256975Z [0., 0., 0.]]], grad_fn=) 2025-09-09T14:47:21.4257278Z converted model pt2e: GraphModule( 2025-09-09T14:47:21.4257533Z (conv): Module() 2025-09-09T14:47:21.4257735Z (bn): Module() 2025-09-09T14:47:21.4257923Z ) 2025-09-09T14:47:21.4258024Z 2025-09-09T14:47:21.4258028Z 2025-09-09T14:47:21.4258036Z 2025-09-09T14:47:21.4258121Z def forward(self, x): 2025-09-09T14:47:21.4258402Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:47:21.4258781Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:47:21.4259489Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:47:21.4260713Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:47:21.4261751Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:47:21.4262215Z _scale_0 = self._scale_0 2025-09-09T14:47:21.4262467Z _zero_point_0 = self._zero_point_0 2025-09-09T14:47:21.4262767Z quantize_per_channel = self._frozen_param0 2025-09-09T14:47:40.3544340Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T14:47:40.3545532Z conv_weight_bias = self.conv.weight_bias 2025-09-09T14:47:40.3546584Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_weight_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_weight_bias = None 2025-09-09T14:47:40.3547720Z relu = torch.ops.aten.relu.default(conv1d_2); conv1d_2 = None 2025-09-09T14:47:40.3548701Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.0054035005159676075, -128, -128, 127, torch.int8); relu = None 2025-09-09T14:47:40.3550570Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.0054035005159676075, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:47:40.3551859Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T14:47:40.3552540Z 2025-09-09T14:47:40.3552879Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:47:40.3553342Z onverted model fx: GraphModule( 2025-09-09T14:47:40.3553649Z (conv): ConvReLU1d( 2025-09-09T14:47:40.3554052Z (0): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T14:47:40.3554497Z (1): ReLU() 2025-09-09T14:47:40.3554736Z ) 2025-09-09T14:47:40.3554941Z ) 2025-09-09T14:47:40.3555066Z 2025-09-09T14:47:40.3555071Z 2025-09-09T14:47:40.3555076Z 2025-09-09T14:47:40.3555181Z def forward(self, x): 2025-09-09T14:47:40.3555950Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:47:40.3557518Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:47:40.3558801Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:47:40.3559883Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.0054035005159676075, -128, -128, 127, torch.int8); conv = None 2025-09-09T14:47:40.3561507Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.0054035005159676075, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:47:40.3562633Z return dequantize_per_tensor_default_1 2025-09-09T14:47:40.3562969Z 2025-09-09T14:47:40.3563318Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:47:40.3563760Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:47:40.3564057Z [0., 0., 0.], 2025-09-09T14:47:40.3564318Z [0., 0., 0.]]]) 2025-09-09T14:47:40.3564591Z model pt2e: GraphModule( 2025-09-09T14:47:40.3564877Z (conv): Module() 2025-09-09T14:47:40.3565128Z (bn): Module() 2025-09-09T14:47:40.3565497Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:40.3566663Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:47:40.3568039Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:47:40.3568678Z ) 2025-09-09T14:47:40.3569012Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:40.3570192Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0026]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:47:40.3571585Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.32764676213264465, max_val=0.3298276662826538) 2025-09-09T14:47:40.3572234Z ) 2025-09-09T14:47:40.3572571Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:40.3573725Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0054]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:47:40.3575046Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.3749419450759888) 2025-09-09T14:47:40.3575627Z ) 2025-09-09T14:47:40.3575838Z ) 2025-09-09T14:47:40.3576061Z 2025-09-09T14:47:40.3576067Z 2025-09-09T14:47:40.3576071Z 2025-09-09T14:47:40.3576184Z def forward(self, x): 2025-09-09T14:47:40.3576628Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:47:40.3577051Z conv_weight = self.conv.weight 2025-09-09T14:47:40.3577387Z bn_weight = self.bn.weight 2025-09-09T14:47:40.3577787Z bn_bias = self.bn.bias 2025-09-09T14:47:40.3578102Z bn_running_mean = self.bn.running_mean 2025-09-09T14:47:40.3578472Z bn_running_var = self.bn.running_var 2025-09-09T14:47:40.3578879Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:47:40.3579415Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:47:40.3580142Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:47:40.3580786Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:47:40.3581263Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:47:40.3581755Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:47:40.3582293Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T14:47:40.3582915Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:47:40.3583602Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:47:40.3584655Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T14:47:40.3585677Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T14:47:40.3586334Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T14:47:40.3587442Z batch_norm_1 = torch.ops.aten.batch_norm.default(div_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); div_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:47:40.3588492Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T14:47:40.3589149Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T14:47:40.3589817Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:47:40.3590308Z 2025-09-09T14:47:40.3590648Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:47:40.3591101Z model fx: GraphModule( 2025-09-09T14:47:40.3591494Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:40.3592657Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:47:40.3594035Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:47:40.3594665Z ) 2025-09-09T14:47:40.3594899Z (conv): ConvBnReLU1d( 2025-09-09T14:47:40.3595214Z 3, 3, kernel_size=(3,), stride=(1,), bias=False 2025-09-09T14:47:40.3595744Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:47:40.3596329Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:40.3597784Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0026]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:47:40.3599198Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.32764676213264465, max_val=0.3298276662826538) 2025-09-09T14:47:40.3599898Z ) 2025-09-09T14:47:40.3600126Z ) 2025-09-09T14:47:40.3600488Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:40.3601615Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0054]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:47:40.3602698Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.3749419450759888) 2025-09-09T14:47:40.3603273Z ) 2025-09-09T14:47:40.3603448Z ) 2025-09-09T14:47:40.3603547Z 2025-09-09T14:47:40.3603552Z 2025-09-09T14:47:40.3603556Z 2025-09-09T14:47:40.3603649Z def forward(self, x): 2025-09-09T14:47:40.3603999Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:47:40.3604537Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:47:40.3605079Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:47:40.3605511Z return activation_post_process_1 2025-09-09T14:47:40.3605797Z 2025-09-09T14:47:40.3606103Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:47:40.3606470Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:47:40.3606713Z [0., 0., 0.], 2025-09-09T14:47:40.3606965Z [0., 0., 0.]]], grad_fn=) 2025-09-09T14:47:40.3607269Z converted model pt2e: GraphModule( 2025-09-09T14:47:40.3607539Z (conv): Module() 2025-09-09T14:47:40.3607752Z (bn): Module() 2025-09-09T14:47:40.3607952Z ) 2025-09-09T14:47:40.3608053Z 2025-09-09T14:47:40.3608058Z 2025-09-09T14:47:40.3608062Z 2025-09-09T14:47:40.3608147Z def forward(self, x): 2025-09-09T14:47:40.3608435Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:47:40.3608816Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:47:40.3609526Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:47:40.3610760Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:47:40.3611804Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:47:40.3612304Z quantize_per_tensor = self._frozen_param0 2025-09-09T14:47:41.7840078Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.002597068203613162, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T14:47:41.7841144Z conv_weight_bias = self.conv.weight_bias 2025-09-09T14:47:41.7842202Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_weight_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_weight_bias = None 2025-09-09T14:47:41.7843330Z relu = torch.ops.aten.relu.default(conv1d_2); conv1d_2 = None 2025-09-09T14:47:41.7844315Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.0053919292986392975, -128, -128, 127, torch.int8); relu = None 2025-09-09T14:47:41.7845972Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.0053919292986392975, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:47:41.7847321Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:47:41.7847832Z 2025-09-09T14:47:41.7848184Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:47:41.7848645Z onverted model fx: GraphModule( 2025-09-09T14:47:41.7848966Z (conv): ConvReLU1d( 2025-09-09T14:47:41.7849375Z (0): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T14:47:41.7849823Z (1): ReLU() 2025-09-09T14:47:41.7850063Z ) 2025-09-09T14:47:41.7850270Z ) 2025-09-09T14:47:41.7850390Z 2025-09-09T14:47:41.7850402Z 2025-09-09T14:47:41.7850407Z 2025-09-09T14:47:41.7850516Z def forward(self, x): 2025-09-09T14:47:41.7851477Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:47:41.7853048Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:47:41.7854445Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:47:41.7855521Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.0053919292986392975, -128, -128, 127, torch.int8); conv = None 2025-09-09T14:47:41.7857279Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.0053919292986392975, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:47:41.7858402Z return dequantize_per_tensor_default_1 2025-09-09T14:47:41.7858738Z 2025-09-09T14:47:41.7859092Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:47:41.7859541Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:47:41.7859834Z [0., 0., 0.], 2025-09-09T14:47:41.7860107Z [0., 0., 0.]]]) 2025-09-09T14:47:41.7860600Z PASSED 2025-09-09T14:47:41.7861304Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_no_bias model pt2e: GraphModule( 2025-09-09T14:47:41.7862067Z (conv): Module() 2025-09-09T14:47:41.7862447Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:41.7863682Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024, 0.0022, 0.0021]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:47:41.7865311Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.3119, -0.2799, -0.2618]), max_val=tensor([0.1970, 0.1855, 0.2308])) 2025-09-09T14:47:41.7866122Z ) 2025-09-09T14:47:41.7866468Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:41.7867646Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:47:41.7869016Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:47:41.7869664Z ) 2025-09-09T14:47:41.7870004Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:41.7871173Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0038]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:47:41.7872504Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.9578298330307007) 2025-09-09T14:47:41.7873097Z ) 2025-09-09T14:47:41.7873312Z ) 2025-09-09T14:47:41.7873434Z 2025-09-09T14:47:41.7873445Z 2025-09-09T14:47:41.7873450Z 2025-09-09T14:47:41.7873557Z def forward(self, x): 2025-09-09T14:47:41.7873911Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:47:41.7874332Z conv_weight = self.conv.weight 2025-09-09T14:47:41.7874893Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T14:47:41.7875626Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:47:41.7876671Z conv1d = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T14:47:41.7877627Z relu = torch.ops.aten.relu.default(conv1d); conv1d = None 2025-09-09T14:47:41.7878319Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T14:47:41.7878995Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:47:41.7879477Z 2025-09-09T14:47:41.7879899Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:47:41.7880351Z model fx: GraphModule( 2025-09-09T14:47:41.7880743Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:41.7881917Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:47:41.7883292Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:47:41.7883928Z ) 2025-09-09T14:47:41.7884157Z (conv): ConvReLU1d( 2025-09-09T14:47:41.7884459Z 3, 3, kernel_size=(3,), stride=(1,), bias=False 2025-09-09T14:47:41.7884916Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:41.7886116Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024, 0.0022, 0.0021]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:47:41.7887780Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.3119, -0.2799, -0.2618]), max_val=tensor([0.1970, 0.1855, 0.2308])) 2025-09-09T14:47:41.7888592Z ) 2025-09-09T14:47:41.7888806Z ) 2025-09-09T14:47:41.7889147Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:41.7890327Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0038]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:47:41.7891663Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.9578298330307007) 2025-09-09T14:47:41.7892252Z ) 2025-09-09T14:47:41.7892458Z ) 2025-09-09T14:47:41.7892587Z 2025-09-09T14:47:41.7892600Z 2025-09-09T14:47:41.7892605Z 2025-09-09T14:47:41.7892714Z def forward(self, x): 2025-09-09T14:47:41.7893146Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:47:41.7893810Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:47:41.7894498Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:47:41.7895027Z return activation_post_process_1 2025-09-09T14:47:41.7895356Z 2025-09-09T14:47:41.7895701Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:47:41.7896268Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:47:41.7896572Z [0., 0., 0.], 2025-09-09T14:47:41.7896886Z [0., 0., 0.]]], grad_fn=) 2025-09-09T14:47:41.7897217Z converted model pt2e: GraphModule( 2025-09-09T14:47:41.7897657Z (conv): Module() 2025-09-09T14:47:41.7897866Z ) 2025-09-09T14:47:41.7897967Z 2025-09-09T14:47:41.7897971Z 2025-09-09T14:47:41.7897983Z 2025-09-09T14:47:41.7898072Z def forward(self, x): 2025-09-09T14:47:41.7898365Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:47:41.7898700Z _scale_0 = self._scale_0 2025-09-09T14:47:41.7898963Z _zero_point_0 = self._zero_point_0 2025-09-09T14:47:41.7899292Z quantize_per_channel_default = self._frozen_param0 2025-09-09T14:47:41.7900298Z dequantize_per_channel_default = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel_default, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel_default = _scale_0 = _zero_point_0 = None 2025-09-09T14:47:41.7901641Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:47:41.7903030Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:47:41.7904490Z conv1d = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_channel_default); dequantize_per_tensor_default = dequantize_per_channel_default = None 2025-09-09T14:47:41.7905332Z relu = torch.ops.aten.relu.default(conv1d); conv1d = None 2025-09-09T14:47:41.7906142Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.0037561955396085978, -128, -128, 127, torch.int8); relu = None 2025-09-09T14:47:41.7907471Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.0037561955396085978, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:47:41.7908509Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T14:47:41.7908925Z 2025-09-09T14:47:41.7909226Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:47:41.7909618Z onverted model fx: GraphModule( 2025-09-09T14:47:41.7909888Z (conv): ConvReLU1d( 2025-09-09T14:47:41.7910256Z (0): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,), bias=False) 2025-09-09T14:47:41.7910660Z (1): ReLU() 2025-09-09T14:47:41.7910858Z ) 2025-09-09T14:47:41.7911041Z ) 2025-09-09T14:47:41.7911141Z 2025-09-09T14:47:44.4238198Z 2025-09-09T14:47:44.4238209Z 2025-09-09T14:47:44.4238588Z def forward(self, x): 2025-09-09T14:47:44.4239433Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:47:44.4241052Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:47:44.4242325Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:47:44.4243436Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.0037561955396085978, -128, -128, 127, torch.int8); conv = None 2025-09-09T14:47:44.4245097Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.0037561955396085978, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:47:44.4246227Z return dequantize_per_tensor_default_1 2025-09-09T14:47:44.4246624Z 2025-09-09T14:47:44.4246964Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:47:44.4247425Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:47:44.4247716Z [0., 0., 0.], 2025-09-09T14:47:44.4247981Z [0., 0., 0.]]]) 2025-09-09T14:47:44.4248263Z model pt2e: GraphModule( 2025-09-09T14:47:44.4248552Z (conv): Module() 2025-09-09T14:47:44.4248917Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:44.4250121Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:47:44.4251561Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3119288384914398, max_val=0.23078612983226776) 2025-09-09T14:47:44.4252209Z ) 2025-09-09T14:47:44.4252549Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:44.4253712Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:47:44.4255450Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:47:44.4256177Z ) 2025-09-09T14:47:44.4256548Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:44.4257895Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0037]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:47:44.4259234Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.9556508660316467) 2025-09-09T14:47:44.4259817Z ) 2025-09-09T14:47:44.4260031Z ) 2025-09-09T14:47:44.4260154Z 2025-09-09T14:47:44.4260159Z 2025-09-09T14:47:44.4260164Z 2025-09-09T14:47:44.4260269Z def forward(self, x): 2025-09-09T14:47:44.4260617Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:47:44.4261032Z conv_weight = self.conv.weight 2025-09-09T14:47:44.4261610Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T14:47:44.4262341Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:47:44.4263355Z conv1d = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T14:47:44.4264301Z relu = torch.ops.aten.relu.default(conv1d); conv1d = None 2025-09-09T14:47:44.4264894Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T14:47:44.4265576Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:47:44.4266055Z 2025-09-09T14:47:44.4266403Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:47:44.4266873Z model fx: GraphModule( 2025-09-09T14:47:44.4267292Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:44.4268468Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:47:44.4269852Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:47:44.4270498Z ) 2025-09-09T14:47:44.4270732Z (conv): ConvReLU1d( 2025-09-09T14:47:44.4271032Z 3, 3, kernel_size=(3,), stride=(1,), bias=False 2025-09-09T14:47:44.4271469Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:44.4272616Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:47:44.4274028Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3119288384914398, max_val=0.23078612983226776) 2025-09-09T14:47:44.4274671Z ) 2025-09-09T14:47:44.4274897Z ) 2025-09-09T14:47:44.4275229Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:44.4276398Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0037]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:47:44.4277517Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.9556508660316467) 2025-09-09T14:47:44.4277982Z ) 2025-09-09T14:47:44.4278157Z ) 2025-09-09T14:47:44.4278256Z 2025-09-09T14:47:44.4278260Z 2025-09-09T14:47:44.4278264Z 2025-09-09T14:47:44.4278355Z def forward(self, x): 2025-09-09T14:47:44.4278700Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:47:44.4279240Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:47:44.4279866Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:47:44.4280302Z return activation_post_process_1 2025-09-09T14:47:44.4280562Z 2025-09-09T14:47:44.4280844Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:47:44.4283260Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:47:44.4283496Z [0., 0., 0.], 2025-09-09T14:47:44.4283738Z [0., 0., 0.]]], grad_fn=) 2025-09-09T14:47:44.4284041Z converted model pt2e: GraphModule( 2025-09-09T14:47:44.4284308Z (conv): Module() 2025-09-09T14:47:44.4284503Z ) 2025-09-09T14:47:44.4284608Z 2025-09-09T14:47:44.4284612Z 2025-09-09T14:47:44.4284616Z 2025-09-09T14:47:44.4284702Z def forward(self, x): 2025-09-09T14:47:44.4284985Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:47:44.4285365Z quantize_per_tensor_default = self._frozen_param0 2025-09-09T14:47:44.4286283Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.0024561325553804636, 0, -127, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:47:44.4287582Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:47:44.4288829Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:47:44.4290159Z conv1d = torch.ops.aten.conv1d.default(dequantize_per_tensor_default_1, dequantize_per_tensor_default); dequantize_per_tensor_default_1 = dequantize_per_tensor_default = None 2025-09-09T14:47:44.4290987Z relu = torch.ops.aten.relu.default(conv1d); conv1d = None 2025-09-09T14:47:44.4291757Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.003747650422155857, -128, -128, 127, torch.int8); relu = None 2025-09-09T14:47:44.4293042Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.003747650422155857, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:47:44.4294066Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:47:44.4294483Z 2025-09-09T14:47:44.4294764Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:47:44.4295147Z onverted model fx: GraphModule( 2025-09-09T14:47:44.4295407Z (conv): ConvReLU1d( 2025-09-09T14:47:44.4295780Z (0): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,), bias=False) 2025-09-09T14:47:44.4296241Z (1): ReLU() 2025-09-09T14:47:44.4296441Z ) 2025-09-09T14:47:44.4296612Z ) 2025-09-09T14:47:44.4296719Z 2025-09-09T14:47:44.4296723Z 2025-09-09T14:47:44.4296727Z 2025-09-09T14:47:44.4296813Z def forward(self, x): 2025-09-09T14:47:44.4297653Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:47:44.4298881Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:47:44.4299900Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:47:44.4300765Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.003747650422155857, -128, -128, 127, torch.int8); conv = None 2025-09-09T14:47:44.4302043Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.003747650422155857, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:47:44.4302933Z return dequantize_per_tensor_default_1 2025-09-09T14:47:44.4303210Z 2025-09-09T14:47:44.4303633Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:47:44.4304007Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:47:44.4304343Z [0., 0., 0.], 2025-09-09T14:47:44.4304556Z [0., 0., 0.]]]) 2025-09-09T14:47:44.4304786Z model pt2e: GraphModule( 2025-09-09T14:47:44.4305024Z (conv): Module() 2025-09-09T14:47:44.4305327Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:44.4306324Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0022, 0.0025, 0.0024]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:47:45.8447321Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2203, -0.3233, -0.3086]), max_val=tensor([0.2796, 0.3026, 0.2405])) 2025-09-09T14:47:45.8448218Z ) 2025-09-09T14:47:45.8448571Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:45.8449745Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:47:45.8451153Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:47:45.8451797Z ) 2025-09-09T14:47:45.8452142Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:45.8453319Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0087]), zero_point=tensor([-25], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:47:45.8454717Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.8935509324073792, max_val=1.3209781646728516) 2025-09-09T14:47:45.8455363Z ) 2025-09-09T14:47:45.8455575Z ) 2025-09-09T14:47:45.8455695Z 2025-09-09T14:47:45.8455706Z 2025-09-09T14:47:45.8455711Z 2025-09-09T14:47:45.8455816Z def forward(self, x): 2025-09-09T14:47:45.8456286Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:47:45.8456706Z conv_weight = self.conv.weight 2025-09-09T14:47:45.8457272Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T14:47:45.8458058Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:47:45.8459072Z conv1d = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T14:47:45.8460110Z activation_post_process_2 = self.activation_post_process_2(conv1d); conv1d = None 2025-09-09T14:47:45.8460804Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:47:45.8461285Z 2025-09-09T14:47:45.8461633Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:47:45.8462087Z model fx: GraphModule( 2025-09-09T14:47:45.8462474Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:45.8463645Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:47:45.8465015Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:47:45.8465659Z ) 2025-09-09T14:47:45.8473955Z (conv): Conv1d( 2025-09-09T14:47:45.8474299Z 3, 3, kernel_size=(3,), stride=(1,), bias=False 2025-09-09T14:47:45.8474755Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:45.8476168Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0022, 0.0025, 0.0024]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:47:45.8477876Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2203, -0.3233, -0.3086]), max_val=tensor([0.2796, 0.3026, 0.2405])) 2025-09-09T14:47:45.8478826Z ) 2025-09-09T14:47:45.8479050Z ) 2025-09-09T14:47:45.8479392Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:45.8480576Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0087]), zero_point=tensor([-25], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:47:45.8481988Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.8935509324073792, max_val=1.3209781646728516) 2025-09-09T14:47:45.8482638Z ) 2025-09-09T14:47:45.8482849Z ) 2025-09-09T14:47:45.8482969Z 2025-09-09T14:47:45.8482975Z 2025-09-09T14:47:45.8482987Z 2025-09-09T14:47:45.8483100Z def forward(self, x): 2025-09-09T14:47:45.8483528Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:47:45.8484201Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:47:45.8484878Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:47:45.8485411Z return activation_post_process_1 2025-09-09T14:47:45.8485734Z 2025-09-09T14:47:45.8486073Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:47:45.8486527Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:47:45.8486808Z [0., 0., 0.], 2025-09-09T14:47:45.8487099Z [0., 0., 0.]]], grad_fn=) 2025-09-09T14:47:45.8487493Z converted model pt2e: GraphModule( 2025-09-09T14:47:45.8487844Z (conv): Module() 2025-09-09T14:47:45.8488083Z ) 2025-09-09T14:47:45.8488213Z 2025-09-09T14:47:45.8488218Z 2025-09-09T14:47:45.8488223Z 2025-09-09T14:47:45.8488331Z def forward(self, x): 2025-09-09T14:47:45.8488684Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:47:45.8489091Z _scale_0 = self._scale_0 2025-09-09T14:47:45.8489407Z _zero_point_0 = self._zero_point_0 2025-09-09T14:47:45.8489798Z quantize_per_channel_default = self._frozen_param0 2025-09-09T14:47:45.8491058Z dequantize_per_channel_default = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel_default, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel_default = _scale_0 = _zero_point_0 = None 2025-09-09T14:47:45.8492739Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:47:45.8494297Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:47:45.8496128Z conv1d = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_channel_default); dequantize_per_tensor_default = dequantize_per_channel_default = None 2025-09-09T14:47:45.8497759Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d, 0.008684427477419376, -25, -128, 127, torch.int8); conv1d = None 2025-09-09T14:47:45.8499057Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.008684427477419376, -25, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:47:45.8500072Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T14:47:45.8500483Z 2025-09-09T14:47:45.8500771Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:47:45.8501155Z onverted model fx: GraphModule( 2025-09-09T14:47:45.8501706Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,), bias=False) 2025-09-09T14:47:45.8502125Z ) 2025-09-09T14:47:45.8502226Z 2025-09-09T14:47:45.8502230Z 2025-09-09T14:47:45.8502335Z 2025-09-09T14:47:45.8502424Z def forward(self, x): 2025-09-09T14:47:45.8503053Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:47:45.8504299Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:47:45.8505326Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:47:45.8506190Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.008684427477419376, -25, -128, 127, torch.int8); conv = None 2025-09-09T14:47:45.8507474Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.008684427477419376, -25, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:47:45.8508370Z return dequantize_per_tensor_default_1 2025-09-09T14:47:45.8508650Z 2025-09-09T14:47:45.8508930Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:47:45.8509303Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:47:45.8509538Z [0., 0., 0.], 2025-09-09T14:47:45.8509757Z [0., 0., 0.]]]) 2025-09-09T14:47:45.8509989Z model pt2e: GraphModule( 2025-09-09T14:47:45.8510223Z (conv): Module() 2025-09-09T14:47:45.8510531Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:45.8511490Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0025]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:47:45.8512621Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3232726454734802, max_val=0.30256539583206177) 2025-09-09T14:47:45.8513142Z ) 2025-09-09T14:47:45.8513427Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:45.8514358Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:47:45.8515453Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:47:45.8515968Z ) 2025-09-09T14:47:45.8516251Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:47:45.8517197Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0087]), zero_point=tensor([-26], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:47:45.8518342Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.887858510017395, max_val=1.3209781646728516) 2025-09-09T14:47:45.8518860Z ) 2025-09-09T14:47:45.8519032Z ) 2025-09-09T14:47:45.8519130Z 2025-09-09T14:47:45.8519134Z 2025-09-09T14:47:45.8519138Z 2025-09-09T14:47:45.8519224Z def forward(self, x): 2025-09-09T14:47:45.8519518Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:47:45.8519855Z conv_weight = self.conv.weight 2025-09-09T14:47:45.8520321Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T14:47:45.8520904Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:47:45.8521720Z conv1d = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T14:48:58.0153793Z activation_post_process_2 = self.activation_post_process_2(conv1d); conv1d = None 2025-09-09T14:48:58.0156248Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:48:58.0157195Z 2025-09-09T14:48:58.0157591Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:48:58.0157968Z model fx: GraphModule( 2025-09-09T14:48:58.0158298Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:48:58.0159246Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:48:58.0160618Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:48:58.0161152Z ) 2025-09-09T14:48:58.0161340Z (conv): Conv1d( 2025-09-09T14:48:58.0161589Z 3, 3, kernel_size=(3,), stride=(1,), bias=False 2025-09-09T14:48:58.0161967Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:48:58.0162890Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0025]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:48:58.0164028Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3232726454734802, max_val=0.30256539583206177) 2025-09-09T14:48:58.0164546Z ) 2025-09-09T14:48:58.0164722Z ) 2025-09-09T14:48:58.0165006Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:48:58.0165945Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0087]), zero_point=tensor([-26], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:48:58.0167049Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.887858510017395, max_val=1.3209781646728516) 2025-09-09T14:48:58.0167553Z ) 2025-09-09T14:48:58.0167727Z ) 2025-09-09T14:48:58.0167826Z 2025-09-09T14:48:58.0167834Z 2025-09-09T14:48:58.0167838Z 2025-09-09T14:48:58.0167931Z def forward(self, x): 2025-09-09T14:48:58.0168281Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:48:58.0168818Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:48:58.0169364Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:48:58.0169798Z return activation_post_process_1 2025-09-09T14:48:58.0170062Z 2025-09-09T14:48:58.0170345Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:48:58.0170761Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:48:58.0171004Z [0., 0., 0.], 2025-09-09T14:48:58.0171249Z [0., 0., 0.]]], grad_fn=) 2025-09-09T14:48:58.0171554Z converted model pt2e: GraphModule( 2025-09-09T14:48:58.0171832Z (conv): Module() 2025-09-09T14:48:58.0172032Z ) 2025-09-09T14:48:58.0172134Z 2025-09-09T14:48:58.0172144Z 2025-09-09T14:48:58.0172153Z 2025-09-09T14:48:58.0172238Z def forward(self, x): 2025-09-09T14:48:58.0172524Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:48:58.0172905Z quantize_per_tensor_default = self._frozen_param0 2025-09-09T14:48:58.0173813Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.002545453840866685, 0, -127, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:48:58.0175046Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:48:58.0176488Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:48:58.0177849Z conv1d = torch.ops.aten.conv1d.default(dequantize_per_tensor_default_1, dequantize_per_tensor_default); dequantize_per_tensor_default_1 = dequantize_per_tensor_default = None 2025-09-09T14:48:58.0179106Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d, 0.008662103675305843, -26, -128, 127, torch.int8); conv1d = None 2025-09-09T14:48:58.0180447Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.008662103675305843, -26, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:48:58.0181460Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:48:58.0181870Z 2025-09-09T14:48:58.0182154Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:48:58.0182529Z onverted model fx: GraphModule( 2025-09-09T14:48:58.0182946Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,), bias=False) 2025-09-09T14:48:58.0183346Z ) 2025-09-09T14:48:58.0183453Z 2025-09-09T14:48:58.0183462Z 2025-09-09T14:48:58.0183467Z 2025-09-09T14:48:58.0183553Z def forward(self, x): 2025-09-09T14:48:58.0184173Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:48:58.0185400Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:48:58.0186411Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:48:58.0187261Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.008662103675305843, -26, -128, 127, torch.int8); conv = None 2025-09-09T14:48:58.0188549Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.008662103675305843, -26, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:48:58.0189440Z return dequantize_per_tensor_default_1 2025-09-09T14:48:58.0189713Z 2025-09-09T14:48:58.0190000Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:48:58.0190395Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:48:58.0190668Z [0., 0., 0.], 2025-09-09T14:48:58.0190878Z [0., 0., 0.]]]) 2025-09-09T14:48:58.0191317Z PASSED 2025-09-09T14:48:58.0191992Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_transpose_bn PASSED 2025-09-09T14:48:58.0193024Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_transpose_bn_relu PASSED 2025-09-09T14:48:58.0193990Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_inplace_add_relu model pt2e: GraphModule( 2025-09-09T14:48:58.0194611Z (conv): Module() 2025-09-09T14:48:58.0194927Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:48:58.0195894Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0010]), zero_point=tensor([127], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:48:58.0197077Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2457]), max_val=tensor([-0.2457])) 2025-09-09T14:48:58.0197839Z ) 2025-09-09T14:48:58.0198120Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:48:58.0199240Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0089]), zero_point=tensor([41], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:48:58.0200346Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.498010516166687, max_val=0.7672448754310608) 2025-09-09T14:48:58.0201013Z ) 2025-09-09T14:48:58.0201295Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:48:58.0202221Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0022]), zero_point=tensor([46], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:48:58.0203324Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3805955946445465, max_val=0.17587313055992126) 2025-09-09T14:48:58.0203841Z ) 2025-09-09T14:48:58.0204115Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:48:58.0205067Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:48:58.0206134Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.3842603862285614) 2025-09-09T14:48:58.0206606Z ) 2025-09-09T14:48:58.0206779Z ) 2025-09-09T14:48:58.0206878Z 2025-09-09T14:48:58.0206883Z 2025-09-09T14:48:58.0206887Z 2025-09-09T14:48:58.0206973Z def forward(self, x): 2025-09-09T14:48:58.0207262Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:48:58.0207599Z conv_weight = self.conv.weight 2025-09-09T14:48:58.0208065Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T14:48:58.0208539Z conv_bias = self.conv.bias 2025-09-09T14:48:58.0208911Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:48:58.0209699Z conv1d = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, conv_bias); activation_post_process_1 = conv_bias = None 2025-09-09T14:48:58.0210514Z activation_post_process_2 = self.activation_post_process_2(conv1d); conv1d = None 2025-09-09T14:48:58.0211320Z add_ = torch.ops.aten.add_.Tensor(activation_post_process_2, activation_post_process_0); activation_post_process_2 = activation_post_process_0 = None 2025-09-09T14:48:58.0212042Z relu_ = torch.ops.aten.relu_.default(add_); add_ = None 2025-09-09T14:48:58.0212520Z activation_post_process_3 = self.activation_post_process_3(relu_); relu_ = None 2025-09-09T14:48:58.0213089Z return pytree.tree_unflatten((activation_post_process_3,), self._out_spec) 2025-09-09T14:48:58.0213487Z 2025-09-09T14:48:58.0213768Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:48:58.0214139Z model fx: GraphModule( 2025-09-09T14:48:58.0214465Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:48:59.5025243Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0089]), zero_point=tensor([41], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:48:59.5026588Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.498010516166687, max_val=0.7672448754310608) 2025-09-09T14:48:59.5027120Z ) 2025-09-09T14:48:59.5027315Z (conv): Conv1d( 2025-09-09T14:48:59.5027539Z 1, 1, kernel_size=(1,), stride=(1,) 2025-09-09T14:48:59.5027891Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:48:59.5028836Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0010]), zero_point=tensor([127], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:48:59.5030045Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2457]), max_val=tensor([-0.2457])) 2025-09-09T14:48:59.5030641Z ) 2025-09-09T14:48:59.5030819Z ) 2025-09-09T14:48:59.5031280Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:48:59.5032243Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0022]), zero_point=tensor([46], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:48:59.5033477Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3805955946445465, max_val=0.17587313055992126) 2025-09-09T14:48:59.5034004Z ) 2025-09-09T14:48:59.5034211Z (relu): ReLU(inplace=True) 2025-09-09T14:48:59.5034567Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:48:59.5035526Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:48:59.5036778Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.3842603862285614) 2025-09-09T14:48:59.5037320Z ) 2025-09-09T14:48:59.5037497Z ) 2025-09-09T14:48:59.5037598Z 2025-09-09T14:48:59.5037602Z 2025-09-09T14:48:59.5037612Z 2025-09-09T14:48:59.5037707Z def forward(self, x): 2025-09-09T14:48:59.5038065Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:48:59.5038623Z conv = self.conv(activation_post_process_0) 2025-09-09T14:48:59.5039129Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:48:59.5039835Z add = activation_post_process_1 + activation_post_process_0; activation_post_process_1 = activation_post_process_0 = None 2025-09-09T14:48:59.5040404Z relu = self.relu(add); add = None 2025-09-09T14:48:59.5040827Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T14:48:59.5041263Z return activation_post_process_2 2025-09-09T14:48:59.5041527Z 2025-09-09T14:48:59.5041819Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:48:59.5042241Z diff: tensor([[[0., 0., 0.]]], grad_fn=) 2025-09-09T14:48:59.5042584Z converted model pt2e: GraphModule( 2025-09-09T14:48:59.5042862Z (conv): Module() 2025-09-09T14:48:59.5043073Z ) 2025-09-09T14:48:59.5043174Z 2025-09-09T14:48:59.5043178Z 2025-09-09T14:48:59.5043181Z 2025-09-09T14:48:59.5043269Z def forward(self, x): 2025-09-09T14:48:59.5043567Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:48:59.5043909Z _scale_0 = self._scale_0 2025-09-09T14:48:59.5044168Z _zero_point_0 = self._zero_point_0 2025-09-09T14:48:59.5044503Z quantize_per_channel_default = self._frozen_param0 2025-09-09T14:48:59.5045522Z dequantize_per_channel_default = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel_default, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel_default = _scale_0 = _zero_point_0 = None 2025-09-09T14:48:59.5046494Z conv_bias = self.conv.bias 2025-09-09T14:48:59.5047152Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.008883354254066944, 41, -128, 127, torch.int8); x = None 2025-09-09T14:48:59.5048295Z dequantize_per_tensor_default_4 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.008883354254066944, 41, -128, 127, torch.int8) 2025-09-09T14:48:59.5049631Z dequantize_per_tensor_default_3 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.008883354254066944, 41, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:48:59.5051111Z conv1d = torch.ops.aten.conv1d.default(dequantize_per_tensor_default_3, dequantize_per_channel_default, conv_bias); dequantize_per_tensor_default_3 = dequantize_per_channel_default = conv_bias = None 2025-09-09T14:48:59.5052478Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d, 0.0021822303533554077, 46, -128, 127, torch.int8); conv1d = None 2025-09-09T14:48:59.5053784Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.0021822303533554077, 46, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:48:59.5055190Z add_ = torch.ops.aten.add_.Tensor(dequantize_per_tensor_default_1, dequantize_per_tensor_default_4); dequantize_per_tensor_default_1 = dequantize_per_tensor_default_4 = None 2025-09-09T14:48:59.5056051Z relu_ = torch.ops.aten.relu_.default(add_); add_ = None 2025-09-09T14:48:59.5056827Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu_, 0.0015069034416228533, -128, -128, 127, torch.int8); relu_ = None 2025-09-09T14:48:59.5058141Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.0015069034416228533, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:48:59.5059176Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:48:59.5059601Z 2025-09-09T14:48:59.5059885Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:48:59.5060273Z onverted model fx: GraphModule( 2025-09-09T14:48:59.5060654Z (conv): QuantizedConv1d(Reference)(1, 1, kernel_size=(1,), stride=(1,)) 2025-09-09T14:48:59.5061105Z (relu): ReLU(inplace=True) 2025-09-09T14:48:59.5061339Z ) 2025-09-09T14:48:59.5061449Z 2025-09-09T14:48:59.5061453Z 2025-09-09T14:48:59.5061457Z 2025-09-09T14:48:59.5061546Z def forward(self, x): 2025-09-09T14:48:59.5062178Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.008883354254066944, 41, -128, 127, torch.int8); x = None 2025-09-09T14:48:59.5063439Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.008883354254066944, 41, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:48:59.5064343Z conv = self.conv(dequantize_per_tensor_default) 2025-09-09T14:48:59.5065083Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.0021822303533554077, 46, -128, 127, torch.int8); conv = None 2025-09-09T14:48:59.5066369Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.0021822303533554077, 46, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:48:59.5067584Z add = dequantize_per_tensor_default_1 + dequantize_per_tensor_default; dequantize_per_tensor_default_1 = dequantize_per_tensor_default = None 2025-09-09T14:48:59.5068216Z relu = self.relu(add); add = None 2025-09-09T14:48:59.5068930Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.0015069034416228533, -128, -128, 127, torch.int8); relu = None 2025-09-09T14:48:59.5070230Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.0015069034416228533, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:48:59.5071176Z return dequantize_per_tensor_default_2 2025-09-09T14:48:59.5071455Z 2025-09-09T14:48:59.5071740Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:48:59.5072120Z diff: tensor([[[0., 0., 0.]]]) 2025-09-09T14:48:59.5072377Z model pt2e: GraphModule( 2025-09-09T14:48:59.5072615Z (conv): Module() 2025-09-09T14:48:59.5072926Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:48:59.5073886Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0010]), zero_point=tensor([127], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:48:59.5075132Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.24565386772155762, max_val=-0.24565386772155762) 2025-09-09T14:48:59.5075662Z ) 2025-09-09T14:48:59.5076054Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:48:59.5077000Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0089]), zero_point=tensor([41], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:48:59.5078100Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.498010516166687, max_val=0.7672448754310608) 2025-09-09T14:48:59.5078622Z ) 2025-09-09T14:48:59.5078904Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:48:59.5079843Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0022]), zero_point=tensor([46], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:48:59.5080956Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3805955946445465, max_val=0.17587313055992126) 2025-09-09T14:48:59.5081482Z ) 2025-09-09T14:48:59.5081767Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:48:59.5082710Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:49:37.3996778Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.3842603862285614) 2025-09-09T14:49:37.3997620Z ) 2025-09-09T14:49:37.3997837Z ) 2025-09-09T14:49:37.3997957Z 2025-09-09T14:49:37.3997963Z 2025-09-09T14:49:37.3997967Z 2025-09-09T14:49:37.3998073Z def forward(self, x): 2025-09-09T14:49:37.3998431Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:49:37.3998879Z conv_weight = self.conv.weight 2025-09-09T14:49:37.3999444Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T14:49:37.4000055Z conv_bias = self.conv.bias 2025-09-09T14:49:37.4000512Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:49:37.4001493Z conv1d = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, conv_bias); activation_post_process_1 = conv_bias = None 2025-09-09T14:49:37.4002496Z activation_post_process_2 = self.activation_post_process_2(conv1d); conv1d = None 2025-09-09T14:49:37.4003549Z add_ = torch.ops.aten.add_.Tensor(activation_post_process_2, activation_post_process_0); activation_post_process_2 = activation_post_process_0 = None 2025-09-09T14:49:37.4004441Z relu_ = torch.ops.aten.relu_.default(add_); add_ = None 2025-09-09T14:49:37.4005025Z activation_post_process_3 = self.activation_post_process_3(relu_); relu_ = None 2025-09-09T14:49:37.4005712Z return pytree.tree_unflatten((activation_post_process_3,), self._out_spec) 2025-09-09T14:49:37.4006216Z 2025-09-09T14:49:37.4006552Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:49:37.4007006Z model fx: GraphModule( 2025-09-09T14:49:37.4007392Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:49:37.4008584Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0089]), zero_point=tensor([41], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:49:37.4009970Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.498010516166687, max_val=0.7672448754310608) 2025-09-09T14:49:37.4010605Z ) 2025-09-09T14:49:37.4010831Z (conv): Conv1d( 2025-09-09T14:49:37.4011095Z 1, 1, kernel_size=(1,), stride=(1,) 2025-09-09T14:49:37.4011509Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:49:37.4012962Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0010]), zero_point=tensor([127], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:49:37.4014544Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.24565386772155762, max_val=-0.24565386772155762) 2025-09-09T14:49:37.4015201Z ) 2025-09-09T14:49:37.4015409Z ) 2025-09-09T14:49:37.4015745Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:49:37.4017005Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0022]), zero_point=tensor([46], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:49:37.4018388Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3805955946445465, max_val=0.17587313055992126) 2025-09-09T14:49:37.4019041Z ) 2025-09-09T14:49:37.4019279Z (relu): ReLU(inplace=True) 2025-09-09T14:49:37.4019695Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:49:37.4020863Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:49:37.4022200Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.3842603862285614) 2025-09-09T14:49:37.4022797Z ) 2025-09-09T14:49:37.4023040Z ) 2025-09-09T14:49:37.4023159Z 2025-09-09T14:49:37.4023164Z 2025-09-09T14:49:37.4023169Z 2025-09-09T14:49:37.4023282Z def forward(self, x): 2025-09-09T14:49:37.4023707Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:49:37.4024237Z conv = self.conv(activation_post_process_0) 2025-09-09T14:49:37.4024767Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:49:37.4025629Z add = activation_post_process_1 + activation_post_process_0; activation_post_process_1 = activation_post_process_0 = None 2025-09-09T14:49:37.4026346Z relu = self.relu(add); add = None 2025-09-09T14:49:37.4026842Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T14:49:37.4027368Z return activation_post_process_2 2025-09-09T14:49:37.4027682Z 2025-09-09T14:49:37.4028022Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:49:37.4028524Z diff: tensor([[[0., 0., 0.]]], grad_fn=) 2025-09-09T14:49:37.4028933Z converted model pt2e: GraphModule( 2025-09-09T14:49:37.4029250Z (conv): Module() 2025-09-09T14:49:37.4029494Z ) 2025-09-09T14:49:37.4029613Z 2025-09-09T14:49:37.4029618Z 2025-09-09T14:49:37.4029623Z 2025-09-09T14:49:37.4029736Z def forward(self, x): 2025-09-09T14:49:37.4030077Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:49:37.4030541Z quantize_per_tensor_default = self._frozen_param0 2025-09-09T14:49:37.4031671Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.0019342823652550578, 0, -127, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:49:37.4032758Z conv_bias = self.conv.bias 2025-09-09T14:49:37.4033625Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.008883354254066944, 41, -128, 127, torch.int8); x = None 2025-09-09T14:49:37.4035113Z dequantize_per_tensor_default_5 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.008883354254066944, 41, -128, 127, torch.int8) 2025-09-09T14:49:37.4036449Z dequantize_per_tensor_default_4 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.008883354254066944, 41, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:49:37.4037969Z conv1d = torch.ops.aten.conv1d.default(dequantize_per_tensor_default_4, dequantize_per_tensor_default, conv_bias); dequantize_per_tensor_default_4 = dequantize_per_tensor_default = conv_bias = None 2025-09-09T14:49:37.4039308Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d, 0.0021822303533554077, 46, -128, 127, torch.int8); conv1d = None 2025-09-09T14:49:37.4040613Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.0021822303533554077, 46, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:49:37.4041936Z add_ = torch.ops.aten.add_.Tensor(dequantize_per_tensor_default_2, dequantize_per_tensor_default_5); dequantize_per_tensor_default_2 = dequantize_per_tensor_default_5 = None 2025-09-09T14:49:37.4042727Z relu_ = torch.ops.aten.relu_.default(add_); add_ = None 2025-09-09T14:49:37.4043509Z quantize_per_tensor_default_3 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu_, 0.0015069034416228533, -128, -128, 127, torch.int8); relu_ = None 2025-09-09T14:49:37.4044805Z dequantize_per_tensor_default_3 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_3, 0.0015069034416228533, -128, -128, 127, torch.int8); quantize_per_tensor_default_3 = None 2025-09-09T14:49:37.4045855Z return pytree.tree_unflatten((dequantize_per_tensor_default_3,), self._out_spec) 2025-09-09T14:49:37.4046273Z 2025-09-09T14:49:37.4046555Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:49:37.4046936Z onverted model fx: GraphModule( 2025-09-09T14:49:37.4047311Z (conv): QuantizedConv1d(Reference)(1, 1, kernel_size=(1,), stride=(1,)) 2025-09-09T14:49:37.4047712Z (relu): ReLU(inplace=True) 2025-09-09T14:49:37.4047944Z ) 2025-09-09T14:49:37.4048054Z 2025-09-09T14:49:37.4048058Z 2025-09-09T14:49:37.4048062Z 2025-09-09T14:49:37.4048145Z def forward(self, x): 2025-09-09T14:49:37.4048775Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.008883354254066944, 41, -128, 127, torch.int8); x = None 2025-09-09T14:49:37.4050029Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.008883354254066944, 41, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:49:37.4050924Z conv = self.conv(dequantize_per_tensor_default) 2025-09-09T14:49:37.4051659Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.0021822303533554077, 46, -128, 127, torch.int8); conv = None 2025-09-09T14:49:37.4052998Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.0021822303533554077, 46, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:49:37.4062227Z add = dequantize_per_tensor_default_1 + dequantize_per_tensor_default; dequantize_per_tensor_default_1 = dequantize_per_tensor_default = None 2025-09-09T14:49:37.4062917Z relu = self.relu(add); add = None 2025-09-09T14:49:37.4063690Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.0015069034416228533, -128, -128, 127, torch.int8); relu = None 2025-09-09T14:49:37.4064977Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.0015069034416228533, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:49:37.4065875Z return dequantize_per_tensor_default_2 2025-09-09T14:49:37.4066153Z 2025-09-09T14:49:37.4066431Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:49:37.4066797Z diff: tensor([[[0., 0., 0.]]]) 2025-09-09T14:49:37.4067229Z PASSED 2025-09-09T14:49:37.4068072Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_per_channel_weight_custom_dtype PASSED 2025-09-09T14:49:37.4069155Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_preserve_source_fn_stack PASSED 2025-09-09T14:49:37.4070191Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_update_shared_qspec model pt2e: GraphModule( 2025-09-09T14:49:37.4070812Z (conv): Module() 2025-09-09T14:49:37.4071014Z (bn): Module() 2025-09-09T14:49:37.4071322Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:49:51.3901176Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:49:51.3902612Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:49:51.3903252Z ) 2025-09-09T14:49:51.3903619Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:49:51.3904843Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0025, 0.0024, 0.0026]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:49:51.3906471Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.3156, -0.1982, -0.3261]), max_val=tensor([0.2827, 0.2978, 0.2359])) 2025-09-09T14:49:51.3907279Z ) 2025-09-09T14:49:51.3907611Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:49:51.3908769Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0106]), zero_point=tensor([-5], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:49:51.3910137Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3043057918548584, max_val=1.399786114692688) 2025-09-09T14:49:51.3910774Z ) 2025-09-09T14:49:51.3911111Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:49:51.3912273Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0106]), zero_point=tensor([-5], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:49:51.3913636Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3043057918548584, max_val=1.399786114692688) 2025-09-09T14:49:51.3914265Z ) 2025-09-09T14:49:51.3914474Z ) 2025-09-09T14:49:51.3914594Z 2025-09-09T14:49:51.3914599Z 2025-09-09T14:49:51.3914604Z 2025-09-09T14:49:51.3914715Z def forward(self, x): 2025-09-09T14:49:51.3915058Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:49:51.3915476Z conv_weight = self.conv.weight 2025-09-09T14:49:51.3915808Z conv_bias = self.conv.bias 2025-09-09T14:49:51.3916127Z bn_weight = self.bn.weight 2025-09-09T14:49:51.3916433Z bn_bias = self.bn.bias 2025-09-09T14:49:51.3916756Z bn_running_mean = self.bn.running_mean 2025-09-09T14:49:51.3917134Z bn_running_var = self.bn.running_var 2025-09-09T14:49:51.3917590Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:49:51.3918131Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:49:51.3918847Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:49:51.3919497Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:49:51.3919967Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:49:51.3920467Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:49:51.3920994Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T14:49:51.3921611Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:49:51.3922561Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:49:51.3923327Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:49:51.3924701Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:49:51.3925789Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T14:49:51.3926438Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T14:49:51.3927141Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T14:49:51.3927863Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:49:51.3928958Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:49:51.3930116Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T14:49:51.3931039Z hardtanh = torch.ops.aten.hardtanh.default(activation_post_process_2, -1.0, 1.0); activation_post_process_2 = None 2025-09-09T14:49:51.3931924Z activation_post_process_3 = self.activation_post_process_3(hardtanh); hardtanh = None 2025-09-09T14:49:51.3932631Z return pytree.tree_unflatten((activation_post_process_3,), self._out_spec) 2025-09-09T14:49:51.3933110Z 2025-09-09T14:49:51.3933447Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:49:51.3933901Z model fx: GraphModule( 2025-09-09T14:49:51.3934288Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:49:51.3935468Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:49:51.3936968Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:49:51.3937654Z ) 2025-09-09T14:49:51.3937880Z (conv): ConvBn1d( 2025-09-09T14:49:51.3938150Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T14:49:51.3938648Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:49:51.3939213Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:49:51.3940415Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0025, 0.0024, 0.0026]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:49:51.3942044Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.3156, -0.1982, -0.3261]), max_val=tensor([0.2827, 0.2978, 0.2359])) 2025-09-09T14:49:51.3942853Z ) 2025-09-09T14:49:51.3943074Z ) 2025-09-09T14:49:51.3943403Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:49:51.3944574Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0106]), zero_point=tensor([-5], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:49:51.3945946Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3043057918548584, max_val=1.399786114692688) 2025-09-09T14:49:51.3946576Z ) 2025-09-09T14:49:51.3946858Z (hardtanh): Hardtanh(min_val=-1.0, max_val=1.0) 2025-09-09T14:49:51.3947385Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:49:51.3948648Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0106]), zero_point=tensor([-5], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:49:51.3950027Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3043057918548584, max_val=1.399786114692688) 2025-09-09T14:49:51.3950736Z ) 2025-09-09T14:49:51.3950948Z ) 2025-09-09T14:49:51.3951067Z 2025-09-09T14:49:51.3951072Z 2025-09-09T14:49:51.3951077Z 2025-09-09T14:49:51.3951182Z def forward(self, x): 2025-09-09T14:49:51.3951610Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:49:51.3952268Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:49:51.3952938Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:49:51.3953659Z hardtanh = self.hardtanh(activation_post_process_1); activation_post_process_1 = None 2025-09-09T14:49:51.3954409Z activation_post_process_2 = self.activation_post_process_2(hardtanh); hardtanh = None 2025-09-09T14:49:51.3954977Z return activation_post_process_2 2025-09-09T14:49:51.3955321Z 2025-09-09T14:49:51.3955692Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:49:51.3956183Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:49:51.3956469Z [0., 0., 0.], 2025-09-09T14:49:51.3956711Z [0., 0., 0.]]], grad_fn=) 2025-09-09T14:49:51.3957026Z converted model pt2e: GraphModule( 2025-09-09T14:49:51.3957332Z (conv): Module() 2025-09-09T14:49:51.3957537Z (bn): Module() 2025-09-09T14:49:51.3957733Z ) 2025-09-09T14:49:51.3957834Z 2025-09-09T14:49:51.3957838Z 2025-09-09T14:49:51.3957842Z 2025-09-09T14:49:51.3957929Z def forward(self, x): 2025-09-09T14:49:51.3958216Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:49:51.3958556Z conv_bias = self.conv.bias 2025-09-09T14:49:51.3958854Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:49:51.3959570Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:49:51.3960799Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:49:51.3961845Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:49:51.3962315Z _scale_0 = self._scale_0 2025-09-09T14:49:51.3962572Z _zero_point_0 = self._zero_point_0 2025-09-09T14:49:51.3962880Z quantize_per_channel = self._frozen_param0 2025-09-09T14:49:51.3963761Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T14:49:51.3965118Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T14:49:51.3966331Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_2, 0.010604282841086388, -5, -128, 127, torch.int8); conv1d_2 = None 2025-09-09T14:49:51.3967683Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.010604282841086388, -5, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:49:54.1832996Z hardtanh = torch.ops.aten.hardtanh.default(dequantize_per_tensor_default_1, -1.0, 1.0); dequantize_per_tensor_default_1 = None 2025-09-09T14:49:54.1834340Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(hardtanh, 0.010604282841086388, -5, -128, 127, torch.int8); hardtanh = None 2025-09-09T14:49:54.1836249Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.010604282841086388, -5, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:49:54.1837676Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:49:54.1838210Z 2025-09-09T14:49:54.1838593Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:49:54.1839057Z onverted model fx: GraphModule( 2025-09-09T14:49:54.1839521Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T14:49:54.1840050Z (hardtanh): Hardtanh(min_val=-1.0, max_val=1.0) 2025-09-09T14:49:54.1840404Z ) 2025-09-09T14:49:54.1840530Z 2025-09-09T14:49:54.1840535Z 2025-09-09T14:49:54.1840547Z 2025-09-09T14:49:54.1840656Z def forward(self, x): 2025-09-09T14:49:54.1841418Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:49:54.1842991Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:49:54.1844275Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:49:54.1845349Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.010604282841086388, -5, -128, 127, torch.int8); conv = None 2025-09-09T14:49:54.1846967Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.010604282841086388, -5, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:49:54.1848318Z hardtanh = self.hardtanh(dequantize_per_tensor_default_1); dequantize_per_tensor_default_1 = None 2025-09-09T14:49:54.1849531Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(hardtanh, 0.010604282841086388, -5, -128, 127, torch.int8); hardtanh = None 2025-09-09T14:49:54.1851180Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.010604282841086388, -5, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:49:54.1852299Z return dequantize_per_tensor_default_2 2025-09-09T14:49:54.1852632Z 2025-09-09T14:49:54.1852978Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:49:54.1853431Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:49:54.1853727Z [0., 0., 0.], 2025-09-09T14:49:54.1853983Z [0., 0., 0.]]]) 2025-09-09T14:49:54.1854267Z model pt2e: GraphModule( 2025-09-09T14:49:54.1854544Z (conv): Module() 2025-09-09T14:49:54.1854795Z (bn): Module() 2025-09-09T14:49:54.1855156Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:49:54.1856403Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:49:54.1857908Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:49:54.1858598Z ) 2025-09-09T14:49:54.1858901Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:49:54.1859844Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0026]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:49:54.1860955Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3261372148990631, max_val=0.297783762216568) 2025-09-09T14:49:54.1861469Z ) 2025-09-09T14:49:54.1861749Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:49:54.1862782Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0106]), zero_point=tensor([-4], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:49:54.1864187Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3146827220916748, max_val=1.399786114692688) 2025-09-09T14:49:54.1864790Z ) 2025-09-09T14:49:54.1865101Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:49:54.1866225Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0106]), zero_point=tensor([-4], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:49:54.1867559Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3146827220916748, max_val=1.399786114692688) 2025-09-09T14:49:54.1868157Z ) 2025-09-09T14:49:54.1868360Z ) 2025-09-09T14:49:54.1868486Z 2025-09-09T14:49:54.1868497Z 2025-09-09T14:49:54.1868501Z 2025-09-09T14:49:54.1868595Z def forward(self, x): 2025-09-09T14:49:54.1868905Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:49:54.1869295Z conv_weight = self.conv.weight 2025-09-09T14:49:54.1869596Z conv_bias = self.conv.bias 2025-09-09T14:49:54.1869884Z bn_weight = self.bn.weight 2025-09-09T14:49:54.1870154Z bn_bias = self.bn.bias 2025-09-09T14:49:54.1870439Z bn_running_mean = self.bn.running_mean 2025-09-09T14:49:54.1870776Z bn_running_var = self.bn.running_var 2025-09-09T14:49:54.1871154Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:49:54.1871661Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:49:54.1872350Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:49:54.1872971Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:49:54.1873422Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:49:54.1873892Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:49:54.1874398Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T14:49:54.1874980Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:49:54.1875643Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:49:54.1876365Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:49:54.1877547Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:49:54.1878633Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T14:49:54.1879287Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T14:49:54.1879963Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T14:49:54.1880630Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:49:54.1881678Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:49:54.1882809Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T14:49:54.1883693Z hardtanh = torch.ops.aten.hardtanh.default(activation_post_process_2, -1.0, 1.0); activation_post_process_2 = None 2025-09-09T14:49:54.1884551Z activation_post_process_3 = self.activation_post_process_3(hardtanh); hardtanh = None 2025-09-09T14:49:54.1885226Z return pytree.tree_unflatten((activation_post_process_3,), self._out_spec) 2025-09-09T14:49:54.1885670Z 2025-09-09T14:49:54.1886068Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:49:54.1886434Z model fx: GraphModule( 2025-09-09T14:49:54.1886841Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:49:54.1887784Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:49:54.1888943Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T14:49:54.1889458Z ) 2025-09-09T14:49:54.1889640Z (conv): ConvBn1d( 2025-09-09T14:49:54.1889871Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T14:49:54.1890283Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:49:54.1890765Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:49:54.1891687Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0026]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:49:54.1892812Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3261372148990631, max_val=0.297783762216568) 2025-09-09T14:49:54.1893335Z ) 2025-09-09T14:49:54.1893509Z ) 2025-09-09T14:49:54.1893790Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:49:54.1894728Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0106]), zero_point=tensor([-4], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:49:54.1895847Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3146827220916748, max_val=1.399786114692688) 2025-09-09T14:49:54.1896433Z ) 2025-09-09T14:49:54.1896662Z (hardtanh): Hardtanh(min_val=-1.0, max_val=1.0) 2025-09-09T14:49:54.1897069Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:49:54.1898367Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0106]), zero_point=tensor([-4], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:50:54.0964060Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3146827220916748, max_val=1.399786114692688) 2025-09-09T14:50:54.0966099Z ) 2025-09-09T14:50:54.0966381Z ) 2025-09-09T14:50:54.0966524Z 2025-09-09T14:50:54.0966530Z 2025-09-09T14:50:54.0966535Z 2025-09-09T14:50:54.0966660Z def forward(self, x): 2025-09-09T14:50:54.0967112Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:50:54.0967798Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:50:54.0968513Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:50:54.0969242Z hardtanh = self.hardtanh(activation_post_process_1); activation_post_process_1 = None 2025-09-09T14:50:54.0970023Z activation_post_process_2 = self.activation_post_process_2(hardtanh); hardtanh = None 2025-09-09T14:50:54.0970606Z return activation_post_process_2 2025-09-09T14:50:54.0970938Z 2025-09-09T14:50:54.0971285Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:50:54.0971750Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:50:54.0972042Z [0., 0., 0.], 2025-09-09T14:50:54.0972340Z [0., 0., 0.]]], grad_fn=) 2025-09-09T14:50:54.0972718Z converted model pt2e: GraphModule( 2025-09-09T14:50:54.0973080Z (conv): Module() 2025-09-09T14:50:54.0973357Z (bn): Module() 2025-09-09T14:50:54.0973603Z ) 2025-09-09T14:50:54.0973728Z 2025-09-09T14:50:54.0973733Z 2025-09-09T14:50:54.0973738Z 2025-09-09T14:50:54.0973858Z def forward(self, x): 2025-09-09T14:50:54.0974568Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:50:54.0975001Z conv_bias = self.conv.bias 2025-09-09T14:50:54.0975620Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:50:54.0976623Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:50:54.0978190Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:50:54.0979495Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:50:54.0980113Z quantize_per_tensor = self._frozen_param0 2025-09-09T14:50:54.0981117Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.002568009542301297, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T14:50:54.0982697Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T14:50:54.0984216Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_2, 0.010644976049661636, -4, -128, 127, torch.int8); conv1d_2 = None 2025-09-09T14:50:54.0985853Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.010644976049661636, -4, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:50:54.0987322Z hardtanh = torch.ops.aten.hardtanh.default(dequantize_per_tensor_default_2, -1.0, 1.0); dequantize_per_tensor_default_2 = None 2025-09-09T14:50:54.0988616Z quantize_per_tensor_default_3 = torch.ops.quantized_decomposed.quantize_per_tensor.default(hardtanh, 0.010644976049661636, -4, -128, 127, torch.int8); hardtanh = None 2025-09-09T14:50:54.0990267Z dequantize_per_tensor_default_3 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_3, 0.010644976049661636, -4, -128, 127, torch.int8); quantize_per_tensor_default_3 = None 2025-09-09T14:50:54.0991545Z return pytree.tree_unflatten((dequantize_per_tensor_default_3,), self._out_spec) 2025-09-09T14:50:54.0992071Z 2025-09-09T14:50:54.0992419Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:50:54.0992889Z onverted model fx: GraphModule( 2025-09-09T14:50:54.0993355Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T14:50:54.0993885Z (hardtanh): Hardtanh(min_val=-1.0, max_val=1.0) 2025-09-09T14:50:54.0994249Z ) 2025-09-09T14:50:54.0994373Z 2025-09-09T14:50:54.0994378Z 2025-09-09T14:50:54.0994383Z 2025-09-09T14:50:54.0994490Z def forward(self, x): 2025-09-09T14:50:54.0995266Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T14:50:54.0996820Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:50:54.0998351Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:50:54.0999425Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.010644976049661636, -4, -128, 127, torch.int8); conv = None 2025-09-09T14:50:54.1001134Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.010644976049661636, -4, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:50:54.1002372Z hardtanh = self.hardtanh(dequantize_per_tensor_default_1); dequantize_per_tensor_default_1 = None 2025-09-09T14:50:54.1003311Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(hardtanh, 0.010644976049661636, -4, -128, 127, torch.int8); hardtanh = None 2025-09-09T14:50:54.1004729Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.010644976049661636, -4, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:50:54.1005617Z return dequantize_per_tensor_default_2 2025-09-09T14:50:54.1005894Z 2025-09-09T14:50:54.1006186Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:50:54.1006563Z diff: tensor([[[0., 0., 0.], 2025-09-09T14:50:54.1006807Z [0., 0., 0.], 2025-09-09T14:50:54.1007051Z [0., 0., 0.]]]) 2025-09-09T14:50:54.1007507Z PASSED 2025-09-09T14:50:54.1008183Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_fold_bn_erases_bn_node PASSED 2025-09-09T14:50:54.1009228Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_bn_bias_derived_qspec PASSED 2025-09-09T14:50:54.1010205Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_bn_fusion model pt2e: GraphModule( 2025-09-09T14:50:54.1010819Z (conv): Module() 2025-09-09T14:50:54.1011029Z (bn): Module() 2025-09-09T14:50:54.1011346Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:50:54.1012292Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:50:54.1013406Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:50:54.1013915Z ) 2025-09-09T14:50:54.1014217Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:50:54.1015222Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:50:54.1016577Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1894, -0.1850, -0.1857]), max_val=tensor([0.1520, 0.1622, 0.1895])) 2025-09-09T14:50:54.1017247Z ) 2025-09-09T14:50:54.1017537Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:50:54.1018488Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0161]), zero_point=tensor([-4], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:50:54.1019599Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.9920659065246582, max_val=2.112386465072632) 2025-09-09T14:50:54.1020111Z ) 2025-09-09T14:50:54.1020294Z ) 2025-09-09T14:50:54.1020396Z 2025-09-09T14:50:54.1020400Z 2025-09-09T14:50:54.1020408Z 2025-09-09T14:50:54.1020503Z def forward(self, x): 2025-09-09T14:50:54.1020794Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:50:54.1021144Z conv_weight = self.conv.weight 2025-09-09T14:50:54.1021424Z conv_bias = self.conv.bias 2025-09-09T14:50:54.1021692Z bn_weight = self.bn.weight 2025-09-09T14:50:54.1021947Z bn_bias = self.bn.bias 2025-09-09T14:50:54.1022213Z bn_running_mean = self.bn.running_mean 2025-09-09T14:50:54.1022521Z bn_running_var = self.bn.running_var 2025-09-09T14:50:54.1022867Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:50:54.1023310Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:50:54.1023912Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:50:54.1024539Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:50:54.1024936Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:50:54.1025436Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:50:54.1025888Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T14:50:54.1026406Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:50:54.1026987Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:50:54.1027655Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:50:54.1028638Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:50:54.1029550Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T14:50:54.1030099Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T14:50:54.1030698Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T14:50:54.1031260Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:51:11.0590623Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:51:11.0591854Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T14:51:11.0592662Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:51:11.0593143Z 2025-09-09T14:51:11.0593491Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:51:11.0593936Z model fx: GraphModule( 2025-09-09T14:51:11.0594345Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:51:11.0595512Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:51:11.0596912Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:51:11.0597743Z ) 2025-09-09T14:51:11.0597971Z (conv): ConvBn2d( 2025-09-09T14:51:11.0598256Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T14:51:11.0598759Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:51:11.0599337Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:51:11.0600544Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:51:11.0602189Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1894, -0.1850, -0.1857]), max_val=tensor([0.1520, 0.1622, 0.1895])) 2025-09-09T14:51:11.0603066Z ) 2025-09-09T14:51:11.0603276Z ) 2025-09-09T14:51:11.0603615Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:51:11.0604790Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0161]), zero_point=tensor([-4], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:51:11.0606180Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.9920659065246582, max_val=2.112386465072632) 2025-09-09T14:51:11.0606827Z ) 2025-09-09T14:51:11.0607035Z ) 2025-09-09T14:51:11.0607153Z 2025-09-09T14:51:11.0607158Z 2025-09-09T14:51:11.0607418Z 2025-09-09T14:51:11.0607531Z def forward(self, x): 2025-09-09T14:51:11.0607960Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:51:11.0608830Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:51:11.0609516Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:51:11.0610050Z return activation_post_process_1 2025-09-09T14:51:11.0610373Z 2025-09-09T14:51:11.0610709Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:51:11.0611165Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:51:11.0611453Z [0., 0., 0.], 2025-09-09T14:51:11.0611718Z [0., 0., 0.]], 2025-09-09T14:51:11.0611888Z 2025-09-09T14:51:11.0611983Z [[0., 0., 0.], 2025-09-09T14:51:11.0612258Z [0., 0., 0.], 2025-09-09T14:51:11.0612558Z [0., 0., 0.]], 2025-09-09T14:51:11.0620917Z 2025-09-09T14:51:11.0621043Z [[0., 0., 0.], 2025-09-09T14:51:11.0621329Z [0., 0., 0.], 2025-09-09T14:51:11.0621637Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T14:51:11.0622033Z converted model pt2e: GraphModule( 2025-09-09T14:51:11.0622392Z (conv): Module() 2025-09-09T14:51:11.0622674Z (bn): Module() 2025-09-09T14:51:11.0622916Z ) 2025-09-09T14:51:11.0623038Z 2025-09-09T14:51:11.0623043Z 2025-09-09T14:51:11.0623048Z 2025-09-09T14:51:11.0623159Z def forward(self, x): 2025-09-09T14:51:11.0623501Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:51:11.0623922Z conv_bias = self.conv.bias 2025-09-09T14:51:11.0624725Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:51:11.0626301Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:51:11.0627382Z _scale_0 = self._scale_0 2025-09-09T14:51:11.0627700Z _zero_point_0 = self._zero_point_0 2025-09-09T14:51:11.0628081Z quantize_per_channel = self._frozen_param0 2025-09-09T14:51:11.0629199Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T14:51:11.0630916Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T14:51:11.0632444Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_2, 0.01609589159488678, -4, -128, 127, torch.int8); conv2d_2 = None 2025-09-09T14:51:11.0634218Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.01609589159488678, -4, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:51:11.0635369Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T14:51:11.0635802Z 2025-09-09T14:51:11.0636087Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:51:11.0636476Z onverted model fx: GraphModule( 2025-09-09T14:51:11.0636868Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T14:51:11.0637256Z ) 2025-09-09T14:51:11.0637359Z 2025-09-09T14:51:11.0637363Z 2025-09-09T14:51:11.0637367Z 2025-09-09T14:51:11.0637456Z def forward(self, x): 2025-09-09T14:51:11.0638082Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:51:11.0639446Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:51:11.0640463Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:51:11.0641407Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.01609589159488678, -4, -128, 127, torch.int8); conv = None 2025-09-09T14:51:11.0642679Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.01609589159488678, -4, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:51:11.0643561Z return dequantize_per_tensor_default_1 2025-09-09T14:51:11.0643842Z 2025-09-09T14:51:11.0644123Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:51:11.0644503Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:51:11.0644741Z [0., 0., 0.], 2025-09-09T14:51:11.0644963Z [0., 0., 0.]], 2025-09-09T14:51:11.0645113Z 2025-09-09T14:51:11.0645191Z [[0., 0., 0.], 2025-09-09T14:51:11.0645407Z [0., 0., 0.], 2025-09-09T14:51:11.0645633Z [0., 0., 0.]], 2025-09-09T14:51:11.0645774Z 2025-09-09T14:51:11.0645852Z [[0., 0., 0.], 2025-09-09T14:51:11.0646067Z [0., 0., 0.], 2025-09-09T14:51:11.0646279Z [0., 0., 0.]]]]) 2025-09-09T14:51:11.0646523Z model pt2e: GraphModule( 2025-09-09T14:51:11.0646756Z (conv): Module() 2025-09-09T14:51:11.0646968Z (bn): Module() 2025-09-09T14:51:11.0647272Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:51:11.0648231Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:51:11.0649341Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:51:11.0649849Z ) 2025-09-09T14:51:11.0650134Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:51:11.0651084Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:51:11.0652210Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.18937721848487854, max_val=0.18946029245853424) 2025-09-09T14:51:11.0652788Z ) 2025-09-09T14:51:11.0653070Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:51:11.0654003Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0161]), zero_point=tensor([-4], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:51:11.0655101Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.9922423362731934, max_val=2.1162424087524414) 2025-09-09T14:51:11.0655626Z ) 2025-09-09T14:51:11.0655800Z ) 2025-09-09T14:51:11.0655950Z 2025-09-09T14:51:11.0655960Z 2025-09-09T14:51:11.0655964Z 2025-09-09T14:51:11.0656051Z def forward(self, x): 2025-09-09T14:51:11.0656345Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:51:11.0656684Z conv_weight = self.conv.weight 2025-09-09T14:51:11.0656967Z conv_bias = self.conv.bias 2025-09-09T14:51:11.0657222Z bn_weight = self.bn.weight 2025-09-09T14:51:11.0657477Z bn_bias = self.bn.bias 2025-09-09T14:51:11.0657744Z bn_running_mean = self.bn.running_mean 2025-09-09T14:51:11.0658049Z bn_running_var = self.bn.running_var 2025-09-09T14:51:11.0658391Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:51:11.0658834Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:51:11.0659546Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:51:11.0660078Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:51:11.0660479Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:51:11.0660975Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:51:11.0661427Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T14:51:11.0661942Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:51:11.0662505Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:51:11.0663174Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:51:11.0664150Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:51:11.0665055Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T14:51:36.6852598Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T14:51:36.6853255Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T14:51:36.6853832Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:51:36.6854705Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:51:36.6855640Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T14:51:36.6856320Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:51:36.6856717Z 2025-09-09T14:51:36.6857008Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:51:36.6857389Z model fx: GraphModule( 2025-09-09T14:51:36.6857727Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:51:36.6858675Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:51:36.6860087Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:51:36.6860692Z ) 2025-09-09T14:51:36.6860884Z (conv): ConvBn2d( 2025-09-09T14:51:36.6861135Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T14:51:36.6861600Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:51:36.6862143Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:51:36.6863268Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:51:36.6864677Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.18937721848487854, max_val=0.18946029245853424) 2025-09-09T14:51:36.6865304Z ) 2025-09-09T14:51:36.6865490Z ) 2025-09-09T14:51:36.6865799Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:51:36.6866948Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0161]), zero_point=tensor([-4], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:51:36.6868298Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.9922423362731934, max_val=2.1162424087524414) 2025-09-09T14:51:36.6868915Z ) 2025-09-09T14:51:36.6869094Z ) 2025-09-09T14:51:36.6869208Z 2025-09-09T14:51:36.6869213Z 2025-09-09T14:51:36.6869216Z 2025-09-09T14:51:36.6869595Z def forward(self, x): 2025-09-09T14:51:36.6869968Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:51:36.6870662Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:51:36.6871218Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:51:36.6871650Z return activation_post_process_1 2025-09-09T14:51:36.6871922Z 2025-09-09T14:51:36.6872211Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:51:36.6872591Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:51:36.6872839Z [0., 0., 0.], 2025-09-09T14:51:36.6873052Z [0., 0., 0.]], 2025-09-09T14:51:36.6873198Z 2025-09-09T14:51:36.6873287Z [[0., 0., 0.], 2025-09-09T14:51:36.6873505Z [0., 0., 0.], 2025-09-09T14:51:36.6873723Z [0., 0., 0.]], 2025-09-09T14:51:36.6873864Z 2025-09-09T14:51:36.6873942Z [[0., 0., 0.], 2025-09-09T14:51:36.6874163Z [0., 0., 0.], 2025-09-09T14:51:36.6874401Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T14:51:36.6874728Z converted model pt2e: GraphModule( 2025-09-09T14:51:36.6875006Z (conv): Module() 2025-09-09T14:51:36.6875207Z (bn): Module() 2025-09-09T14:51:36.6875408Z ) 2025-09-09T14:51:36.6875509Z 2025-09-09T14:51:36.6875513Z 2025-09-09T14:51:36.6875517Z 2025-09-09T14:51:36.6875605Z def forward(self, x): 2025-09-09T14:51:36.6875901Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:51:36.6876238Z conv_bias = self.conv.bias 2025-09-09T14:51:36.6876886Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:51:36.6878145Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:51:36.6879043Z quantize_per_tensor = self._frozen_param0 2025-09-09T14:51:36.6879844Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.0014918133383616805, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T14:51:36.6881104Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T14:51:36.6882298Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_2, 0.01611170545220375, -4, -128, 127, torch.int8); conv2d_2 = None 2025-09-09T14:51:36.6883581Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.01611170545220375, -4, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:51:36.6884583Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:51:36.6885000Z 2025-09-09T14:51:36.6885288Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:51:36.6885670Z onverted model fx: GraphModule( 2025-09-09T14:51:36.6886057Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T14:51:36.6886438Z ) 2025-09-09T14:51:36.6886548Z 2025-09-09T14:51:36.6886552Z 2025-09-09T14:51:36.6886555Z 2025-09-09T14:51:36.6886643Z def forward(self, x): 2025-09-09T14:51:36.6887270Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:51:36.6888512Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:51:36.6889613Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:51:36.6890471Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.01611170545220375, -4, -128, 127, torch.int8); conv = None 2025-09-09T14:51:36.6891806Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.01611170545220375, -4, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:51:36.6892689Z return dequantize_per_tensor_default_1 2025-09-09T14:51:36.6892964Z 2025-09-09T14:51:36.6893252Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:51:36.6893621Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:51:36.6893867Z [0., 0., 0.], 2025-09-09T14:51:36.6894085Z [0., 0., 0.]], 2025-09-09T14:51:36.6894240Z 2025-09-09T14:51:36.6894321Z [[0., 0., 0.], 2025-09-09T14:51:36.6894537Z [0., 0., 0.], 2025-09-09T14:51:36.6894756Z [0., 0., 0.]], 2025-09-09T14:51:36.6894899Z 2025-09-09T14:51:36.6894989Z [[0., 0., 0.], 2025-09-09T14:51:36.6895198Z [0., 0., 0.], 2025-09-09T14:51:36.6895422Z [0., 0., 0.]]]]) 2025-09-09T14:51:36.6895893Z PASSED 2025-09-09T14:51:36.6896496Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_bn_fusion_cuda model pt2e: GraphModule( 2025-09-09T14:51:36.6897121Z (conv): Module() 2025-09-09T14:51:36.6897607Z (bn): Module() 2025-09-09T14:51:36.6897915Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:51:36.6899038Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0183], device='cuda:0'), zero_point=tensor([10], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:51:36.6900318Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:51:36.6900882Z ) 2025-09-09T14:51:36.6901169Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:51:36.6902346Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0015, 0.0015, 0.0014], device='cuda:0'), zero_point=tensor([0, 0, 0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:51:36.6903893Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1860, -0.1897, -0.1787], device='cuda:0'), max_val=tensor([0.1824, 0.1870, 0.1478], device='cuda:0')) 2025-09-09T14:51:36.6904628Z ) 2025-09-09T14:51:36.6904903Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:51:36.6906020Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0165], device='cuda:0'), zero_point=tensor([2], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:51:36.6907294Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1394810676574707, max_val=2.0564441680908203) 2025-09-09T14:51:36.6907803Z ) 2025-09-09T14:51:36.6907980Z ) 2025-09-09T14:51:36.6908079Z 2025-09-09T14:51:36.6908083Z 2025-09-09T14:51:36.6908087Z 2025-09-09T14:51:36.6908180Z def forward(self, x): 2025-09-09T14:51:36.6908465Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:51:36.6908811Z conv_weight = self.conv.weight 2025-09-09T14:51:36.6909085Z conv_bias = self.conv.bias 2025-09-09T14:51:36.6909346Z bn_weight = self.bn.weight 2025-09-09T14:51:36.6909597Z bn_bias = self.bn.bias 2025-09-09T14:51:36.6909860Z bn_running_mean = self.bn.running_mean 2025-09-09T14:51:36.6910181Z bn_running_var = self.bn.running_var 2025-09-09T14:51:36.6910642Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:51:36.6911138Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:51:36.6911828Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:51:36.6912365Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:51:36.6912769Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:51:36.6913180Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:51:36.6913635Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T14:51:53.1688200Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:51:53.1689720Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:51:53.1691281Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:51:53.1693342Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:51:53.1694990Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T14:51:53.1696090Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T14:51:53.1696755Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T14:51:53.1697576Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:51:53.1698456Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:51:53.1699403Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T14:51:53.1699993Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:51:53.1700390Z 2025-09-09T14:51:53.1700675Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:51:53.1701051Z model fx: GraphModule( 2025-09-09T14:51:53.1701379Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:51:53.1702503Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0183], device='cuda:0'), zero_point=tensor([10], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:51:53.1703779Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:51:53.1704290Z ) 2025-09-09T14:51:53.1704482Z (conv): ConvBn2d( 2025-09-09T14:51:53.1704729Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T14:51:53.1705146Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:51:53.1705622Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:51:53.1706756Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0015, 0.0015, 0.0014], device='cuda:0'), zero_point=tensor([0, 0, 0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:51:53.1708309Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1860, -0.1897, -0.1787], device='cuda:0'), max_val=tensor([0.1824, 0.1870, 0.1478], device='cuda:0')) 2025-09-09T14:51:53.1709056Z ) 2025-09-09T14:51:53.1709233Z ) 2025-09-09T14:51:53.1709522Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:51:53.1710883Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0165], device='cuda:0'), zero_point=tensor([2], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:51:53.1712319Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1394810676574707, max_val=2.0564441680908203) 2025-09-09T14:51:53.1712839Z ) 2025-09-09T14:51:53.1713010Z ) 2025-09-09T14:51:53.1713110Z 2025-09-09T14:51:53.1713115Z 2025-09-09T14:51:53.1713118Z 2025-09-09T14:51:53.1713211Z def forward(self, x): 2025-09-09T14:51:53.1713569Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:51:53.1714110Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:51:53.1714655Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:51:53.1715093Z return activation_post_process_1 2025-09-09T14:51:53.1715364Z 2025-09-09T14:51:53.1715640Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:51:53.1716057Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:51:53.1716301Z [0., 0., 0.], 2025-09-09T14:51:53.1716519Z [0., 0., 0.]], 2025-09-09T14:51:53.1716660Z 2025-09-09T14:51:53.1716737Z [[0., 0., 0.], 2025-09-09T14:51:53.1716953Z [0., 0., 0.], 2025-09-09T14:51:53.1717158Z [0., 0., 0.]], 2025-09-09T14:51:53.1717305Z 2025-09-09T14:51:53.1717381Z [[0., 0., 0.], 2025-09-09T14:51:53.1717591Z [0., 0., 0.], 2025-09-09T14:51:53.1717854Z [0., 0., 0.]]]], device='cuda:0', grad_fn=) 2025-09-09T14:51:53.1718190Z converted model pt2e: GraphModule( 2025-09-09T14:51:53.1718454Z (conv): Module() 2025-09-09T14:51:53.1718665Z (bn): Module() 2025-09-09T14:51:53.1718854Z ) 2025-09-09T14:51:53.1718960Z 2025-09-09T14:51:53.1718965Z 2025-09-09T14:51:53.1718969Z 2025-09-09T14:51:53.1719058Z def forward(self, x): 2025-09-09T14:51:53.1719339Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:51:53.1719675Z conv_bias = self.conv.bias 2025-09-09T14:51:53.1720324Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:51:53.1721554Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:51:53.1722406Z _scale_0 = self._scale_0 2025-09-09T14:51:53.1722660Z _zero_point_0 = self._zero_point_0 2025-09-09T14:51:53.1722968Z quantize_per_channel = self._frozen_param0 2025-09-09T14:51:53.1723862Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T14:51:53.1725206Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T14:51:53.1726453Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_2, 0.016454609110951424, 2, -128, 127, torch.int8); conv2d_2 = None 2025-09-09T14:51:53.1727736Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.016454609110951424, 2, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:51:53.1728731Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T14:51:53.1729145Z 2025-09-09T14:51:53.1729424Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:51:53.1729804Z onverted model fx: GraphModule( 2025-09-09T14:51:53.1730273Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T14:51:53.1730656Z ) 2025-09-09T14:51:53.1730756Z 2025-09-09T14:51:53.1730839Z 2025-09-09T14:51:53.1730843Z 2025-09-09T14:51:53.1730939Z def forward(self, x): 2025-09-09T14:51:53.1731554Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:51:53.1732786Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:51:53.1733795Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:51:53.1734637Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.016454609110951424, 2, -128, 127, torch.int8); conv = None 2025-09-09T14:51:53.1735957Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.016454609110951424, 2, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:51:53.1736837Z return dequantize_per_tensor_default_1 2025-09-09T14:51:53.1737114Z 2025-09-09T14:51:53.1737398Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:51:53.1737764Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:51:53.1738005Z [0., 0., 0.], 2025-09-09T14:51:53.1738218Z [0., 0., 0.]], 2025-09-09T14:51:53.1738361Z 2025-09-09T14:51:53.1738448Z [[0., 0., 0.], 2025-09-09T14:51:53.1738655Z [0., 0., 0.], 2025-09-09T14:51:53.1738869Z [0., 0., 0.]], 2025-09-09T14:51:53.1739008Z 2025-09-09T14:51:53.1739086Z [[0., 0., 0.], 2025-09-09T14:51:53.1739299Z [0., 0., 0.], 2025-09-09T14:51:53.1739531Z [0., 0., 0.]]]], device='cuda:0') 2025-09-09T14:51:53.1739818Z model pt2e: GraphModule( 2025-09-09T14:51:53.1740051Z (conv): Module() 2025-09-09T14:51:53.1740252Z (bn): Module() 2025-09-09T14:51:53.1740560Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:51:53.1741679Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0183], device='cuda:0'), zero_point=tensor([10], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:51:53.1742946Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:51:53.1743456Z ) 2025-09-09T14:51:53.1743739Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:51:53.1744863Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0015], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:51:53.1746195Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.18965020775794983, max_val=0.1870359182357788) 2025-09-09T14:51:53.1746724Z ) 2025-09-09T14:51:53.1747008Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:51:53.1748105Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0164], device='cuda:0'), zero_point=tensor([2], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:51:53.1749386Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.137371778488159, max_val=2.0522286891937256) 2025-09-09T14:51:53.1749891Z ) 2025-09-09T14:51:53.1750068Z ) 2025-09-09T14:51:53.1750167Z 2025-09-09T14:51:53.1750171Z 2025-09-09T14:51:53.1750175Z 2025-09-09T14:51:53.1750355Z def forward(self, x): 2025-09-09T14:52:18.4911145Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:52:18.4913604Z conv_weight = self.conv.weight 2025-09-09T14:52:18.4915927Z conv_bias = self.conv.bias 2025-09-09T14:52:18.4916255Z bn_weight = self.bn.weight 2025-09-09T14:52:18.4916559Z bn_bias = self.bn.bias 2025-09-09T14:52:18.4916877Z bn_running_mean = self.bn.running_mean 2025-09-09T14:52:18.4917246Z bn_running_var = self.bn.running_var 2025-09-09T14:52:18.4917658Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:52:18.4918195Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:52:18.4918932Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:52:18.4919584Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:52:18.4920057Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:52:18.4920572Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:52:18.4921108Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T14:52:18.4921740Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:52:18.4922429Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:52:18.4923189Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:52:18.4924455Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:52:18.4925555Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T14:52:18.4926218Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T14:52:18.4926936Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T14:52:18.4927625Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:52:18.4928700Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:52:18.4929843Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T14:52:18.4930580Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:52:18.4931054Z 2025-09-09T14:52:18.4931399Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:52:18.4931851Z model fx: GraphModule( 2025-09-09T14:52:18.4932236Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:52:18.4933634Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0183], device='cuda:0'), zero_point=tensor([10], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:52:18.4935287Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:52:18.4935996Z ) 2025-09-09T14:52:18.4936228Z (conv): ConvBn2d( 2025-09-09T14:52:18.4936508Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T14:52:18.4937017Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:52:18.4937585Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:52:18.4939111Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0015], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:52:18.4940742Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.18965020775794983, max_val=0.1870359182357788) 2025-09-09T14:52:18.4941473Z ) 2025-09-09T14:52:18.4941691Z ) 2025-09-09T14:52:18.4942025Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:52:18.4943409Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0164], device='cuda:0'), zero_point=tensor([2], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:52:18.4944993Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.137371778488159, max_val=2.0522286891937256) 2025-09-09T14:52:18.4945623Z ) 2025-09-09T14:52:18.4945834Z ) 2025-09-09T14:52:18.4945952Z 2025-09-09T14:52:18.4945957Z 2025-09-09T14:52:18.4945962Z 2025-09-09T14:52:18.4946074Z def forward(self, x): 2025-09-09T14:52:18.4946503Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:52:18.4947154Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:52:18.4947830Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:52:18.4948360Z return activation_post_process_1 2025-09-09T14:52:18.4948673Z 2025-09-09T14:52:18.4949010Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:52:18.4949456Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:52:18.4949748Z [0., 0., 0.], 2025-09-09T14:52:18.4950001Z [0., 0., 0.]], 2025-09-09T14:52:18.4950182Z 2025-09-09T14:52:18.4950282Z [[0., 0., 0.], 2025-09-09T14:52:18.4950537Z [0., 0., 0.], 2025-09-09T14:52:18.4950789Z [0., 0., 0.]], 2025-09-09T14:52:18.4950957Z 2025-09-09T14:52:18.4951056Z [[0., 0., 0.], 2025-09-09T14:52:18.4951317Z [0., 0., 0.], 2025-09-09T14:52:18.4951651Z [0., 0., 0.]]]], device='cuda:0', grad_fn=) 2025-09-09T14:52:18.4952060Z converted model pt2e: GraphModule( 2025-09-09T14:52:18.4960048Z (conv): Module() 2025-09-09T14:52:18.4960302Z (bn): Module() 2025-09-09T14:52:18.4960500Z ) 2025-09-09T14:52:18.4960603Z 2025-09-09T14:52:18.4960615Z 2025-09-09T14:52:18.4960619Z 2025-09-09T14:52:18.4960711Z def forward(self, x): 2025-09-09T14:52:18.4961006Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:52:18.4961355Z conv_bias = self.conv.bias 2025-09-09T14:52:18.4962013Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:52:18.4963267Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:52:18.4964235Z quantize_per_tensor = self._frozen_param0 2025-09-09T14:52:18.4965049Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.0014933086931705475, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T14:52:18.4966347Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T14:52:18.4967577Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_2, 0.016429806128144264, 2, -128, 127, torch.int8); conv2d_2 = None 2025-09-09T14:52:18.4968881Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.016429806128144264, 2, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:52:18.4970051Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:52:18.4970472Z 2025-09-09T14:52:18.4970765Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:52:18.4971228Z onverted model fx: GraphModule( 2025-09-09T14:52:18.4971615Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T14:52:18.4972006Z ) 2025-09-09T14:52:18.4972106Z 2025-09-09T14:52:18.4972111Z 2025-09-09T14:52:18.4972115Z 2025-09-09T14:52:18.4972203Z def forward(self, x): 2025-09-09T14:52:18.4972844Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:52:18.4974116Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:52:18.4975158Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:52:18.4976102Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.016429806128144264, 2, -128, 127, torch.int8); conv = None 2025-09-09T14:52:18.4977398Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.016429806128144264, 2, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:52:18.4978312Z return dequantize_per_tensor_default_1 2025-09-09T14:52:18.4978594Z 2025-09-09T14:52:18.4978873Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:52:18.4979253Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:52:18.4979492Z [0., 0., 0.], 2025-09-09T14:52:18.4979710Z [0., 0., 0.]], 2025-09-09T14:52:18.4979854Z 2025-09-09T14:52:18.4979932Z [[0., 0., 0.], 2025-09-09T14:52:18.4980145Z [0., 0., 0.], 2025-09-09T14:52:18.4980358Z [0., 0., 0.]], 2025-09-09T14:52:18.4980503Z 2025-09-09T14:52:18.4980580Z [[0., 0., 0.], 2025-09-09T14:52:18.4980794Z [0., 0., 0.], 2025-09-09T14:52:18.4981023Z [0., 0., 0.]]]], device='cuda:0') 2025-09-09T14:52:18.4981493Z PASSED 2025-09-09T14:52:18.4982124Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_bn_fusion_literal_args model pt2e: GraphModule( 2025-09-09T14:52:18.4982788Z (conv): Module() 2025-09-09T14:52:18.4982992Z (bn): Module() 2025-09-09T14:52:18.4983300Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:52:18.4984311Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0165]), zero_point=tensor([-18], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:52:18.4985419Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.8076945543289185, max_val=2.388113498687744) 2025-09-09T14:52:18.4985936Z ) 2025-09-09T14:52:18.4986215Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:52:18.4987209Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:52:32.3143731Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1897, -0.1774, -0.1913]), max_val=tensor([0.1806, 0.1870, 0.1478])) 2025-09-09T14:52:32.3144610Z ) 2025-09-09T14:52:32.3144953Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:52:32.3146129Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0279]), zero_point=tensor([-3], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:52:32.3147750Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-3.481316328048706, max_val=3.622279405593872) 2025-09-09T14:52:32.3148395Z ) 2025-09-09T14:52:32.3148773Z ) 2025-09-09T14:52:32.3148895Z 2025-09-09T14:52:32.3148901Z 2025-09-09T14:52:32.3148905Z 2025-09-09T14:52:32.3149015Z def forward(self, x): 2025-09-09T14:52:32.3149368Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:52:32.3149785Z conv_weight = self.conv.weight 2025-09-09T14:52:32.3150119Z conv_bias = self.conv.bias 2025-09-09T14:52:32.3150438Z bn_weight = self.bn.weight 2025-09-09T14:52:32.3150742Z bn_bias = self.bn.bias 2025-09-09T14:52:32.3151064Z bn_running_mean = self.bn.running_mean 2025-09-09T14:52:32.3151429Z bn_running_var = self.bn.running_var 2025-09-09T14:52:32.3151835Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:52:32.3152368Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:52:32.3153092Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:52:32.3153741Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:52:32.3154219Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:52:32.3154727Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:52:32.3155263Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T14:52:32.3155889Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:52:32.3156575Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:52:32.3157334Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:52:32.3158613Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like, [2, 2], [4, 4]); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:52:32.3159725Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T14:52:32.3160387Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T14:52:32.3161110Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T14:52:32.3161787Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:52:32.3162858Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:52:32.3164000Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T14:52:32.3164744Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:52:32.3165224Z 2025-09-09T14:52:32.3165567Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:52:32.3166021Z model fx: GraphModule( 2025-09-09T14:52:32.3166410Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:52:32.3167583Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0165]), zero_point=tensor([-18], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:52:32.3168998Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.8076945543289185, max_val=2.388113498687744) 2025-09-09T14:52:32.3169632Z ) 2025-09-09T14:52:32.3169858Z (conv): ConvBn2d( 2025-09-09T14:52:32.3170170Z 3, 3, kernel_size=(3, 3), stride=(2, 2), padding=(4, 4) 2025-09-09T14:52:32.3170722Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:52:32.3171287Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:52:32.3172586Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:52:32.3174308Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1897, -0.1774, -0.1913]), max_val=tensor([0.1806, 0.1870, 0.1478])) 2025-09-09T14:52:32.3175111Z ) 2025-09-09T14:52:32.3175327Z ) 2025-09-09T14:52:32.3175658Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:52:32.3176936Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0279]), zero_point=tensor([-3], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:52:32.3178306Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-3.481316328048706, max_val=3.622279405593872) 2025-09-09T14:52:32.3178950Z ) 2025-09-09T14:52:32.3179160Z ) 2025-09-09T14:52:32.3179278Z 2025-09-09T14:52:32.3179283Z 2025-09-09T14:52:32.3179288Z 2025-09-09T14:52:32.3179403Z def forward(self, x): 2025-09-09T14:52:32.3179831Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:52:32.3180483Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:52:32.3181161Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:52:32.3181691Z return activation_post_process_1 2025-09-09T14:52:32.3182004Z 2025-09-09T14:52:32.3182347Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:52:32.3182798Z diff: tensor([[[[0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:32.3183137Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:32.3183436Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:32.3183749Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:32.3184050Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:32.3184352Z [0., 0., 0., 0., 0., 0.]], 2025-09-09T14:52:32.3184558Z 2025-09-09T14:52:32.3184663Z [[0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:32.3184964Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:32.3185265Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:32.3185567Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:32.3185867Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:32.3186159Z [0., 0., 0., 0., 0., 0.]], 2025-09-09T14:52:32.3186366Z 2025-09-09T14:52:32.3186465Z [[0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:32.3186759Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:32.3187056Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:32.3187348Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:32.3187645Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:32.3187990Z [0., 0., 0., 0., 0., 0.]]]], grad_fn=) 2025-09-09T14:52:32.3188400Z converted model pt2e: GraphModule( 2025-09-09T14:52:32.3188757Z (conv): Module() 2025-09-09T14:52:32.3189006Z (bn): Module() 2025-09-09T14:52:32.3189243Z ) 2025-09-09T14:52:32.3189363Z 2025-09-09T14:52:32.3189375Z 2025-09-09T14:52:32.3189380Z 2025-09-09T14:52:32.3189488Z def forward(self, x): 2025-09-09T14:52:32.3189839Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:52:32.3190250Z conv_bias = self.conv.bias 2025-09-09T14:52:32.3191054Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.016454149037599564, -18, -128, 127, torch.int8); x = None 2025-09-09T14:52:32.3192618Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.016454149037599564, -18, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:52:32.3193685Z _scale_0 = self._scale_0 2025-09-09T14:52:32.3194001Z _zero_point_0 = self._zero_point_0 2025-09-09T14:52:32.3194464Z quantize_per_channel = self._frozen_param0 2025-09-09T14:52:32.3195640Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T14:52:32.3197560Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias, [2, 2], [4, 4]); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T14:52:32.3198831Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_2, 0.027857238426804543, -3, -128, 127, torch.int8); conv2d_2 = None 2025-09-09T14:52:32.3200121Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.027857238426804543, -3, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:52:32.3201136Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T14:52:32.3201547Z 2025-09-09T14:52:32.3201838Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:52:32.3202223Z onverted model fx: GraphModule( 2025-09-09T14:52:32.3202656Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(2, 2), padding=(4, 4)) 2025-09-09T14:52:32.3203087Z ) 2025-09-09T14:52:32.3203190Z 2025-09-09T14:52:32.3203194Z 2025-09-09T14:52:32.3203198Z 2025-09-09T14:52:32.3203285Z def forward(self, x): 2025-09-09T14:52:32.3203915Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.016454149037599564, -18, -128, 127, torch.int8); x = None 2025-09-09T14:52:32.3205159Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.016454149037599564, -18, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:52:32.3206176Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:52:32.3207034Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.027857238426804543, -3, -128, 127, torch.int8); conv = None 2025-09-09T14:52:32.3208320Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.027857238426804543, -3, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:52:32.3209217Z return dequantize_per_tensor_default_1 2025-09-09T14:52:32.3209496Z 2025-09-09T14:52:32.3209773Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:52:32.3210149Z diff: tensor([[[[0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:32.3210422Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:32.3210675Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:32.3210919Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:32.3211164Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8614674Z [0., 0., 0., 0., 0., 0.]], 2025-09-09T14:52:48.8614938Z 2025-09-09T14:52:48.8615055Z [[0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8615429Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8615743Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8616135Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8616438Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8616771Z [0., 0., 0., 0., 0., 0.]], 2025-09-09T14:52:48.8616977Z 2025-09-09T14:52:48.8617085Z [[0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8617383Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8617686Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8617982Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8618283Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8618587Z [0., 0., 0., 0., 0., 0.]]]]) 2025-09-09T14:52:48.8618935Z model pt2e: GraphModule( 2025-09-09T14:52:48.8619217Z (conv): Module() 2025-09-09T14:52:48.8619807Z (bn): Module() 2025-09-09T14:52:48.8620181Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:52:48.8621567Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0165]), zero_point=tensor([-18], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:52:48.8622961Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.8076945543289185, max_val=2.388113498687744) 2025-09-09T14:52:48.8623618Z ) 2025-09-09T14:52:48.8623984Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:52:48.8625163Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:52:48.8626563Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.19127479195594788, max_val=0.1870359182357788) 2025-09-09T14:52:48.8627209Z ) 2025-09-09T14:52:48.8627541Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:52:48.8628709Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0278]), zero_point=tensor([-3], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:52:48.8630062Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-3.4796082973480225, max_val=3.620413064956665) 2025-09-09T14:52:48.8630698Z ) 2025-09-09T14:52:48.8630905Z ) 2025-09-09T14:52:48.8631024Z 2025-09-09T14:52:48.8631029Z 2025-09-09T14:52:48.8631034Z 2025-09-09T14:52:48.8631138Z def forward(self, x): 2025-09-09T14:52:48.8631485Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:52:48.8631892Z conv_weight = self.conv.weight 2025-09-09T14:52:48.8632231Z conv_bias = self.conv.bias 2025-09-09T14:52:48.8632547Z bn_weight = self.bn.weight 2025-09-09T14:52:48.8632855Z bn_bias = self.bn.bias 2025-09-09T14:52:48.8633171Z bn_running_mean = self.bn.running_mean 2025-09-09T14:52:48.8633541Z bn_running_var = self.bn.running_var 2025-09-09T14:52:48.8633999Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:52:48.8634533Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:52:48.8635256Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:52:48.8635901Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:52:48.8636382Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:52:48.8636878Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:52:48.8637412Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T14:52:48.8638043Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:52:48.8638726Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:52:48.8639491Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:52:48.8640708Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like, [2, 2], [4, 4]); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:52:48.8641832Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T14:52:48.8642494Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T14:52:48.8643209Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T14:52:48.8643945Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:52:48.8645101Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:52:48.8646346Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T14:52:48.8647086Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:52:48.8647563Z 2025-09-09T14:52:48.8647904Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:52:48.8648343Z model fx: GraphModule( 2025-09-09T14:52:48.8648738Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:52:48.8649910Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0165]), zero_point=tensor([-18], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:52:48.8651282Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.8076945543289185, max_val=2.388113498687744) 2025-09-09T14:52:48.8651921Z ) 2025-09-09T14:52:48.8652134Z (conv): ConvBn2d( 2025-09-09T14:52:48.8652463Z 3, 3, kernel_size=(3, 3), stride=(2, 2), padding=(4, 4) 2025-09-09T14:52:48.8653007Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:52:48.8653586Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:52:48.8654730Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:52:48.8656228Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.19127479195594788, max_val=0.1870359182357788) 2025-09-09T14:52:48.8656879Z ) 2025-09-09T14:52:48.8657090Z ) 2025-09-09T14:52:48.8657426Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:52:48.8658602Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0278]), zero_point=tensor([-3], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:52:48.8659978Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-3.4796082973480225, max_val=3.620413064956665) 2025-09-09T14:52:48.8660614Z ) 2025-09-09T14:52:48.8660827Z ) 2025-09-09T14:52:48.8660951Z 2025-09-09T14:52:48.8660956Z 2025-09-09T14:52:48.8660961Z 2025-09-09T14:52:48.8661069Z def forward(self, x): 2025-09-09T14:52:48.8661493Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:52:48.8662148Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:52:48.8662822Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:52:48.8663344Z return activation_post_process_1 2025-09-09T14:52:48.8663692Z 2025-09-09T14:52:48.8664048Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:52:48.8664503Z diff: tensor([[[[0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8664841Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8665149Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8665456Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8665759Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8666060Z [0., 0., 0., 0., 0., 0.]], 2025-09-09T14:52:48.8666264Z 2025-09-09T14:52:48.8666362Z [[0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8666664Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8666957Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8667254Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8667549Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8667858Z [0., 0., 0., 0., 0., 0.]], 2025-09-09T14:52:48.8668062Z 2025-09-09T14:52:48.8668166Z [[0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8668552Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8668861Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8669247Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8669552Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:52:48.8669894Z [0., 0., 0., 0., 0., 0.]]]], grad_fn=) 2025-09-09T14:52:48.8670318Z converted model pt2e: GraphModule( 2025-09-09T14:52:48.8670662Z (conv): Module() 2025-09-09T14:52:48.8670925Z (bn): Module() 2025-09-09T14:52:48.8671169Z ) 2025-09-09T14:52:48.8671298Z 2025-09-09T14:52:48.8671304Z 2025-09-09T14:52:48.8671309Z 2025-09-09T14:52:48.8671418Z def forward(self, x): 2025-09-09T14:52:48.8671739Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:52:48.8672073Z conv_bias = self.conv.bias 2025-09-09T14:52:48.8672736Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.016454149037599564, -18, -128, 127, torch.int8); x = None 2025-09-09T14:52:48.8674046Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.016454149037599564, -18, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:52:48.8674936Z quantize_per_tensor = self._frozen_param0 2025-09-09T14:52:48.8675728Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.0015061007579788566, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T14:52:48.8677008Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias, [2, 2], [4, 4]); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T14:52:48.8678226Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_2, 0.02784322015941143, -3, -128, 127, torch.int8); conv2d_2 = None 2025-09-09T14:52:48.8679521Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.02784322015941143, -3, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:52:48.8680527Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:52:48.8680941Z 2025-09-09T14:52:48.8681221Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:52:48.8681601Z onverted model fx: GraphModule( 2025-09-09T14:52:48.8682030Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(2, 2), padding=(4, 4)) 2025-09-09T14:52:48.8682449Z ) 2025-09-09T14:52:48.8682548Z 2025-09-09T14:52:48.8682552Z 2025-09-09T14:52:48.8682556Z 2025-09-09T14:52:48.8682652Z def forward(self, x): 2025-09-09T14:53:14.0552118Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.016454149037599564, -18, -128, 127, torch.int8); x = None 2025-09-09T14:53:14.0553416Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.016454149037599564, -18, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:53:14.0554657Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:53:14.0555635Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.02784322015941143, -3, -128, 127, torch.int8); conv = None 2025-09-09T14:53:14.0557317Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.02784322015941143, -3, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:53:14.0558323Z return dequantize_per_tensor_default_1 2025-09-09T14:53:14.0558610Z 2025-09-09T14:53:14.0558905Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:53:14.0559281Z diff: tensor([[[[0., 0., 0., 0., 0., 0.], 2025-09-09T14:53:14.0559808Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:53:14.0560104Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:53:14.0560350Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:53:14.0560753Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:53:14.0561000Z [0., 0., 0., 0., 0., 0.]], 2025-09-09T14:53:14.0561175Z 2025-09-09T14:53:14.0561259Z [[0., 0., 0., 0., 0., 0.], 2025-09-09T14:53:14.0561512Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:53:14.0561791Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:53:14.0562058Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:53:14.0562305Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:53:14.0562555Z [0., 0., 0., 0., 0., 0.]], 2025-09-09T14:53:14.0562723Z 2025-09-09T14:53:14.0562806Z [[0., 0., 0., 0., 0., 0.], 2025-09-09T14:53:14.0563056Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:53:14.0563299Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:53:14.0563555Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:53:14.0563807Z [0., 0., 0., 0., 0., 0.], 2025-09-09T14:53:14.0564056Z [0., 0., 0., 0., 0., 0.]]]]) 2025-09-09T14:53:14.0564524Z PASSED 2025-09-09T14:53:14.0565147Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_bn_fusion_no_conv_bias model pt2e: GraphModule( 2025-09-09T14:53:14.0565808Z (conv): Module() 2025-09-09T14:53:14.0566014Z (bn): Module() 2025-09-09T14:53:14.0566325Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:14.0567277Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0189]), zero_point=tensor([-17], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:53:14.0568405Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.0985729694366455, max_val=2.7226178646087646) 2025-09-09T14:53:14.0568924Z ) 2025-09-09T14:53:14.0569219Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:14.0570217Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0014]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:53:14.0571525Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1860, -0.1897, -0.1787]), max_val=tensor([0.1824, 0.1870, 0.1478])) 2025-09-09T14:53:14.0572217Z ) 2025-09-09T14:53:14.0572506Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:14.0573440Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0177]), zero_point=tensor([-7], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:53:14.0574562Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1459269523620605, max_val=2.376943349838257) 2025-09-09T14:53:14.0575073Z ) 2025-09-09T14:53:14.0575247Z ) 2025-09-09T14:53:14.0575355Z 2025-09-09T14:53:14.0575364Z 2025-09-09T14:53:14.0575368Z 2025-09-09T14:53:14.0575457Z def forward(self, x): 2025-09-09T14:53:14.0575742Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:53:14.0576201Z conv_weight = self.conv.weight 2025-09-09T14:53:14.0576483Z bn_weight = self.bn.weight 2025-09-09T14:53:14.0576734Z bn_bias = self.bn.bias 2025-09-09T14:53:14.0576997Z bn_running_mean = self.bn.running_mean 2025-09-09T14:53:14.0577300Z bn_running_var = self.bn.running_var 2025-09-09T14:53:14.0577637Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:53:14.0578075Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:53:14.0578671Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:53:14.0579291Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:53:14.0579697Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:53:14.0580191Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:53:14.0580634Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T14:53:14.0581147Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:53:14.0581713Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:53:14.0582559Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T14:53:14.0583398Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T14:53:14.0583936Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T14:53:14.0584835Z batch_norm_1 = torch.ops.aten.batch_norm.default(div_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); div_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:53:14.0585762Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T14:53:14.0586363Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:53:14.0586755Z 2025-09-09T14:53:14.0587037Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:53:14.0587411Z model fx: GraphModule( 2025-09-09T14:53:14.0587738Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:14.0588683Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0189]), zero_point=tensor([-17], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:53:14.0589793Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.0985729694366455, max_val=2.7226178646087646) 2025-09-09T14:53:14.0590316Z ) 2025-09-09T14:53:14.0590504Z (conv): ConvBn2d( 2025-09-09T14:53:14.0590761Z 3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False 2025-09-09T14:53:14.0591201Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:53:14.0591704Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:14.0592748Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0014]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:53:14.0594056Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1860, -0.1897, -0.1787]), max_val=tensor([0.1824, 0.1870, 0.1478])) 2025-09-09T14:53:14.0602439Z ) 2025-09-09T14:53:14.0602636Z ) 2025-09-09T14:53:14.0602936Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:14.0603916Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0177]), zero_point=tensor([-7], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:53:14.0605042Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1459269523620605, max_val=2.376943349838257) 2025-09-09T14:53:14.0605561Z ) 2025-09-09T14:53:14.0605740Z ) 2025-09-09T14:53:14.0605845Z 2025-09-09T14:53:14.0605849Z 2025-09-09T14:53:14.0605853Z 2025-09-09T14:53:14.0605946Z def forward(self, x): 2025-09-09T14:53:14.0606339Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:53:14.0606962Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:53:14.0607598Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:53:14.0608253Z return activation_post_process_1 2025-09-09T14:53:14.0608522Z 2025-09-09T14:53:14.0608812Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:53:14.0609301Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:53:14.0609544Z [0., 0., 0.], 2025-09-09T14:53:14.0609762Z [0., 0., 0.]], 2025-09-09T14:53:14.0609903Z 2025-09-09T14:53:14.0609981Z [[0., 0., 0.], 2025-09-09T14:53:14.0610196Z [0., 0., 0.], 2025-09-09T14:53:14.0610402Z [0., 0., 0.]], 2025-09-09T14:53:14.0610550Z 2025-09-09T14:53:14.0610627Z [[0., 0., 0.], 2025-09-09T14:53:14.0610833Z [0., 0., 0.], 2025-09-09T14:53:14.0611053Z [0., 0., 0.]]], 2025-09-09T14:53:14.0611200Z 2025-09-09T14:53:14.0611204Z 2025-09-09T14:53:14.0611287Z [[[0., 0., 0.], 2025-09-09T14:53:14.0611493Z [0., 0., 0.], 2025-09-09T14:53:14.0611706Z [0., 0., 0.]], 2025-09-09T14:53:14.0611847Z 2025-09-09T14:53:14.0611930Z [[0., 0., 0.], 2025-09-09T14:53:14.0612142Z [0., 0., 0.], 2025-09-09T14:53:14.0612349Z [0., 0., 0.]], 2025-09-09T14:53:14.0612495Z 2025-09-09T14:53:14.0612578Z [[0., 0., 0.], 2025-09-09T14:53:14.0612787Z [0., 0., 0.], 2025-09-09T14:53:14.0613004Z [0., 0., 0.]]], 2025-09-09T14:53:14.0613147Z 2025-09-09T14:53:14.0613151Z 2025-09-09T14:53:14.0613235Z [[[0., 0., 0.], 2025-09-09T14:53:14.0613439Z [0., 0., 0.], 2025-09-09T14:53:14.0613650Z [0., 0., 0.]], 2025-09-09T14:53:14.0613790Z 2025-09-09T14:53:14.0613867Z [[0., 0., 0.], 2025-09-09T14:53:14.0614076Z [0., 0., 0.], 2025-09-09T14:53:14.0614282Z [0., 0., 0.]], 2025-09-09T14:53:14.0614428Z 2025-09-09T14:53:14.0614504Z [[0., 0., 0.], 2025-09-09T14:53:14.0614708Z [0., 0., 0.], 2025-09-09T14:53:14.0614957Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T14:53:14.0615274Z converted model pt2e: GraphModule( 2025-09-09T14:53:14.0615540Z (conv): Module() 2025-09-09T14:53:14.0615752Z (bn): Module() 2025-09-09T14:53:14.0616019Z ) 2025-09-09T14:53:14.0616120Z 2025-09-09T14:53:14.0616129Z 2025-09-09T14:53:14.0616139Z 2025-09-09T14:53:14.0616226Z def forward(self, x): 2025-09-09T14:53:14.0616510Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:53:14.0617242Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01890663057565689, -17, -128, 127, torch.int8); x = None 2025-09-09T14:53:14.0618482Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01890663057565689, -17, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:53:14.0619332Z _scale_0 = self._scale_0 2025-09-09T14:53:14.0619593Z _zero_point_0 = self._zero_point_0 2025-09-09T14:53:14.0619898Z quantize_per_channel = self._frozen_param0 2025-09-09T14:53:16.7769211Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T14:53:16.7770368Z conv_weight_bias = self.conv.weight_bias 2025-09-09T14:53:16.7771430Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_weight_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_weight_bias = None 2025-09-09T14:53:16.7773071Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_2, 0.01773674599826336, -7, -128, 127, torch.int8); conv2d_2 = None 2025-09-09T14:53:16.7774701Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.01773674599826336, -7, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:53:16.7776488Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T14:53:16.7777006Z 2025-09-09T14:53:16.7777358Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:53:16.7777995Z onverted model fx: GraphModule( 2025-09-09T14:53:16.7778474Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T14:53:16.7778953Z ) 2025-09-09T14:53:16.7779077Z 2025-09-09T14:53:16.7779082Z 2025-09-09T14:53:16.7779087Z 2025-09-09T14:53:16.7779196Z def forward(self, x): 2025-09-09T14:53:16.7779973Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01890663057565689, -17, -128, 127, torch.int8); x = None 2025-09-09T14:53:16.7781525Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01890663057565689, -17, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:53:16.7782808Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:53:16.7783875Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.01773674599826336, -7, -128, 127, torch.int8); conv = None 2025-09-09T14:53:16.7785474Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.01773674599826336, -7, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:53:16.7786581Z return dequantize_per_tensor_default_1 2025-09-09T14:53:16.7786923Z 2025-09-09T14:53:16.7787263Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:53:16.7787723Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:53:16.7788013Z [0., 0., 0.], 2025-09-09T14:53:16.7788277Z [0., 0., 0.]], 2025-09-09T14:53:16.7788452Z 2025-09-09T14:53:16.7788550Z [[0., 0., 0.], 2025-09-09T14:53:16.7788813Z [0., 0., 0.], 2025-09-09T14:53:16.7789064Z [0., 0., 0.]], 2025-09-09T14:53:16.7789251Z 2025-09-09T14:53:16.7789349Z [[0., 0., 0.], 2025-09-09T14:53:16.7789612Z [0., 0., 0.], 2025-09-09T14:53:16.7789873Z [0., 0., 0.]]], 2025-09-09T14:53:16.7790047Z 2025-09-09T14:53:16.7790052Z 2025-09-09T14:53:16.7790153Z [[[0., 0., 0.], 2025-09-09T14:53:16.7790440Z [0., 0., 0.], 2025-09-09T14:53:16.7790697Z [0., 0., 0.]], 2025-09-09T14:53:16.7790870Z 2025-09-09T14:53:16.7790974Z [[0., 0., 0.], 2025-09-09T14:53:16.7791228Z [0., 0., 0.], 2025-09-09T14:53:16.7791487Z [0., 0., 0.]], 2025-09-09T14:53:16.7791656Z 2025-09-09T14:53:16.7791750Z [[0., 0., 0.], 2025-09-09T14:53:16.7792009Z [0., 0., 0.], 2025-09-09T14:53:16.7792262Z [0., 0., 0.]]], 2025-09-09T14:53:16.7792441Z 2025-09-09T14:53:16.7792446Z 2025-09-09T14:53:16.7792542Z [[[0., 0., 0.], 2025-09-09T14:53:16.7792797Z [0., 0., 0.], 2025-09-09T14:53:16.7793083Z [0., 0., 0.]], 2025-09-09T14:53:16.7793280Z 2025-09-09T14:53:16.7793382Z [[0., 0., 0.], 2025-09-09T14:53:16.7793632Z [0., 0., 0.], 2025-09-09T14:53:16.7793889Z [0., 0., 0.]], 2025-09-09T14:53:16.7794066Z 2025-09-09T14:53:16.7794161Z [[0., 0., 0.], 2025-09-09T14:53:16.7794421Z [0., 0., 0.], 2025-09-09T14:53:16.7794694Z [0., 0., 0.]]]]) 2025-09-09T14:53:16.7795010Z model pt2e: GraphModule( 2025-09-09T14:53:16.7795307Z (conv): Module() 2025-09-09T14:53:16.7795580Z (bn): Module() 2025-09-09T14:53:16.7795957Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:16.7796913Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0189]), zero_point=tensor([-17], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:53:16.7798277Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.0985729694366455, max_val=2.7226178646087646) 2025-09-09T14:53:16.7798939Z ) 2025-09-09T14:53:16.7799233Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:16.7800187Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:53:16.7801410Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.18965020775794983, max_val=0.1870359182357788) 2025-09-09T14:53:16.7801933Z ) 2025-09-09T14:53:16.7802213Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:16.7803195Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0177]), zero_point=tensor([-7], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:53:16.7804326Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1475415229797363, max_val=2.368046283721924) 2025-09-09T14:53:16.7804837Z ) 2025-09-09T14:53:16.7805020Z ) 2025-09-09T14:53:16.7805121Z 2025-09-09T14:53:16.7805131Z 2025-09-09T14:53:16.7805135Z 2025-09-09T14:53:16.7805222Z def forward(self, x): 2025-09-09T14:53:16.7805523Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:53:16.7805867Z conv_weight = self.conv.weight 2025-09-09T14:53:16.7806153Z bn_weight = self.bn.weight 2025-09-09T14:53:16.7806420Z bn_bias = self.bn.bias 2025-09-09T14:53:16.7806686Z bn_running_mean = self.bn.running_mean 2025-09-09T14:53:16.7806999Z bn_running_var = self.bn.running_var 2025-09-09T14:53:16.7807331Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:53:16.7807777Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:53:16.7808367Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:53:16.7808910Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:53:16.7809311Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:53:16.7809731Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:53:16.7810185Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T14:53:16.7810692Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:53:16.7811270Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:53:16.7812115Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T14:53:16.7812950Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T14:53:16.7813496Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T14:53:16.7814393Z batch_norm_1 = torch.ops.aten.batch_norm.default(div_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); div_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:53:16.7815334Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T14:53:16.7815991Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:53:16.7816384Z 2025-09-09T14:53:16.7816672Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:53:16.7817037Z model fx: GraphModule( 2025-09-09T14:53:16.7817377Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:16.7818319Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0189]), zero_point=tensor([-17], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:53:16.7819515Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.0985729694366455, max_val=2.7226178646087646) 2025-09-09T14:53:16.7820038Z ) 2025-09-09T14:53:16.7820226Z (conv): ConvBn2d( 2025-09-09T14:53:16.7820568Z 3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False 2025-09-09T14:53:16.7821001Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:53:16.7821472Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:16.7822391Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:53:16.7823557Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.18965020775794983, max_val=0.1870359182357788) 2025-09-09T14:53:16.7824079Z ) 2025-09-09T14:53:16.7824255Z ) 2025-09-09T14:53:16.7824547Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:16.7825499Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0177]), zero_point=tensor([-7], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:53:16.7826623Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1475415229797363, max_val=2.368046283721924) 2025-09-09T14:53:16.7827136Z ) 2025-09-09T14:53:16.7827311Z ) 2025-09-09T14:53:16.7827410Z 2025-09-09T14:53:16.7827414Z 2025-09-09T14:53:16.7827418Z 2025-09-09T14:53:16.7827511Z def forward(self, x): 2025-09-09T14:53:16.7827865Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:53:16.7828409Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:53:16.7828960Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:53:16.7829404Z return activation_post_process_1 2025-09-09T14:53:16.7829677Z 2025-09-09T14:53:16.7829957Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:53:16.7830344Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:53:16.7830597Z [0., 0., 0.], 2025-09-09T14:53:16.7830823Z [0., 0., 0.]], 2025-09-09T14:53:16.7830967Z 2025-09-09T14:53:16.7831050Z [[0., 0., 0.], 2025-09-09T14:53:16.7831271Z [0., 0., 0.], 2025-09-09T14:53:16.7831483Z [0., 0., 0.]], 2025-09-09T14:53:16.7831632Z 2025-09-09T14:53:16.7831714Z [[0., 0., 0.], 2025-09-09T14:53:16.7831931Z [0., 0., 0.], 2025-09-09T14:53:16.7832144Z [0., 0., 0.]]], 2025-09-09T14:53:16.7832290Z 2025-09-09T14:53:16.7832294Z 2025-09-09T14:53:16.7832383Z [[[0., 0., 0.], 2025-09-09T14:53:16.7832596Z [0., 0., 0.], 2025-09-09T14:53:16.7832818Z [0., 0., 0.]], 2025-09-09T14:53:16.7832971Z 2025-09-09T14:53:16.7833053Z [[0., 0., 0.], 2025-09-09T14:53:16.7833272Z [0., 0., 0.], 2025-09-09T14:53:16.7833489Z [0., 0., 0.]], 2025-09-09T14:53:16.7833634Z 2025-09-09T14:53:16.7833712Z [[0., 0., 0.], 2025-09-09T14:53:16.7833940Z [0., 0., 0.], 2025-09-09T14:53:16.7834153Z [0., 0., 0.]]], 2025-09-09T14:53:16.7834299Z 2025-09-09T14:53:16.7834303Z 2025-09-09T14:53:16.7834387Z [[[0., 0., 0.], 2025-09-09T14:53:16.7834593Z [0., 0., 0.], 2025-09-09T14:53:16.7834814Z [0., 0., 0.]], 2025-09-09T14:53:16.7834955Z 2025-09-09T14:53:33.3720124Z [[0., 0., 0.], 2025-09-09T14:53:33.3720517Z [0., 0., 0.], 2025-09-09T14:53:33.3720778Z [0., 0., 0.]], 2025-09-09T14:53:33.3720959Z 2025-09-09T14:53:33.3721066Z [[0., 0., 0.], 2025-09-09T14:53:33.3721324Z [0., 0., 0.], 2025-09-09T14:53:33.3721629Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T14:53:33.3722019Z converted model pt2e: GraphModule( 2025-09-09T14:53:33.3722353Z (conv): Module() 2025-09-09T14:53:33.3723064Z (bn): Module() 2025-09-09T14:53:33.3723321Z ) 2025-09-09T14:53:33.3723449Z 2025-09-09T14:53:33.3723455Z 2025-09-09T14:53:33.3723460Z 2025-09-09T14:53:33.3723792Z def forward(self, x): 2025-09-09T14:53:33.3724148Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:53:33.3725068Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01890663057565689, -17, -128, 127, torch.int8); x = None 2025-09-09T14:53:33.3726629Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01890663057565689, -17, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:53:33.3727747Z quantize_per_tensor = self._frozen_param0 2025-09-09T14:53:33.3728799Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.0014933086931705475, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T14:53:33.3729805Z conv_weight_bias = self.conv.weight_bias 2025-09-09T14:53:33.3730865Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_weight_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_weight_bias = None 2025-09-09T14:53:33.3732449Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_2, 0.01770818792283535, -7, -128, 127, torch.int8); conv2d_2 = None 2025-09-09T14:53:33.3734071Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.01770818792283535, -7, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:53:33.3735343Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:53:33.3735967Z 2025-09-09T14:53:33.3736325Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:53:33.3736794Z onverted model fx: GraphModule( 2025-09-09T14:53:33.3737266Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T14:53:33.3737743Z ) 2025-09-09T14:53:33.3737877Z 2025-09-09T14:53:33.3737882Z 2025-09-09T14:53:33.3737886Z 2025-09-09T14:53:33.3737996Z def forward(self, x): 2025-09-09T14:53:33.3738827Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01890663057565689, -17, -128, 127, torch.int8); x = None 2025-09-09T14:53:33.3740387Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01890663057565689, -17, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:53:33.3741651Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:53:33.3742716Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.01770818792283535, -7, -128, 127, torch.int8); conv = None 2025-09-09T14:53:33.3744304Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.01770818792283535, -7, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:53:33.3745414Z return dequantize_per_tensor_default_1 2025-09-09T14:53:33.3745756Z 2025-09-09T14:53:33.3746120Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:53:33.3746621Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:53:33.3746926Z [0., 0., 0.], 2025-09-09T14:53:33.3747209Z [0., 0., 0.]], 2025-09-09T14:53:33.3747388Z 2025-09-09T14:53:33.3747470Z [[0., 0., 0.], 2025-09-09T14:53:33.3747688Z [0., 0., 0.], 2025-09-09T14:53:33.3747898Z [0., 0., 0.]], 2025-09-09T14:53:33.3748044Z 2025-09-09T14:53:33.3748122Z [[0., 0., 0.], 2025-09-09T14:53:33.3748338Z [0., 0., 0.], 2025-09-09T14:53:33.3748547Z [0., 0., 0.]]], 2025-09-09T14:53:33.3748693Z 2025-09-09T14:53:33.3748797Z 2025-09-09T14:53:33.3748884Z [[[0., 0., 0.], 2025-09-09T14:53:33.3749099Z [0., 0., 0.], 2025-09-09T14:53:33.3749398Z [0., 0., 0.]], 2025-09-09T14:53:33.3749539Z 2025-09-09T14:53:33.3749619Z [[0., 0., 0.], 2025-09-09T14:53:33.3749836Z [0., 0., 0.], 2025-09-09T14:53:33.3750043Z [0., 0., 0.]], 2025-09-09T14:53:33.3750196Z 2025-09-09T14:53:33.3750274Z [[0., 0., 0.], 2025-09-09T14:53:33.3750491Z [0., 0., 0.], 2025-09-09T14:53:33.3750708Z [0., 0., 0.]]], 2025-09-09T14:53:33.3750855Z 2025-09-09T14:53:33.3750867Z 2025-09-09T14:53:33.3750945Z [[[0., 0., 0.], 2025-09-09T14:53:33.3751154Z [0., 0., 0.], 2025-09-09T14:53:33.3751373Z [0., 0., 0.]], 2025-09-09T14:53:33.3751518Z 2025-09-09T14:53:33.3751596Z [[0., 0., 0.], 2025-09-09T14:53:33.3751816Z [0., 0., 0.], 2025-09-09T14:53:33.3752031Z [0., 0., 0.]], 2025-09-09T14:53:33.3752175Z 2025-09-09T14:53:33.3752258Z [[0., 0., 0.], 2025-09-09T14:53:33.3752481Z [0., 0., 0.], 2025-09-09T14:53:33.3752693Z [0., 0., 0.]]]]) 2025-09-09T14:53:33.3752949Z model pt2e: GraphModule( 2025-09-09T14:53:33.3753181Z (conv1): Module() 2025-09-09T14:53:33.3753396Z (bn1): Module() 2025-09-09T14:53:33.3753601Z (conv2): Module() 2025-09-09T14:53:33.3753816Z (bn2): Module() 2025-09-09T14:53:33.3754122Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:33.3755074Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0189]), zero_point=tensor([-17], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:53:33.3756207Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.0985729694366455, max_val=2.7226178646087646) 2025-09-09T14:53:33.3756728Z ) 2025-09-09T14:53:33.3757022Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:33.3758024Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:53:33.3759325Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1921, -0.1899, -0.1895]), max_val=tensor([0.1769, 0.1726, 0.1697])) 2025-09-09T14:53:33.3759984Z ) 2025-09-09T14:53:33.3760274Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:33.3761264Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0014, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:53:33.3762555Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1913, -0.1469, -0.1921]), max_val=tensor([0.1740, 0.1746, 0.1810])) 2025-09-09T14:53:33.3763197Z ) 2025-09-09T14:53:33.3763489Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:33.3764424Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0174]), zero_point=tensor([-29], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:53:33.3765525Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.7323514223098755, max_val=2.7138354778289795) 2025-09-09T14:53:33.3766042Z ) 2025-09-09T14:53:33.3766320Z (activation_post_process_4): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:33.3767254Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0110]), zero_point=tensor([-1], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:53:33.3768487Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.4053945541381836, max_val=1.4082176685333252) 2025-09-09T14:53:33.3769005Z ) 2025-09-09T14:53:33.3769183Z ) 2025-09-09T14:53:33.3769360Z 2025-09-09T14:53:33.3769364Z 2025-09-09T14:53:33.3769368Z 2025-09-09T14:53:33.3769458Z def forward(self, x): 2025-09-09T14:53:33.3769750Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:53:33.3770092Z conv1_weight = self.conv1.weight 2025-09-09T14:53:33.3770385Z bn1_weight = self.bn1.weight 2025-09-09T14:53:33.3770649Z bn1_bias = self.bn1.bias 2025-09-09T14:53:33.3770913Z conv2_weight = self.conv2.weight 2025-09-09T14:53:33.3771199Z conv2_bias = self.conv2.bias 2025-09-09T14:53:33.3771462Z bn2_weight = self.bn2.weight 2025-09-09T14:53:33.3771722Z bn2_bias = self.bn2.bias 2025-09-09T14:53:33.3771998Z bn1_running_mean = self.bn1.running_mean 2025-09-09T14:53:33.3772314Z bn1_running_var = self.bn1.running_var 2025-09-09T14:53:33.3772659Z bn1_num_batches_tracked = self.bn1.num_batches_tracked 2025-09-09T14:53:33.3773023Z bn2_running_mean = self.bn2.running_mean 2025-09-09T14:53:33.3773330Z bn2_running_var = self.bn2.running_var 2025-09-09T14:53:33.3773679Z bn2_num_batches_tracked = self.bn2.num_batches_tracked 2025-09-09T14:53:33.3774129Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:53:33.3774725Z add_ = torch.ops.aten.add_.Tensor(bn1_num_batches_tracked, 1); bn1_num_batches_tracked = add_ = None 2025-09-09T14:53:33.3775399Z add__1 = torch.ops.aten.add_.Tensor(bn2_num_batches_tracked, 1); bn2_num_batches_tracked = add__1 = None 2025-09-09T14:53:33.3776059Z add = torch.ops.aten.add.Tensor(bn2_running_var, 1e-05) 2025-09-09T14:53:33.3776470Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:53:33.3776891Z div = torch.ops.aten.div.Tensor(bn2_weight, sqrt); sqrt = None 2025-09-09T14:53:33.3777351Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T14:53:33.3777881Z mul = torch.ops.aten.mul.Tensor(conv2_weight, reshape); conv2_weight = reshape = None 2025-09-09T14:53:33.3778459Z activation_post_process_3 = self.activation_post_process_3(mul); mul = None 2025-09-09T14:53:33.3779095Z zeros_like = torch.ops.aten.zeros_like.default(conv2_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:53:33.3779661Z add_2 = torch.ops.aten.add.Tensor(bn1_running_var, 1e-05) 2025-09-09T14:53:33.3780091Z sqrt_1 = torch.ops.aten.sqrt.default(add_2); add_2 = None 2025-09-09T14:53:33.3780554Z div_2 = torch.ops.aten.div.Tensor(bn1_weight, sqrt_1); sqrt_1 = None 2025-09-09T14:53:33.3781032Z reshape_3 = torch.ops.aten.reshape.default(div_2, [-1, 1, 1, 1]) 2025-09-09T14:53:33.3781580Z mul_1 = torch.ops.aten.mul.Tensor(conv1_weight, reshape_3); conv1_weight = reshape_3 = None 2025-09-09T14:53:33.3782180Z activation_post_process_1 = self.activation_post_process_1(mul_1); mul_1 = None 2025-09-09T14:53:33.3783057Z conv2d_3 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T14:53:33.3783924Z reshape_4 = torch.ops.aten.reshape.default(div_2, [1, -1, 1, 1]); div_2 = None 2025-09-09T14:53:33.3784486Z div_3 = torch.ops.aten.div.Tensor(conv2d_3, reshape_4); conv2d_3 = reshape_4 = None 2025-09-09T14:53:50.5733605Z batch_norm_3 = torch.ops.aten.batch_norm.default(div_3, bn1_weight, bn1_bias, bn1_running_mean, bn1_running_var, True, 0.1, 1e-05, True); div_3 = bn1_weight = bn1_bias = bn1_running_mean = bn1_running_var = None 2025-09-09T14:53:50.5734662Z activation_post_process_2 = self.activation_post_process_2(batch_norm_3); batch_norm_3 = None 2025-09-09T14:53:50.5735615Z conv2d_2 = torch.ops.aten.conv2d.default(activation_post_process_2, activation_post_process_3, zeros_like); activation_post_process_2 = activation_post_process_3 = zeros_like = None 2025-09-09T14:53:50.5737016Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T14:53:50.5737579Z div_1 = torch.ops.aten.div.Tensor(conv2d_2, reshape_1); conv2d_2 = reshape_1 = None 2025-09-09T14:53:50.5738388Z reshape_2 = torch.ops.aten.reshape.default(conv2_bias, [1, -1, 1, 1]); conv2_bias = None 2025-09-09T14:53:50.5738961Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:53:50.5739902Z batch_norm_2 = torch.ops.aten.batch_norm.default(add_1, bn2_weight, bn2_bias, bn2_running_mean, bn2_running_var, True, 0.1, 1e-05, True); add_1 = bn2_weight = bn2_bias = bn2_running_mean = bn2_running_var = None 2025-09-09T14:53:50.5741241Z activation_post_process_4 = self.activation_post_process_4(batch_norm_2); batch_norm_2 = None 2025-09-09T14:53:50.5742046Z return pytree.tree_unflatten((activation_post_process_4,), self._out_spec) 2025-09-09T14:53:50.5742537Z 2025-09-09T14:53:50.5742822Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:53:50.5743192Z model fx: GraphModule( 2025-09-09T14:53:50.5743568Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:50.5744507Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0189]), zero_point=tensor([-17], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:53:50.5745640Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.0985729694366455, max_val=2.7226178646087646) 2025-09-09T14:53:50.5746155Z ) 2025-09-09T14:53:50.5746337Z (conv1): ConvBn2d( 2025-09-09T14:53:50.5746602Z 3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False 2025-09-09T14:53:50.5747032Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:53:50.5747499Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:50.5748461Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0014, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:53:50.5749748Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1913, -0.1469, -0.1921]), max_val=tensor([0.1740, 0.1746, 0.1810])) 2025-09-09T14:53:50.5750392Z ) 2025-09-09T14:53:50.5750564Z ) 2025-09-09T14:53:50.5750845Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:50.5751788Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0174]), zero_point=tensor([-29], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:53:50.5752888Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.7323514223098755, max_val=2.7138354778289795) 2025-09-09T14:53:50.5753426Z ) 2025-09-09T14:53:50.5753631Z (conv2): ConvBn2d( 2025-09-09T14:53:50.5753874Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T14:53:50.5754276Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:53:50.5754747Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:50.5755715Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:53:50.5756993Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1921, -0.1899, -0.1895]), max_val=tensor([0.1769, 0.1726, 0.1697])) 2025-09-09T14:53:50.5757640Z ) 2025-09-09T14:53:50.5757809Z ) 2025-09-09T14:53:50.5758093Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:50.5759151Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0110]), zero_point=tensor([-1], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:53:50.5760325Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.4053945541381836, max_val=1.4082176685333252) 2025-09-09T14:53:50.5760834Z ) 2025-09-09T14:53:50.5761002Z ) 2025-09-09T14:53:50.5761108Z 2025-09-09T14:53:50.5761112Z 2025-09-09T14:53:50.5761116Z 2025-09-09T14:53:50.5761201Z def forward(self, x): 2025-09-09T14:53:50.5761551Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:53:50.5762100Z conv1 = self.conv1(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:53:50.5762658Z activation_post_process_1 = self.activation_post_process_1(conv1); conv1 = None 2025-09-09T14:53:50.5763203Z conv2 = self.conv2(activation_post_process_1); activation_post_process_1 = None 2025-09-09T14:53:50.5763760Z activation_post_process_2 = self.activation_post_process_2(conv2); conv2 = None 2025-09-09T14:53:50.5764188Z return activation_post_process_2 2025-09-09T14:53:50.5764448Z 2025-09-09T14:53:50.5764731Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:53:50.5765093Z diff: tensor([[[[0.]], 2025-09-09T14:53:50.5765234Z 2025-09-09T14:53:50.5765315Z [[0.]], 2025-09-09T14:53:50.5765440Z 2025-09-09T14:53:50.5765516Z [[0.]]], 2025-09-09T14:53:50.5765636Z 2025-09-09T14:53:50.5765645Z 2025-09-09T14:53:50.5765718Z [[[0.]], 2025-09-09T14:53:50.5765861Z 2025-09-09T14:53:50.5765935Z [[0.]], 2025-09-09T14:53:50.5766054Z 2025-09-09T14:53:50.5766128Z [[0.]]], 2025-09-09T14:53:50.5766254Z 2025-09-09T14:53:50.5766258Z 2025-09-09T14:53:50.5774448Z [[[0.]], 2025-09-09T14:53:50.5774614Z 2025-09-09T14:53:50.5774697Z [[0.]], 2025-09-09T14:53:50.5774838Z 2025-09-09T14:53:50.5774949Z [[0.]]]], grad_fn=) 2025-09-09T14:53:50.5775261Z converted model pt2e: GraphModule( 2025-09-09T14:53:50.5775532Z (conv1): Module() 2025-09-09T14:53:50.5775746Z (bn1): Module() 2025-09-09T14:53:50.5776067Z (conv2): Module() 2025-09-09T14:53:50.5776284Z (bn2): Module() 2025-09-09T14:53:50.5776480Z ) 2025-09-09T14:53:50.5776579Z 2025-09-09T14:53:50.5776583Z 2025-09-09T14:53:50.5776587Z 2025-09-09T14:53:50.5776672Z def forward(self, x): 2025-09-09T14:53:50.5776961Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:53:50.5777300Z conv2_bias = self.conv2.bias 2025-09-09T14:53:50.5777952Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01890663057565689, -17, -128, 127, torch.int8); x = None 2025-09-09T14:53:50.5779190Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01890663057565689, -17, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:53:50.5780037Z _scale_0 = self._scale_0 2025-09-09T14:53:50.5780304Z _zero_point_0 = self._zero_point_0 2025-09-09T14:53:50.5780581Z _scale_1 = self._scale_1 2025-09-09T14:53:50.5780839Z _zero_point_1 = self._zero_point_1 2025-09-09T14:53:50.5781137Z quantize_per_channel_1 = self._frozen_param0 2025-09-09T14:53:50.5782039Z dequantize_per_channel_1 = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel_1, _scale_1, _zero_point_1, 0, -127, 127, torch.int8); quantize_per_channel_1 = _scale_1 = _zero_point_1 = None 2025-09-09T14:53:50.5782953Z conv1_weight_bias = self.conv1.weight_bias 2025-09-09T14:53:50.5783860Z conv2d_5 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_channel_1, conv1_weight_bias); dequantize_per_tensor_default = dequantize_per_channel_1 = conv1_weight_bias = None 2025-09-09T14:53:50.5785255Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_5, 0.01743602752685547, -29, -128, 127, torch.int8); conv2d_5 = None 2025-09-09T14:53:50.5786539Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.01743602752685547, -29, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:53:50.5787504Z quantize_per_channel = self._frozen_param1 2025-09-09T14:53:50.5788386Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T14:53:50.5789739Z conv2d_4 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default_1, dequantize_per_channel, conv2_bias); dequantize_per_tensor_default_1 = dequantize_per_channel = conv2_bias = None 2025-09-09T14:53:50.5790974Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_4, 0.011033773422241211, -1, -128, 127, torch.int8); conv2d_4 = None 2025-09-09T14:53:50.5792265Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.011033773422241211, -1, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:53:50.5793268Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:53:50.5793710Z 2025-09-09T14:53:50.5794007Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:53:50.5794393Z onverted model fx: GraphModule( 2025-09-09T14:53:50.5794782Z (conv1): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T14:53:50.5795296Z (conv2): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T14:53:50.5795676Z ) 2025-09-09T14:53:50.5795775Z 2025-09-09T14:53:50.5795779Z 2025-09-09T14:53:50.5795783Z 2025-09-09T14:53:50.5795869Z def forward(self, x): 2025-09-09T14:53:50.5796492Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01890663057565689, -17, -128, 127, torch.int8); x = None 2025-09-09T14:53:50.5798139Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01890663057565689, -17, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:53:50.5799162Z conv1 = self.conv1(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:53:50.5800024Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1, 0.01743602752685547, -29, -128, 127, torch.int8); conv1 = None 2025-09-09T14:53:53.3646671Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.01743602752685547, -29, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:53:53.3648043Z conv2 = self.conv2(dequantize_per_tensor_default_1); dequantize_per_tensor_default_1 = None 2025-09-09T14:53:53.3649170Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2, 0.011033773422241211, -1, -128, 127, torch.int8); conv2 = None 2025-09-09T14:53:53.3650789Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.011033773422241211, -1, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:53:53.3651901Z return dequantize_per_tensor_default_2 2025-09-09T14:53:53.3652256Z 2025-09-09T14:53:53.3652606Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:53:53.3653066Z diff: tensor([[[[0.]], 2025-09-09T14:53:53.3653247Z 2025-09-09T14:53:53.3653345Z [[0.]], 2025-09-09T14:53:53.3653504Z 2025-09-09T14:53:53.3653602Z [[0.]]], 2025-09-09T14:53:53.3653753Z 2025-09-09T14:53:53.3653758Z 2025-09-09T14:53:53.3653854Z [[[0.]], 2025-09-09T14:53:53.3654010Z 2025-09-09T14:53:53.3654479Z [[0.]], 2025-09-09T14:53:53.3654635Z 2025-09-09T14:53:53.3654738Z [[0.]]], 2025-09-09T14:53:53.3654891Z 2025-09-09T14:53:53.3654896Z 2025-09-09T14:53:53.3655165Z [[[0.]], 2025-09-09T14:53:53.3655322Z 2025-09-09T14:53:53.3655417Z [[0.]], 2025-09-09T14:53:53.3655567Z 2025-09-09T14:53:53.3655668Z [[0.]]]]) 2025-09-09T14:53:53.3656038Z model pt2e: GraphModule( 2025-09-09T14:53:53.3656331Z (conv1): Module() 2025-09-09T14:53:53.3656589Z (bn1): Module() 2025-09-09T14:53:53.3656848Z (conv2): Module() 2025-09-09T14:53:53.3657099Z (bn2): Module() 2025-09-09T14:53:53.3657480Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:53.3658657Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0189]), zero_point=tensor([-17], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:53:53.3660130Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.0985729694366455, max_val=2.7226178646087646) 2025-09-09T14:53:53.3660845Z ) 2025-09-09T14:53:53.3661208Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:53.3662262Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:53:53.3663371Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.1921343356370926, max_val=0.1768510341644287) 2025-09-09T14:53:53.3663896Z ) 2025-09-09T14:53:53.3664187Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:53.3665184Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:53:53.3666315Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.19212442636489868, max_val=0.18097376823425293) 2025-09-09T14:53:53.3666836Z ) 2025-09-09T14:53:53.3667134Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:53.3668080Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0174]), zero_point=tensor([-29], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:53:53.3669189Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.7288155555725098, max_val=2.7138354778289795) 2025-09-09T14:53:53.3669716Z ) 2025-09-09T14:53:53.3670000Z (activation_post_process_4): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:53.3670957Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0110]), zero_point=tensor([-1], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:53:53.3672064Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.4025696516036987, max_val=1.4086220264434814) 2025-09-09T14:53:53.3672588Z ) 2025-09-09T14:53:53.3672772Z ) 2025-09-09T14:53:53.3672879Z 2025-09-09T14:53:53.3672883Z 2025-09-09T14:53:53.3672887Z 2025-09-09T14:53:53.3672977Z def forward(self, x): 2025-09-09T14:53:53.3673280Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:53:53.3673633Z conv1_weight = self.conv1.weight 2025-09-09T14:53:53.3673928Z bn1_weight = self.bn1.weight 2025-09-09T14:53:53.3674201Z bn1_bias = self.bn1.bias 2025-09-09T14:53:53.3674466Z conv2_weight = self.conv2.weight 2025-09-09T14:53:53.3674758Z conv2_bias = self.conv2.bias 2025-09-09T14:53:53.3675025Z bn2_weight = self.bn2.weight 2025-09-09T14:53:53.3675293Z bn2_bias = self.bn2.bias 2025-09-09T14:53:53.3675569Z bn1_running_mean = self.bn1.running_mean 2025-09-09T14:53:53.3676060Z bn1_running_var = self.bn1.running_var 2025-09-09T14:53:53.3676415Z bn1_num_batches_tracked = self.bn1.num_batches_tracked 2025-09-09T14:53:53.3676788Z bn2_running_mean = self.bn2.running_mean 2025-09-09T14:53:53.3678542Z bn2_running_var = self.bn2.running_var 2025-09-09T14:53:53.3678893Z bn2_num_batches_tracked = self.bn2.num_batches_tracked 2025-09-09T14:53:53.3679356Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:53:53.3679962Z add_ = torch.ops.aten.add_.Tensor(bn1_num_batches_tracked, 1); bn1_num_batches_tracked = add_ = None 2025-09-09T14:53:53.3680647Z add__1 = torch.ops.aten.add_.Tensor(bn2_num_batches_tracked, 1); bn2_num_batches_tracked = add__1 = None 2025-09-09T14:53:53.3681190Z add = torch.ops.aten.add.Tensor(bn2_running_var, 1e-05) 2025-09-09T14:53:53.3681598Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:53:53.3682020Z div = torch.ops.aten.div.Tensor(bn2_weight, sqrt); sqrt = None 2025-09-09T14:53:53.3682482Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T14:53:53.3683006Z mul = torch.ops.aten.mul.Tensor(conv2_weight, reshape); conv2_weight = reshape = None 2025-09-09T14:53:53.3683588Z activation_post_process_3 = self.activation_post_process_3(mul); mul = None 2025-09-09T14:53:53.3684216Z zeros_like = torch.ops.aten.zeros_like.default(conv2_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:53:53.3684822Z add_2 = torch.ops.aten.add.Tensor(bn1_running_var, 1e-05) 2025-09-09T14:53:53.3685245Z sqrt_1 = torch.ops.aten.sqrt.default(add_2); add_2 = None 2025-09-09T14:53:53.3685691Z div_2 = torch.ops.aten.div.Tensor(bn1_weight, sqrt_1); sqrt_1 = None 2025-09-09T14:53:53.3686163Z reshape_3 = torch.ops.aten.reshape.default(div_2, [-1, 1, 1, 1]) 2025-09-09T14:53:53.3686717Z mul_1 = torch.ops.aten.mul.Tensor(conv1_weight, reshape_3); conv1_weight = reshape_3 = None 2025-09-09T14:53:53.3687314Z activation_post_process_1 = self.activation_post_process_1(mul_1); mul_1 = None 2025-09-09T14:53:53.3688179Z conv2d_3 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T14:53:53.3689032Z reshape_4 = torch.ops.aten.reshape.default(div_2, [1, -1, 1, 1]); div_2 = None 2025-09-09T14:53:53.3689592Z div_3 = torch.ops.aten.div.Tensor(conv2d_3, reshape_4); conv2d_3 = reshape_4 = None 2025-09-09T14:53:53.3690534Z batch_norm_3 = torch.ops.aten.batch_norm.default(div_3, bn1_weight, bn1_bias, bn1_running_mean, bn1_running_var, True, 0.1, 1e-05, True); div_3 = bn1_weight = bn1_bias = bn1_running_mean = bn1_running_var = None 2025-09-09T14:53:53.3691500Z activation_post_process_2 = self.activation_post_process_2(batch_norm_3); batch_norm_3 = None 2025-09-09T14:53:53.3692467Z conv2d_2 = torch.ops.aten.conv2d.default(activation_post_process_2, activation_post_process_3, zeros_like); activation_post_process_2 = activation_post_process_3 = zeros_like = None 2025-09-09T14:53:53.3693361Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T14:53:53.3693903Z div_1 = torch.ops.aten.div.Tensor(conv2d_2, reshape_1); conv2d_2 = reshape_1 = None 2025-09-09T14:53:53.3694521Z reshape_2 = torch.ops.aten.reshape.default(conv2_bias, [1, -1, 1, 1]); conv2_bias = None 2025-09-09T14:53:53.3695147Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:53:53.3696147Z batch_norm_2 = torch.ops.aten.batch_norm.default(add_1, bn2_weight, bn2_bias, bn2_running_mean, bn2_running_var, True, 0.1, 1e-05, True); add_1 = bn2_weight = bn2_bias = bn2_running_mean = bn2_running_var = None 2025-09-09T14:53:53.3697123Z activation_post_process_4 = self.activation_post_process_4(batch_norm_2); batch_norm_2 = None 2025-09-09T14:53:53.3698010Z return pytree.tree_unflatten((activation_post_process_4,), self._out_spec) 2025-09-09T14:53:53.3698412Z 2025-09-09T14:53:53.3698842Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:53:53.3699224Z model fx: GraphModule( 2025-09-09T14:53:53.3699562Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:53.3700620Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0189]), zero_point=tensor([-17], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:53:53.3701736Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.0985729694366455, max_val=2.7226178646087646) 2025-09-09T14:53:53.3702251Z ) 2025-09-09T14:53:53.3702443Z (conv1): ConvBn2d( 2025-09-09T14:53:53.3702706Z 3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False 2025-09-09T14:53:53.3703148Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:53:53.3703622Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:53.3704540Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:53:53.3705669Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.19212442636489868, max_val=0.18097376823425293) 2025-09-09T14:53:53.3706187Z ) 2025-09-09T14:53:53.3706367Z ) 2025-09-09T14:53:53.3706651Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:53:53.3707584Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0174]), zero_point=tensor([-29], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:53:53.3708695Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.7288155555725098, max_val=2.7138354778289795) 2025-09-09T14:53:53.3709214Z ) 2025-09-09T14:54:34.1047374Z (conv2): ConvBn2d( 2025-09-09T14:54:34.1047709Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T14:54:34.1048272Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:54:34.1048870Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:54:34.1050021Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:54:34.1051456Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.1921343356370926, max_val=0.1768510341644287) 2025-09-09T14:54:34.1052110Z ) 2025-09-09T14:54:34.1052331Z ) 2025-09-09T14:54:34.1052675Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:54:34.1053855Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0110]), zero_point=tensor([-1], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:54:34.1055248Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.4025696516036987, max_val=1.4086220264434814) 2025-09-09T14:54:34.1056012Z ) 2025-09-09T14:54:34.1056229Z ) 2025-09-09T14:54:34.1056348Z 2025-09-09T14:54:34.1056354Z 2025-09-09T14:54:34.1056368Z 2025-09-09T14:54:34.1056474Z def forward(self, x): 2025-09-09T14:54:34.1056936Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:54:34.1057599Z conv1 = self.conv1(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:54:34.1058340Z activation_post_process_1 = self.activation_post_process_1(conv1); conv1 = None 2025-09-09T14:54:34.1059024Z conv2 = self.conv2(activation_post_process_1); activation_post_process_1 = None 2025-09-09T14:54:34.1059709Z activation_post_process_2 = self.activation_post_process_2(conv2); conv2 = None 2025-09-09T14:54:34.1060520Z return activation_post_process_2 2025-09-09T14:54:34.1060843Z 2025-09-09T14:54:34.1061190Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:54:34.1061787Z diff: tensor([[[[0.]], 2025-09-09T14:54:34.1061956Z 2025-09-09T14:54:34.1062055Z [[0.]], 2025-09-09T14:54:34.1062220Z 2025-09-09T14:54:34.1062310Z [[0.]]], 2025-09-09T14:54:34.1062455Z 2025-09-09T14:54:34.1062460Z 2025-09-09T14:54:34.1062561Z [[[0.]], 2025-09-09T14:54:34.1062705Z 2025-09-09T14:54:34.1062793Z [[0.]], 2025-09-09T14:54:34.1062936Z 2025-09-09T14:54:34.1063032Z [[0.]]], 2025-09-09T14:54:34.1063175Z 2025-09-09T14:54:34.1063180Z 2025-09-09T14:54:34.1063268Z [[[0.]], 2025-09-09T14:54:34.1063416Z 2025-09-09T14:54:34.1063507Z [[0.]], 2025-09-09T14:54:34.1063653Z 2025-09-09T14:54:34.1063775Z [[0.]]]], grad_fn=) 2025-09-09T14:54:34.1064139Z converted model pt2e: GraphModule( 2025-09-09T14:54:34.1064468Z (conv1): Module() 2025-09-09T14:54:34.1064710Z (bn1): Module() 2025-09-09T14:54:34.1064951Z (conv2): Module() 2025-09-09T14:54:34.1065199Z (bn2): Module() 2025-09-09T14:54:34.1065437Z ) 2025-09-09T14:54:34.1065555Z 2025-09-09T14:54:34.1065560Z 2025-09-09T14:54:34.1065564Z 2025-09-09T14:54:34.1065670Z def forward(self, x): 2025-09-09T14:54:34.1066012Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:54:34.1066416Z conv2_bias = self.conv2.bias 2025-09-09T14:54:34.1067230Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01890663057565689, -17, -128, 127, torch.int8); x = None 2025-09-09T14:54:34.1068828Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01890663057565689, -17, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:54:34.1069914Z quantize_per_tensor_1 = self._frozen_param0 2025-09-09T14:54:34.1070915Z dequantize_per_tensor_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_1, 0.001512790797278285, 0, -127, 127, torch.int8); quantize_per_tensor_1 = None 2025-09-09T14:54:34.1071924Z conv1_weight_bias = self.conv1.weight_bias 2025-09-09T14:54:34.1072982Z conv2d_5 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_tensor_1, conv1_weight_bias); dequantize_per_tensor_default = dequantize_per_tensor_1 = conv1_weight_bias = None 2025-09-09T14:54:34.1074571Z quantize_per_tensor_default_3 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_5, 0.017422160133719444, -29, -128, 127, torch.int8); conv2d_5 = None 2025-09-09T14:54:34.1076192Z dequantize_per_tensor_default_3 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_3, 0.017422160133719444, -29, -128, 127, torch.int8); quantize_per_tensor_default_3 = None 2025-09-09T14:54:34.1077307Z quantize_per_tensor = self._frozen_param1 2025-09-09T14:54:34.1078345Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.001512868795543909, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T14:54:34.1079930Z conv2d_4 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default_3, dequantize_per_tensor, conv2_bias); dequantize_per_tensor_default_3 = dequantize_per_tensor = conv2_bias = None 2025-09-09T14:54:34.1081439Z quantize_per_tensor_default_4 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_4, 0.011024280451238155, -1, -128, 127, torch.int8); conv2d_4 = None 2025-09-09T14:54:34.1083050Z dequantize_per_tensor_default_4 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_4, 0.011024280451238155, -1, -128, 127, torch.int8); quantize_per_tensor_default_4 = None 2025-09-09T14:54:34.1084300Z return pytree.tree_unflatten((dequantize_per_tensor_default_4,), self._out_spec) 2025-09-09T14:54:34.1084904Z 2025-09-09T14:54:34.1085240Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:54:34.1085725Z onverted model fx: GraphModule( 2025-09-09T14:54:34.1086340Z (conv1): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T14:54:34.1086982Z (conv2): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T14:54:34.1087363Z ) 2025-09-09T14:54:34.1087462Z 2025-09-09T14:54:34.1087466Z 2025-09-09T14:54:34.1087470Z 2025-09-09T14:54:34.1087554Z def forward(self, x): 2025-09-09T14:54:34.1088174Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01890663057565689, -17, -128, 127, torch.int8); x = None 2025-09-09T14:54:34.1089409Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01890663057565689, -17, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:54:34.1090419Z conv1 = self.conv1(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:54:34.1091281Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1, 0.017422160133719444, -29, -128, 127, torch.int8); conv1 = None 2025-09-09T14:54:34.1092554Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.017422160133719444, -29, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:54:34.1093583Z conv2 = self.conv2(dequantize_per_tensor_default_1); dequantize_per_tensor_default_1 = None 2025-09-09T14:54:34.1094449Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2, 0.011024280451238155, -1, -128, 127, torch.int8); conv2 = None 2025-09-09T14:54:34.1095717Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.011024280451238155, -1, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:54:34.1096687Z return dequantize_per_tensor_default_2 2025-09-09T14:54:34.1096969Z 2025-09-09T14:54:34.1097242Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:54:34.1097761Z diff: tensor([[[[0.]], 2025-09-09T14:54:34.1097925Z 2025-09-09T14:54:34.1098014Z [[0.]], 2025-09-09T14:54:34.1098148Z 2025-09-09T14:54:34.1098224Z [[0.]]], 2025-09-09T14:54:34.1098345Z 2025-09-09T14:54:34.1098348Z 2025-09-09T14:54:34.1098421Z [[[0.]], 2025-09-09T14:54:34.1098550Z 2025-09-09T14:54:34.1098622Z [[0.]], 2025-09-09T14:54:34.1098737Z 2025-09-09T14:54:34.1098820Z [[0.]]], 2025-09-09T14:54:34.1098938Z 2025-09-09T14:54:34.1098942Z 2025-09-09T14:54:34.1099015Z [[[0.]], 2025-09-09T14:54:34.1099131Z 2025-09-09T14:54:34.1099212Z [[0.]], 2025-09-09T14:54:34.1099326Z 2025-09-09T14:54:34.1099400Z [[0.]]]]) 2025-09-09T14:54:34.1099823Z PASSED 2025-09-09T14:54:34.1100530Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_bn_per_channel_weight_bias PASSED 2025-09-09T14:54:34.1101525Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_bn_relu_fusion model pt2e: GraphModule( 2025-09-09T14:54:34.1102144Z (conv): Module() 2025-09-09T14:54:34.1102345Z (bn): Module() 2025-09-09T14:54:34.1102647Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:54:34.1103582Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:54:34.1104680Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:54:34.1105183Z ) 2025-09-09T14:54:34.1106508Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:54:34.1107510Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0014, 0.0015, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:54:34.1108974Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1835, -0.1822, -0.1883]), max_val=tensor([0.1799, 0.1856, 0.1719])) 2025-09-09T14:54:34.1109619Z ) 2025-09-09T14:54:34.1109898Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:54:34.1110833Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0049]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:54:34.1111898Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.2505061626434326) 2025-09-09T14:54:34.1112361Z ) 2025-09-09T14:54:34.1112532Z ) 2025-09-09T14:54:34.1112627Z 2025-09-09T14:54:34.1112631Z 2025-09-09T14:54:34.1112641Z 2025-09-09T14:54:34.1112734Z def forward(self, x): 2025-09-09T14:54:34.1113014Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:54:34.1113353Z conv_weight = self.conv.weight 2025-09-09T14:54:34.1113628Z conv_bias = self.conv.bias 2025-09-09T14:54:34.1113885Z bn_weight = self.bn.weight 2025-09-09T14:54:34.1114129Z bn_bias = self.bn.bias 2025-09-09T14:54:34.1114392Z bn_running_mean = self.bn.running_mean 2025-09-09T14:54:34.1114687Z bn_running_var = self.bn.running_var 2025-09-09T14:54:34.1115019Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:54:34.1115454Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:54:50.9687906Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:54:50.9688605Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:54:50.9689095Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:54:50.9689623Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:54:50.9690171Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T14:54:50.9690811Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:54:50.9691520Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:54:50.9692284Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:54:50.9693593Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:54:50.9694728Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T14:54:50.9695402Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T14:54:50.9696253Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T14:54:50.9696949Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:54:50.9698207Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:54:50.9699265Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T14:54:50.9699915Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T14:54:50.9700594Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:54:50.9701076Z 2025-09-09T14:54:50.9701659Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:54:50.9702121Z model fx: GraphModule( 2025-09-09T14:54:50.9702515Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:54:50.9703904Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:54:50.9705295Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:54:50.9705929Z ) 2025-09-09T14:54:50.9706165Z (conv): ConvBnReLU2d( 2025-09-09T14:54:50.9706467Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T14:54:50.9706983Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:54:50.9707556Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:54:50.9708765Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0014, 0.0015, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:54:50.9710402Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1835, -0.1822, -0.1883]), max_val=tensor([0.1799, 0.1856, 0.1719])) 2025-09-09T14:54:50.9711217Z ) 2025-09-09T14:54:50.9711442Z ) 2025-09-09T14:54:50.9711779Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:54:50.9712969Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0049]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:54:50.9714304Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.2505061626434326) 2025-09-09T14:54:50.9714895Z ) 2025-09-09T14:54:50.9715117Z ) 2025-09-09T14:54:50.9715242Z 2025-09-09T14:54:50.9715247Z 2025-09-09T14:54:50.9715252Z 2025-09-09T14:54:50.9715360Z def forward(self, x): 2025-09-09T14:54:50.9715802Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:54:50.9716459Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:54:50.9717143Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:54:50.9717679Z return activation_post_process_1 2025-09-09T14:54:50.9718000Z 2025-09-09T14:54:50.9718347Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:54:50.9718800Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:54:50.9719103Z [0., 0., 0.], 2025-09-09T14:54:50.9719364Z [0., 0., 0.]], 2025-09-09T14:54:50.9719546Z 2025-09-09T14:54:50.9719645Z [[0., 0., 0.], 2025-09-09T14:54:50.9719903Z [0., 0., 0.], 2025-09-09T14:54:50.9720173Z [0., 0., 0.]], 2025-09-09T14:54:50.9720348Z 2025-09-09T14:54:50.9720451Z [[0., 0., 0.], 2025-09-09T14:54:50.9720707Z [0., 0., 0.], 2025-09-09T14:54:50.9721014Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T14:54:50.9721397Z converted model pt2e: GraphModule( 2025-09-09T14:54:50.9721728Z (conv): Module() 2025-09-09T14:54:50.9721976Z (bn): Module() 2025-09-09T14:54:50.9722225Z ) 2025-09-09T14:54:50.9722348Z 2025-09-09T14:54:50.9722353Z 2025-09-09T14:54:50.9722358Z 2025-09-09T14:54:50.9722465Z def forward(self, x): 2025-09-09T14:54:50.9722841Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:54:50.9723285Z conv_bias = self.conv.bias 2025-09-09T14:54:50.9724082Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:54:50.9725735Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:54:50.9726814Z _scale_0 = self._scale_0 2025-09-09T14:54:50.9727218Z _zero_point_0 = self._zero_point_0 2025-09-09T14:54:50.9727596Z quantize_per_channel = self._frozen_param0 2025-09-09T14:54:50.9728697Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T14:54:50.9730386Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T14:54:50.9731510Z relu = torch.ops.aten.relu.default(conv2d_2); conv2d_2 = None 2025-09-09T14:54:50.9732449Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.004903945606201887, -128, -128, 127, torch.int8); relu = None 2025-09-09T14:54:50.9733790Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.004903945606201887, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:54:50.9734816Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T14:54:50.9735237Z 2025-09-09T14:54:50.9735515Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:54:50.9735967Z onverted model fx: GraphModule( 2025-09-09T14:54:50.9736227Z (conv): ConvReLU2d( 2025-09-09T14:54:50.9736562Z (0): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T14:54:50.9736939Z (1): ReLU() 2025-09-09T14:54:50.9737125Z ) 2025-09-09T14:54:50.9737304Z ) 2025-09-09T14:54:50.9737400Z 2025-09-09T14:54:50.9737404Z 2025-09-09T14:54:50.9737408Z 2025-09-09T14:54:50.9737492Z def forward(self, x): 2025-09-09T14:54:50.9738114Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:54:50.9739362Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:54:50.9740360Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:54:50.9741222Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.004903945606201887, -128, -128, 127, torch.int8); conv = None 2025-09-09T14:54:50.9742501Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.004903945606201887, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:54:50.9743393Z return dequantize_per_tensor_default_1 2025-09-09T14:54:50.9743668Z 2025-09-09T14:54:50.9743944Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:54:50.9744319Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:54:50.9744551Z [0., 0., 0.], 2025-09-09T14:54:50.9744764Z [0., 0., 0.]], 2025-09-09T14:54:50.9744902Z 2025-09-09T14:54:50.9744977Z [[0., 0., 0.], 2025-09-09T14:54:50.9745185Z [0., 0., 0.], 2025-09-09T14:54:50.9745394Z [0., 0., 0.]], 2025-09-09T14:54:50.9745536Z 2025-09-09T14:54:50.9745614Z [[0., 0., 0.], 2025-09-09T14:54:50.9745824Z [0., 0., 0.], 2025-09-09T14:54:50.9746031Z [0., 0., 0.]]]]) 2025-09-09T14:54:50.9746268Z model pt2e: GraphModule( 2025-09-09T14:54:50.9746497Z (conv): Module() 2025-09-09T14:54:50.9746701Z (bn): Module() 2025-09-09T14:54:50.9746996Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:54:50.9748037Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:54:50.9749197Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:54:50.9749697Z ) 2025-09-09T14:54:50.9757798Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:54:50.9758800Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:54:50.9759934Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.1882954239845276, max_val=0.1855725795030594) 2025-09-09T14:54:50.9760445Z ) 2025-09-09T14:54:50.9760743Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:54:50.9761692Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0049]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:54:50.9762990Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.2396948337554932) 2025-09-09T14:54:50.9763566Z ) 2025-09-09T14:54:50.9763767Z ) 2025-09-09T14:54:50.9763872Z 2025-09-09T14:54:50.9763876Z 2025-09-09T14:54:50.9763880Z 2025-09-09T14:54:50.9763975Z def forward(self, x): 2025-09-09T14:55:16.7300556Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:55:16.7301514Z conv_weight = self.conv.weight 2025-09-09T14:55:16.7301907Z conv_bias = self.conv.bias 2025-09-09T14:55:16.7302268Z bn_weight = self.bn.weight 2025-09-09T14:55:16.7302612Z bn_bias = self.bn.bias 2025-09-09T14:55:16.7302977Z bn_running_mean = self.bn.running_mean 2025-09-09T14:55:16.7303435Z bn_running_var = self.bn.running_var 2025-09-09T14:55:16.7303927Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:55:16.7304459Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:55:16.7305225Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:55:16.7305972Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:55:16.7306504Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:55:16.7307003Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:55:16.7307453Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T14:55:16.7307975Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:55:16.7308551Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:55:16.7309177Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:55:16.7310163Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:55:16.7311115Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T14:55:16.7311678Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T14:55:16.7312277Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T14:55:16.7312842Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:55:16.7313734Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:55:16.7314926Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T14:55:16.7315475Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T14:55:16.7316188Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:55:16.7316583Z 2025-09-09T14:55:16.7316882Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:55:16.7317252Z model fx: GraphModule( 2025-09-09T14:55:16.7317593Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:55:16.7318553Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:55:16.7319699Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:55:16.7320219Z ) 2025-09-09T14:55:16.7320416Z (conv): ConvBnReLU2d( 2025-09-09T14:55:16.7320684Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T14:55:16.7321110Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:55:16.7321640Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:55:16.7322585Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:55:16.7323725Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.1882954239845276, max_val=0.1855725795030594) 2025-09-09T14:55:16.7324255Z ) 2025-09-09T14:55:16.7324439Z ) 2025-09-09T14:55:16.7324737Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:55:16.7325691Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0049]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:55:16.7326771Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.2396948337554932) 2025-09-09T14:55:16.7327260Z ) 2025-09-09T14:55:16.7327434Z ) 2025-09-09T14:55:16.7327561Z 2025-09-09T14:55:16.7327566Z 2025-09-09T14:55:16.7327570Z 2025-09-09T14:55:16.7327661Z def forward(self, x): 2025-09-09T14:55:16.7328030Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:55:16.7328572Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:55:16.7329133Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:55:16.7329571Z return activation_post_process_1 2025-09-09T14:55:16.7329849Z 2025-09-09T14:55:16.7330139Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:55:16.7330522Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:55:16.7330778Z [0., 0., 0.], 2025-09-09T14:55:16.7330994Z [0., 0., 0.]], 2025-09-09T14:55:16.7331138Z 2025-09-09T14:55:16.7331228Z [[0., 0., 0.], 2025-09-09T14:55:16.7331450Z [0., 0., 0.], 2025-09-09T14:55:16.7331669Z [0., 0., 0.]], 2025-09-09T14:55:16.7331818Z 2025-09-09T14:55:16.7331914Z [[0., 0., 0.], 2025-09-09T14:55:16.7332156Z [0., 0., 0.], 2025-09-09T14:55:16.7332405Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T14:55:16.7332721Z converted model pt2e: GraphModule( 2025-09-09T14:55:16.7332996Z (conv): Module() 2025-09-09T14:55:16.7333203Z (bn): Module() 2025-09-09T14:55:16.7333405Z ) 2025-09-09T14:55:16.7333521Z 2025-09-09T14:55:16.7333527Z 2025-09-09T14:55:16.7333532Z 2025-09-09T14:55:16.7333653Z def forward(self, x): 2025-09-09T14:55:16.7334068Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:55:16.7334498Z conv_bias = self.conv.bias 2025-09-09T14:55:16.7335275Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:55:16.7336701Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:55:16.7337583Z quantize_per_tensor = self._frozen_param0 2025-09-09T14:55:16.7338391Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.0014826410915702581, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T14:55:16.7339731Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T14:55:16.7340674Z relu = torch.ops.aten.relu.default(conv2d_2); conv2d_2 = None 2025-09-09T14:55:16.7341527Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.004861548542976379, -128, -128, 127, torch.int8); relu = None 2025-09-09T14:55:16.7342831Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.004861548542976379, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:55:16.7343861Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:55:16.7344284Z 2025-09-09T14:55:16.7344574Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:55:16.7344966Z onverted model fx: GraphModule( 2025-09-09T14:55:16.7345230Z (conv): ConvReLU2d( 2025-09-09T14:55:16.7345580Z (0): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T14:55:16.7345960Z (1): ReLU() 2025-09-09T14:55:16.7346166Z ) 2025-09-09T14:55:16.7346353Z ) 2025-09-09T14:55:16.7346466Z 2025-09-09T14:55:16.7346474Z 2025-09-09T14:55:16.7346478Z 2025-09-09T14:55:16.7346569Z def forward(self, x): 2025-09-09T14:55:16.7347203Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:55:16.7348448Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:55:16.7349475Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:55:16.7350349Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.004861548542976379, -128, -128, 127, torch.int8); conv = None 2025-09-09T14:55:16.7351690Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.004861548542976379, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:55:16.7352591Z return dequantize_per_tensor_default_1 2025-09-09T14:55:16.7352874Z 2025-09-09T14:55:16.7353170Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:55:16.7353554Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:55:16.7353797Z [0., 0., 0.], 2025-09-09T14:55:16.7354022Z [0., 0., 0.]], 2025-09-09T14:55:16.7354166Z 2025-09-09T14:55:16.7354248Z [[0., 0., 0.], 2025-09-09T14:55:16.7354468Z [0., 0., 0.], 2025-09-09T14:55:16.7354680Z [0., 0., 0.]], 2025-09-09T14:55:16.7354827Z 2025-09-09T14:55:16.7354907Z [[0., 0., 0.], 2025-09-09T14:55:16.7355118Z [0., 0., 0.], 2025-09-09T14:55:16.7355338Z [0., 0., 0.]]]]) 2025-09-09T14:55:16.7355795Z PASSED 2025-09-09T14:55:16.7356418Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_bn_relu_fusion_cuda model pt2e: GraphModule( 2025-09-09T14:55:16.7357171Z (conv): Module() 2025-09-09T14:55:16.7357384Z (bn): Module() 2025-09-09T14:55:16.7357702Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:55:16.7358947Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0183], device='cuda:0'), zero_point=tensor([10], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:55:16.7360228Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:55:16.7360750Z ) 2025-09-09T14:55:16.7361036Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:55:16.7362273Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0015, 0.0015, 0.0014], device='cuda:0'), zero_point=tensor([0, 0, 0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:55:30.7716113Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1860, -0.1897, -0.1787], device='cuda:0'), max_val=tensor([0.1824, 0.1870, 0.1478], device='cuda:0')) 2025-09-09T14:55:30.7717020Z ) 2025-09-09T14:55:30.7717320Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:55:30.7718812Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0081], device='cuda:0'), zero_point=tensor([-128], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:55:30.7720258Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=2.0564441680908203) 2025-09-09T14:55:30.7720919Z ) 2025-09-09T14:55:30.7721159Z ) 2025-09-09T14:55:30.7721298Z 2025-09-09T14:55:30.7721312Z 2025-09-09T14:55:30.7721316Z 2025-09-09T14:55:30.7721406Z def forward(self, x): 2025-09-09T14:55:30.7721706Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:55:30.7722052Z conv_weight = self.conv.weight 2025-09-09T14:55:30.7722344Z conv_bias = self.conv.bias 2025-09-09T14:55:30.7722605Z bn_weight = self.bn.weight 2025-09-09T14:55:30.7722866Z bn_bias = self.bn.bias 2025-09-09T14:55:30.7723131Z bn_running_mean = self.bn.running_mean 2025-09-09T14:55:30.7723445Z bn_running_var = self.bn.running_var 2025-09-09T14:55:30.7723782Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:55:30.7724232Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:55:30.7724831Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:55:30.7725366Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:55:30.7725779Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:55:30.7726245Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:55:30.7726708Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T14:55:30.7727222Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:55:30.7727793Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:55:30.7728418Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:55:30.7729393Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:55:30.7730297Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T14:55:30.7731037Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T14:55:30.7731640Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T14:55:30.7732334Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:55:30.7733204Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:55:30.7734055Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T14:55:30.7734577Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T14:55:30.7735123Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:55:30.7735516Z 2025-09-09T14:55:30.7735885Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:55:30.7736262Z model fx: GraphModule( 2025-09-09T14:55:30.7736595Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:55:30.7737718Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0183], device='cuda:0'), zero_point=tensor([10], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:55:30.7739002Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:55:30.7739508Z ) 2025-09-09T14:55:30.7739706Z (conv): ConvBnReLU2d( 2025-09-09T14:55:30.7739952Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T14:55:30.7740370Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:55:30.7740836Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:55:30.7741986Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0015, 0.0015, 0.0014], device='cuda:0'), zero_point=tensor([0, 0, 0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:55:30.7743565Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1860, -0.1897, -0.1787], device='cuda:0'), max_val=tensor([0.1824, 0.1870, 0.1478], device='cuda:0')) 2025-09-09T14:55:30.7744302Z ) 2025-09-09T14:55:30.7744479Z ) 2025-09-09T14:55:30.7744763Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:55:30.7745882Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0081], device='cuda:0'), zero_point=tensor([-128], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:55:30.7747123Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=2.0564441680908203) 2025-09-09T14:55:30.7747599Z ) 2025-09-09T14:55:30.7747770Z ) 2025-09-09T14:55:30.7747882Z 2025-09-09T14:55:30.7747887Z 2025-09-09T14:55:30.7747892Z 2025-09-09T14:55:30.7747981Z def forward(self, x): 2025-09-09T14:55:30.7748333Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:55:30.7748874Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:55:30.7749427Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:55:30.7749860Z return activation_post_process_1 2025-09-09T14:55:30.7750129Z 2025-09-09T14:55:30.7750429Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:55:30.7750814Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:55:30.7751068Z [0., 0., 0.], 2025-09-09T14:55:30.7751282Z [0., 0., 0.]], 2025-09-09T14:55:30.7751424Z 2025-09-09T14:55:30.7751604Z [[0., 0., 0.], 2025-09-09T14:55:30.7751817Z [0., 0., 0.], 2025-09-09T14:55:30.7752031Z [0., 0., 0.]], 2025-09-09T14:55:30.7752250Z 2025-09-09T14:55:30.7752329Z [[0., 0., 0.], 2025-09-09T14:55:30.7752542Z [0., 0., 0.], 2025-09-09T14:55:30.7752807Z [0., 0., 0.]]]], device='cuda:0', grad_fn=) 2025-09-09T14:55:30.7753148Z converted model pt2e: GraphModule( 2025-09-09T14:55:30.7753412Z (conv): Module() 2025-09-09T14:55:30.7753623Z (bn): Module() 2025-09-09T14:55:30.7753821Z ) 2025-09-09T14:55:30.7753922Z 2025-09-09T14:55:30.7753926Z 2025-09-09T14:55:30.7753930Z 2025-09-09T14:55:30.7754017Z def forward(self, x): 2025-09-09T14:55:30.7754310Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:55:30.7754648Z conv_bias = self.conv.bias 2025-09-09T14:55:30.7755309Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:55:30.7756556Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:55:30.7757417Z _scale_0 = self._scale_0 2025-09-09T14:55:30.7757687Z _zero_point_0 = self._zero_point_0 2025-09-09T14:55:30.7757996Z quantize_per_channel = self._frozen_param0 2025-09-09T14:55:30.7758889Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T14:55:30.7760242Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T14:55:30.7761096Z relu = torch.ops.aten.relu.default(conv2d_2); conv2d_2 = None 2025-09-09T14:55:30.7761891Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.008064487017691135, -128, -128, 127, torch.int8); relu = None 2025-09-09T14:55:30.7763191Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.008064487017691135, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:55:30.7764216Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T14:55:30.7764635Z 2025-09-09T14:55:30.7764924Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:55:30.7765311Z onverted model fx: GraphModule( 2025-09-09T14:55:30.7765572Z (conv): ConvReLU2d( 2025-09-09T14:55:30.7765921Z (0): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T14:55:30.7766346Z (1): ReLU() 2025-09-09T14:55:30.7766548Z ) 2025-09-09T14:55:30.7766724Z ) 2025-09-09T14:55:30.7766836Z 2025-09-09T14:55:30.7766840Z 2025-09-09T14:55:30.7766844Z 2025-09-09T14:55:30.7766933Z def forward(self, x): 2025-09-09T14:55:30.7767559Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:55:30.7768801Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:55:30.7769822Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:55:30.7770692Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.008064487017691135, -128, -128, 127, torch.int8); conv = None 2025-09-09T14:55:30.7772062Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.008064487017691135, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:55:30.7772961Z return dequantize_per_tensor_default_1 2025-09-09T14:55:30.7773317Z 2025-09-09T14:55:30.7773612Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:55:30.7773994Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:55:30.7774237Z [0., 0., 0.], 2025-09-09T14:55:30.7774461Z [0., 0., 0.]], 2025-09-09T14:55:30.7774605Z 2025-09-09T14:55:30.7774685Z [[0., 0., 0.], 2025-09-09T14:55:30.7774906Z [0., 0., 0.], 2025-09-09T14:55:30.7775121Z [0., 0., 0.]], 2025-09-09T14:55:30.7775269Z 2025-09-09T14:55:30.7775349Z [[0., 0., 0.], 2025-09-09T14:55:30.7775561Z [0., 0., 0.], 2025-09-09T14:55:47.6260745Z [0., 0., 0.]]]], device='cuda:0') 2025-09-09T14:55:47.6261209Z model pt2e: GraphModule( 2025-09-09T14:55:47.6261575Z (conv): Module() 2025-09-09T14:55:47.6261858Z (bn): Module() 2025-09-09T14:55:47.6262245Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:55:47.6263470Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0183], device='cuda:0'), zero_point=tensor([10], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:55:47.6264832Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:55:47.6265347Z ) 2025-09-09T14:55:47.6265632Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:55:47.6266744Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0015], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:55:47.6268052Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.18965020775794983, max_val=0.1870359182357788) 2025-09-09T14:55:47.6268576Z ) 2025-09-09T14:55:47.6268854Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:55:47.6269962Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0080], device='cuda:0'), zero_point=tensor([-128], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:55:47.6271179Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=2.0522286891937256) 2025-09-09T14:55:47.6271746Z ) 2025-09-09T14:55:47.6271916Z ) 2025-09-09T14:55:47.6272021Z 2025-09-09T14:55:47.6272025Z 2025-09-09T14:55:47.6272029Z 2025-09-09T14:55:47.6272117Z def forward(self, x): 2025-09-09T14:55:47.6272408Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:55:47.6272751Z conv_weight = self.conv.weight 2025-09-09T14:55:47.6273034Z conv_bias = self.conv.bias 2025-09-09T14:55:47.6273288Z bn_weight = self.bn.weight 2025-09-09T14:55:47.6273551Z bn_bias = self.bn.bias 2025-09-09T14:55:47.6273805Z bn_running_mean = self.bn.running_mean 2025-09-09T14:55:47.6274112Z bn_running_var = self.bn.running_var 2025-09-09T14:55:47.6274456Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:55:47.6274892Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:55:47.6275480Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:55:47.6276001Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:55:47.6276487Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:55:47.6276898Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:55:47.6277629Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T14:55:47.6278145Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:55:47.6278935Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:55:47.6279555Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:55:47.6280529Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:55:47.6281427Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T14:55:47.6281971Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T14:55:47.6282549Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T14:55:47.6283119Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:55:47.6283984Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:55:47.6284841Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T14:55:47.6285377Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T14:55:47.6285919Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:55:47.6286315Z 2025-09-09T14:55:47.6286597Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:55:47.6286967Z model fx: GraphModule( 2025-09-09T14:55:47.6287292Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:55:47.6288417Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0183], device='cuda:0'), zero_point=tensor([10], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:55:47.6289697Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:55:47.6290211Z ) 2025-09-09T14:55:47.6290409Z (conv): ConvBnReLU2d( 2025-09-09T14:55:47.6290654Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T14:55:47.6291073Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:55:47.6291538Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:55:47.6292646Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0015], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:55:47.6293969Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.18965020775794983, max_val=0.1870359182357788) 2025-09-09T14:55:47.6294488Z ) 2025-09-09T14:55:47.6294666Z ) 2025-09-09T14:55:47.6294946Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:55:47.6296153Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0080], device='cuda:0'), zero_point=tensor([-128], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:55:47.6297662Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=2.0522286891937256) 2025-09-09T14:55:47.6298139Z ) 2025-09-09T14:55:47.6298318Z ) 2025-09-09T14:55:47.6298417Z 2025-09-09T14:55:47.6298422Z 2025-09-09T14:55:47.6298426Z 2025-09-09T14:55:47.6298517Z def forward(self, x): 2025-09-09T14:55:47.6298998Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:55:47.6299540Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:55:47.6300190Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:55:47.6300621Z return activation_post_process_1 2025-09-09T14:55:47.6300879Z 2025-09-09T14:55:47.6301164Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:55:47.6301560Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:55:47.6301831Z [0., 0., 0.], 2025-09-09T14:55:47.6302052Z [0., 0., 0.]], 2025-09-09T14:55:47.6302193Z 2025-09-09T14:55:47.6302272Z [[0., 0., 0.], 2025-09-09T14:55:47.6302483Z [0., 0., 0.], 2025-09-09T14:55:47.6302691Z [0., 0., 0.]], 2025-09-09T14:55:47.6302835Z 2025-09-09T14:55:47.6302916Z [[0., 0., 0.], 2025-09-09T14:55:47.6303121Z [0., 0., 0.], 2025-09-09T14:55:47.6303396Z [0., 0., 0.]]]], device='cuda:0', grad_fn=) 2025-09-09T14:55:47.6303731Z converted model pt2e: GraphModule( 2025-09-09T14:55:47.6303998Z (conv): Module() 2025-09-09T14:55:47.6304219Z (bn): Module() 2025-09-09T14:55:47.6304411Z ) 2025-09-09T14:55:47.6304511Z 2025-09-09T14:55:47.6304515Z 2025-09-09T14:55:47.6304518Z 2025-09-09T14:55:47.6304615Z def forward(self, x): 2025-09-09T14:55:47.6304898Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:55:47.6305235Z conv_bias = self.conv.bias 2025-09-09T14:55:47.6305878Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:55:47.6307126Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:55:47.6308030Z quantize_per_tensor = self._frozen_param0 2025-09-09T14:55:47.6308825Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.0014933086931705475, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T14:55:47.6310101Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T14:55:47.6310949Z relu = torch.ops.aten.relu.default(conv2d_2); conv2d_2 = None 2025-09-09T14:55:47.6311722Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.00804795604199171, -128, -128, 127, torch.int8); relu = None 2025-09-09T14:55:47.6313008Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.00804795604199171, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:55:47.6314037Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:55:47.6314458Z 2025-09-09T14:55:47.6314750Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:55:47.6315132Z onverted model fx: GraphModule( 2025-09-09T14:55:47.6315392Z (conv): ConvReLU2d( 2025-09-09T14:55:47.6315730Z (0): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T14:55:47.6316112Z (1): ReLU() 2025-09-09T14:55:47.6316302Z ) 2025-09-09T14:55:47.6316489Z ) 2025-09-09T14:55:47.6316588Z 2025-09-09T14:55:47.6316592Z 2025-09-09T14:55:47.6316596Z 2025-09-09T14:55:47.6316682Z def forward(self, x): 2025-09-09T14:55:47.6317306Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:55:47.6318637Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:55:47.6319649Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:56:13.6026174Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.00804795604199171, -128, -128, 127, torch.int8); conv = None 2025-09-09T14:56:13.6028892Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.00804795604199171, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:56:13.6029947Z return dequantize_per_tensor_default_1 2025-09-09T14:56:13.6030339Z 2025-09-09T14:56:13.6038660Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:56:13.6039219Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:56:13.6039568Z [0., 0., 0.], 2025-09-09T14:56:13.6039798Z [0., 0., 0.]], 2025-09-09T14:56:13.6039955Z 2025-09-09T14:56:13.6040046Z [[0., 0., 0.], 2025-09-09T14:56:13.6040255Z [0., 0., 0.], 2025-09-09T14:56:13.6040469Z [0., 0., 0.]], 2025-09-09T14:56:13.6040619Z 2025-09-09T14:56:13.6040699Z [[0., 0., 0.], 2025-09-09T14:56:13.6040991Z [0., 0., 0.], 2025-09-09T14:56:13.6041303Z [0., 0., 0.]]]], device='cuda:0') 2025-09-09T14:56:13.6041909Z PASSED 2025-09-09T14:56:13.6042660Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_bn_relu_fusion_no_conv_bias model pt2e: GraphModule( 2025-09-09T14:56:13.6043321Z (conv): Module() 2025-09-09T14:56:13.6043526Z (bn): Module() 2025-09-09T14:56:13.6043822Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:13.6044792Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:56:13.6045909Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:56:13.6046430Z ) 2025-09-09T14:56:13.6046712Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:13.6047706Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:56:13.6049035Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1866, -0.1825, -0.1912]), max_val=tensor([0.1747, 0.1914, 0.1702])) 2025-09-09T14:56:13.6049725Z ) 2025-09-09T14:56:13.6050002Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:13.6050963Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0078]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:56:13.6052034Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.984864354133606) 2025-09-09T14:56:13.6052497Z ) 2025-09-09T14:56:13.6052670Z ) 2025-09-09T14:56:13.6052767Z 2025-09-09T14:56:13.6052772Z 2025-09-09T14:56:13.6052775Z 2025-09-09T14:56:13.6052860Z def forward(self, x): 2025-09-09T14:56:13.6053149Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:56:13.6053482Z conv_weight = self.conv.weight 2025-09-09T14:56:13.6053764Z bn_weight = self.bn.weight 2025-09-09T14:56:13.6054014Z bn_bias = self.bn.bias 2025-09-09T14:56:13.6054269Z bn_running_mean = self.bn.running_mean 2025-09-09T14:56:13.6054573Z bn_running_var = self.bn.running_var 2025-09-09T14:56:13.6054899Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:56:13.6055664Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:56:13.6056347Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:56:13.6057124Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:56:13.6057521Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:56:13.6057930Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:56:13.6058379Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T14:56:13.6058883Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:56:13.6059487Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:56:13.6060349Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T14:56:13.6061202Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T14:56:13.6061755Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T14:56:13.6062667Z batch_norm_1 = torch.ops.aten.batch_norm.default(div_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); div_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:56:13.6063526Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T14:56:13.6064048Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T14:56:13.6064599Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:56:13.6064994Z 2025-09-09T14:56:13.6065272Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:56:13.6065640Z model fx: GraphModule( 2025-09-09T14:56:13.6065965Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:13.6066928Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:56:13.6068048Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:56:13.6068552Z ) 2025-09-09T14:56:13.6068748Z (conv): ConvBnReLU2d( 2025-09-09T14:56:13.6069008Z 3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False 2025-09-09T14:56:13.6069447Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:56:13.6069910Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:13.6070905Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:56:13.6072229Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1866, -0.1825, -0.1912]), max_val=tensor([0.1747, 0.1914, 0.1702])) 2025-09-09T14:56:13.6072897Z ) 2025-09-09T14:56:13.6073071Z ) 2025-09-09T14:56:13.6073351Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:13.6074318Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0078]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:56:13.6075392Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.984864354133606) 2025-09-09T14:56:13.6075858Z ) 2025-09-09T14:56:13.6076030Z ) 2025-09-09T14:56:13.6076126Z 2025-09-09T14:56:13.6076130Z 2025-09-09T14:56:13.6076134Z 2025-09-09T14:56:13.6076221Z def forward(self, x): 2025-09-09T14:56:13.6076668Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:56:13.6077202Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:56:13.6077835Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:56:13.6078270Z return activation_post_process_1 2025-09-09T14:56:13.6078529Z 2025-09-09T14:56:13.6078811Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:56:13.6079176Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:56:13.6079419Z [0., 0., 0.], 2025-09-09T14:56:13.6079629Z [0., 0., 0.]], 2025-09-09T14:56:13.6079801Z 2025-09-09T14:56:13.6079883Z [[0., 0., 0.], 2025-09-09T14:56:13.6080105Z [0., 0., 0.], 2025-09-09T14:56:13.6080315Z [0., 0., 0.]], 2025-09-09T14:56:13.6080451Z 2025-09-09T14:56:13.6080533Z [[0., 0., 0.], 2025-09-09T14:56:13.6080732Z [0., 0., 0.], 2025-09-09T14:56:13.6080980Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T14:56:13.6081285Z converted model pt2e: GraphModule( 2025-09-09T14:56:13.6081551Z (conv): Module() 2025-09-09T14:56:13.6081756Z (bn): Module() 2025-09-09T14:56:13.6081951Z ) 2025-09-09T14:56:13.6082050Z 2025-09-09T14:56:13.6082054Z 2025-09-09T14:56:13.6082058Z 2025-09-09T14:56:13.6082140Z def forward(self, x): 2025-09-09T14:56:13.6082424Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:56:13.6083151Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:56:13.6084412Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:56:13.6085270Z _scale_0 = self._scale_0 2025-09-09T14:56:13.6085525Z _zero_point_0 = self._zero_point_0 2025-09-09T14:56:13.6085838Z quantize_per_channel = self._frozen_param0 2025-09-09T14:56:13.6086730Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T14:56:13.6087619Z conv_weight_bias = self.conv.weight_bias 2025-09-09T14:56:13.6088462Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_weight_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_weight_bias = None 2025-09-09T14:56:13.6089367Z relu = torch.ops.aten.relu.default(conv2d_2); conv2d_2 = None 2025-09-09T14:56:13.6090207Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.007783781737089157, -128, -128, 127, torch.int8); relu = None 2025-09-09T14:56:13.6091496Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.007783781737089157, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:56:13.6092514Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T14:56:13.6092924Z 2025-09-09T14:56:13.6093200Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:56:13.6093576Z onverted model fx: GraphModule( 2025-09-09T14:56:13.6093836Z (conv): ConvReLU2d( 2025-09-09T14:56:13.6094171Z (0): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T14:56:13.6094542Z (1): ReLU() 2025-09-09T14:56:13.6094729Z ) 2025-09-09T14:56:13.6094906Z ) 2025-09-09T14:56:13.6095002Z 2025-09-09T14:56:13.6095006Z 2025-09-09T14:56:13.6095010Z 2025-09-09T14:56:13.6095094Z def forward(self, x): 2025-09-09T14:56:30.1875205Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:56:30.1876657Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:56:30.1878031Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:56:30.1879048Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.007783781737089157, -128, -128, 127, torch.int8); conv = None 2025-09-09T14:56:30.1880493Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.007783781737089157, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:56:30.1881526Z return dequantize_per_tensor_default_1 2025-09-09T14:56:30.1881880Z 2025-09-09T14:56:30.1882173Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:56:30.1882600Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:56:30.1882888Z [0., 0., 0.], 2025-09-09T14:56:30.1883110Z [0., 0., 0.]], 2025-09-09T14:56:30.1883265Z 2025-09-09T14:56:30.1883347Z [[0., 0., 0.], 2025-09-09T14:56:30.1883646Z [0., 0., 0.], 2025-09-09T14:56:30.1883906Z [0., 0., 0.]], 2025-09-09T14:56:30.1884074Z 2025-09-09T14:56:30.1884152Z [[0., 0., 0.], 2025-09-09T14:56:30.1884411Z [0., 0., 0.], 2025-09-09T14:56:30.1884688Z [0., 0., 0.]]]]) 2025-09-09T14:56:30.1884924Z model pt2e: GraphModule( 2025-09-09T14:56:30.1885177Z (conv): Module() 2025-09-09T14:56:30.1885454Z (bn): Module() 2025-09-09T14:56:30.1885761Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:30.1886798Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:56:30.1888064Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:56:30.1888590Z ) 2025-09-09T14:56:30.1888952Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:30.1889976Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:56:30.1891184Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.19124282896518707, max_val=0.19141820073127747) 2025-09-09T14:56:30.1891793Z ) 2025-09-09T14:56:30.1892088Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:30.1893160Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0078]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:56:30.1894330Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.9838534593582153) 2025-09-09T14:56:30.1894851Z ) 2025-09-09T14:56:30.1895083Z ) 2025-09-09T14:56:30.1895208Z 2025-09-09T14:56:30.1895212Z 2025-09-09T14:56:30.1895217Z 2025-09-09T14:56:30.1895326Z def forward(self, x): 2025-09-09T14:56:30.1895622Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:56:30.1896141Z conv_weight = self.conv.weight 2025-09-09T14:56:30.1896426Z bn_weight = self.bn.weight 2025-09-09T14:56:30.1896739Z bn_bias = self.bn.bias 2025-09-09T14:56:30.1897032Z bn_running_mean = self.bn.running_mean 2025-09-09T14:56:30.1897544Z bn_running_var = self.bn.running_var 2025-09-09T14:56:30.1897964Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:56:30.1898448Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:56:30.1899289Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:56:30.1899903Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:56:30.1900521Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:56:30.1900950Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:56:30.1901484Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T14:56:30.1902067Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:56:30.1902656Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:56:30.1903597Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T14:56:30.1904534Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T14:56:30.1905188Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T14:56:30.1906195Z batch_norm_1 = torch.ops.aten.batch_norm.default(div_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); div_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:56:30.1907149Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T14:56:30.1907756Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T14:56:30.1908386Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:56:30.1908783Z 2025-09-09T14:56:30.1909153Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:56:30.1909526Z model fx: GraphModule( 2025-09-09T14:56:30.1909911Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:30.1910974Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:56:30.1912171Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:56:30.1912772Z ) 2025-09-09T14:56:30.1912964Z (conv): ConvBnReLU2d( 2025-09-09T14:56:30.1913250Z 3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False 2025-09-09T14:56:30.1913767Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:56:30.1914297Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:30.1915328Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:56:30.1916542Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.19124282896518707, max_val=0.19141820073127747) 2025-09-09T14:56:30.1917127Z ) 2025-09-09T14:56:30.1917334Z ) 2025-09-09T14:56:30.1917624Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:30.1918660Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0078]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:56:30.1919759Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.9838534593582153) 2025-09-09T14:56:30.1920234Z ) 2025-09-09T14:56:30.1920418Z ) 2025-09-09T14:56:30.1920520Z 2025-09-09T14:56:30.1920525Z 2025-09-09T14:56:30.1920529Z 2025-09-09T14:56:30.1920619Z def forward(self, x): 2025-09-09T14:56:30.1921062Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:56:30.1921787Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:56:30.1922354Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:56:30.1922875Z return activation_post_process_1 2025-09-09T14:56:30.1923144Z 2025-09-09T14:56:30.1923436Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:56:30.1923809Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:56:30.1924063Z [0., 0., 0.], 2025-09-09T14:56:30.1924280Z [0., 0., 0.]], 2025-09-09T14:56:30.1924432Z 2025-09-09T14:56:30.1924513Z [[0., 0., 0.], 2025-09-09T14:56:30.1924793Z [0., 0., 0.], 2025-09-09T14:56:30.1925041Z [0., 0., 0.]], 2025-09-09T14:56:30.1925203Z 2025-09-09T14:56:30.1925300Z [[0., 0., 0.], 2025-09-09T14:56:30.1925541Z [0., 0., 0.], 2025-09-09T14:56:30.1925788Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T14:56:30.1926103Z converted model pt2e: GraphModule( 2025-09-09T14:56:30.1926380Z (conv): Module() 2025-09-09T14:56:30.1926586Z (bn): Module() 2025-09-09T14:56:30.1926787Z ) 2025-09-09T14:56:30.1926888Z 2025-09-09T14:56:30.1926898Z 2025-09-09T14:56:30.1926902Z 2025-09-09T14:56:30.1926996Z def forward(self, x): 2025-09-09T14:56:30.1927282Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:56:30.1928019Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:56:30.1929264Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:56:30.1930151Z quantize_per_tensor = self._frozen_param0 2025-09-09T14:56:30.1930957Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.001507229870185256, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T14:56:30.1931780Z conv_weight_bias = self.conv.weight_bias 2025-09-09T14:56:30.1932698Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_weight_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_weight_bias = None 2025-09-09T14:56:30.1933615Z relu = torch.ops.aten.relu.default(conv2d_2); conv2d_2 = None 2025-09-09T14:56:30.1934410Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.007779817562550306, -128, -128, 127, torch.int8); relu = None 2025-09-09T14:56:30.1935756Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.007779817562550306, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:56:30.1936845Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:56:30.1937266Z 2025-09-09T14:56:30.1937557Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:56:30.1937946Z onverted model fx: GraphModule( 2025-09-09T14:56:30.1938223Z (conv): ConvReLU2d( 2025-09-09T14:56:30.1938567Z (0): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T14:56:30.1938948Z (1): ReLU() 2025-09-09T14:56:30.1939142Z ) 2025-09-09T14:56:30.1939334Z ) 2025-09-09T14:56:30.1939438Z 2025-09-09T14:56:30.1939442Z 2025-09-09T14:56:30.1939447Z 2025-09-09T14:56:30.1939535Z def forward(self, x): 2025-09-09T14:56:31.6506355Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:56:31.6508156Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:56:31.6509355Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:56:31.6510238Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.007779817562550306, -128, -128, 127, torch.int8); conv = None 2025-09-09T14:56:31.6511625Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.007779817562550306, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:56:31.6512514Z return dequantize_per_tensor_default_1 2025-09-09T14:56:31.6512795Z 2025-09-09T14:56:31.6513078Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:56:31.6513452Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:56:31.6513693Z [0., 0., 0.], 2025-09-09T14:56:31.6513914Z [0., 0., 0.]], 2025-09-09T14:56:31.6514056Z 2025-09-09T14:56:31.6514138Z [[0., 0., 0.], 2025-09-09T14:56:31.6514365Z [0., 0., 0.], 2025-09-09T14:56:31.6514576Z [0., 0., 0.]], 2025-09-09T14:56:31.6514721Z 2025-09-09T14:56:31.6514801Z [[0., 0., 0.], 2025-09-09T14:56:31.6515015Z [0., 0., 0.], 2025-09-09T14:56:31.6515235Z [0., 0., 0.]]]]) 2025-09-09T14:56:31.6515684Z PASSED 2025-09-09T14:56:31.6516263Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_no_bias model pt2e: GraphModule( 2025-09-09T14:56:31.6516867Z (conv): Module() 2025-09-09T14:56:31.6517174Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:31.6518176Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0014]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:56:31.6519488Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1860, -0.1897, -0.1787]), max_val=tensor([0.1824, 0.1870, 0.1478])) 2025-09-09T14:56:31.6520133Z ) 2025-09-09T14:56:31.6520426Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:31.6521371Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:56:31.6522458Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:56:31.6522969Z ) 2025-09-09T14:56:31.6523250Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:31.6524193Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0062]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:56:31.6525256Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.5882599353790283) 2025-09-09T14:56:31.6525726Z ) 2025-09-09T14:56:31.6525903Z ) 2025-09-09T14:56:31.6526011Z 2025-09-09T14:56:31.6526016Z 2025-09-09T14:56:31.6526020Z 2025-09-09T14:56:31.6526109Z def forward(self, x): 2025-09-09T14:56:31.6526402Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:56:31.6526742Z conv_weight = self.conv.weight 2025-09-09T14:56:31.6527213Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T14:56:31.6527810Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:56:31.6528611Z conv2d = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T14:56:31.6529370Z relu = torch.ops.aten.relu.default(conv2d); conv2d = None 2025-09-09T14:56:31.6530985Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T14:56:31.6531550Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:56:31.6532023Z 2025-09-09T14:56:31.6532311Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:56:31.6532682Z model fx: GraphModule( 2025-09-09T14:56:31.6533008Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:31.6533949Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:56:31.6535036Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:56:31.6535573Z ) 2025-09-09T14:56:31.6535781Z (conv): ConvReLU2d( 2025-09-09T14:56:31.6536138Z 3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False 2025-09-09T14:56:31.6536515Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:31.6537468Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0014]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:56:31.6538760Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1860, -0.1897, -0.1787]), max_val=tensor([0.1824, 0.1870, 0.1478])) 2025-09-09T14:56:31.6539402Z ) 2025-09-09T14:56:31.6539587Z ) 2025-09-09T14:56:31.6539869Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:31.6540805Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0062]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:56:31.6541890Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.5882599353790283) 2025-09-09T14:56:31.6542365Z ) 2025-09-09T14:56:31.6542540Z ) 2025-09-09T14:56:31.6542644Z 2025-09-09T14:56:31.6542653Z 2025-09-09T14:56:31.6542657Z 2025-09-09T14:56:31.6542753Z def forward(self, x): 2025-09-09T14:56:31.6543101Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:56:31.6543637Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:56:31.6544184Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:56:31.6544614Z return activation_post_process_1 2025-09-09T14:56:31.6544875Z 2025-09-09T14:56:31.6545160Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:56:31.6545559Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:56:31.6545823Z [0., 0., 0.], 2025-09-09T14:56:31.6546041Z [0., 0., 0.]], 2025-09-09T14:56:31.6546183Z 2025-09-09T14:56:31.6546268Z [[0., 0., 0.], 2025-09-09T14:56:31.6546484Z [0., 0., 0.], 2025-09-09T14:56:31.6546693Z [0., 0., 0.]], 2025-09-09T14:56:31.6546841Z 2025-09-09T14:56:31.6546926Z [[0., 0., 0.], 2025-09-09T14:56:31.6547133Z [0., 0., 0.], 2025-09-09T14:56:31.6547387Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T14:56:31.6547698Z converted model pt2e: GraphModule( 2025-09-09T14:56:31.6547968Z (conv): Module() 2025-09-09T14:56:31.6548171Z ) 2025-09-09T14:56:31.6548272Z 2025-09-09T14:56:31.6548275Z 2025-09-09T14:56:31.6548279Z 2025-09-09T14:56:31.6548367Z def forward(self, x): 2025-09-09T14:56:31.6548659Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:56:31.6548987Z _scale_0 = self._scale_0 2025-09-09T14:56:31.6549251Z _zero_point_0 = self._zero_point_0 2025-09-09T14:56:31.6549580Z quantize_per_channel_default = self._frozen_param0 2025-09-09T14:56:31.6550674Z dequantize_per_channel_default = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel_default, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel_default = _scale_0 = _zero_point_0 = None 2025-09-09T14:56:31.6552088Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:56:31.6553328Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:56:31.6554655Z conv2d = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_channel_default); dequantize_per_tensor_default = dequantize_per_channel_default = None 2025-09-09T14:56:31.6555481Z relu = torch.ops.aten.relu.default(conv2d); conv2d = None 2025-09-09T14:56:31.6556252Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.006228470243513584, -128, -128, 127, torch.int8); relu = None 2025-09-09T14:56:31.6557534Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.006228470243513584, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:56:31.6558559Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T14:56:31.6558968Z 2025-09-09T14:56:31.6559257Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:56:31.6559636Z onverted model fx: GraphModule( 2025-09-09T14:56:31.6559904Z (conv): ConvReLU2d( 2025-09-09T14:56:31.6560275Z (0): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False) 2025-09-09T14:56:31.6560686Z (1): ReLU() 2025-09-09T14:56:31.6560882Z ) 2025-09-09T14:56:31.6561060Z ) 2025-09-09T14:56:31.6561159Z 2025-09-09T14:56:31.6561163Z 2025-09-09T14:56:31.6561166Z 2025-09-09T14:56:31.6561264Z def forward(self, x): 2025-09-09T14:56:31.6561881Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:56:31.6563116Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:56:31.6564121Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:56:31.6564980Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.006228470243513584, -128, -128, 127, torch.int8); conv = None 2025-09-09T14:56:31.6566309Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.006228470243513584, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:56:34.3241299Z return dequantize_per_tensor_default_1 2025-09-09T14:56:34.3241716Z 2025-09-09T14:56:34.3242073Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:56:34.3242561Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:56:34.3242866Z [0., 0., 0.], 2025-09-09T14:56:34.3243141Z [0., 0., 0.]], 2025-09-09T14:56:34.3243319Z 2025-09-09T14:56:34.3243423Z [[0., 0., 0.], 2025-09-09T14:56:34.3243698Z [0., 0., 0.], 2025-09-09T14:56:34.3243972Z [0., 0., 0.]], 2025-09-09T14:56:34.3244148Z 2025-09-09T14:56:34.3244247Z [[0., 0., 0.], 2025-09-09T14:56:34.3244517Z [0., 0., 0.], 2025-09-09T14:56:34.3244781Z [0., 0., 0.]]]]) 2025-09-09T14:56:34.3245081Z model pt2e: GraphModule( 2025-09-09T14:56:34.3245360Z (conv): Module() 2025-09-09T14:56:34.3245732Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:34.3247180Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:56:34.3248730Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.18965116143226624, max_val=0.18703685700893402) 2025-09-09T14:56:34.3249396Z ) 2025-09-09T14:56:34.3249736Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:34.3250904Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:56:34.3252265Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:56:34.3252889Z ) 2025-09-09T14:56:34.3253225Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:34.3254399Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0062]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:56:34.3255733Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.5892566442489624) 2025-09-09T14:56:34.3256444Z ) 2025-09-09T14:56:34.3256654Z ) 2025-09-09T14:56:34.3256774Z 2025-09-09T14:56:34.3256778Z 2025-09-09T14:56:34.3256783Z 2025-09-09T14:56:34.3256896Z def forward(self, x): 2025-09-09T14:56:34.3257236Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:56:34.3257656Z conv_weight = self.conv.weight 2025-09-09T14:56:34.3258214Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T14:56:34.3258939Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:56:34.3259956Z conv2d = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T14:56:34.3260883Z relu = torch.ops.aten.relu.default(conv2d); conv2d = None 2025-09-09T14:56:34.3269652Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T14:56:34.3270344Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:56:34.3270828Z 2025-09-09T14:56:34.3271182Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:56:34.3271628Z model fx: GraphModule( 2025-09-09T14:56:34.3272021Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:34.3273199Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:56:34.3274608Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:56:34.3275254Z ) 2025-09-09T14:56:34.3275475Z (conv): ConvReLU2d( 2025-09-09T14:56:34.3275800Z 3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False 2025-09-09T14:56:34.3276239Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:34.3277398Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:56:34.3278816Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.18965116143226624, max_val=0.18703685700893402) 2025-09-09T14:56:34.3279463Z ) 2025-09-09T14:56:34.3279705Z ) 2025-09-09T14:56:34.3280091Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:34.3281379Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0062]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:56:34.3282514Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.5892566442489624) 2025-09-09T14:56:34.3282982Z ) 2025-09-09T14:56:34.3283161Z ) 2025-09-09T14:56:34.3283261Z 2025-09-09T14:56:34.3283266Z 2025-09-09T14:56:34.3283270Z 2025-09-09T14:56:34.3283357Z def forward(self, x): 2025-09-09T14:56:34.3283710Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:56:34.3284246Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:56:34.3284788Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:56:34.3285217Z return activation_post_process_1 2025-09-09T14:56:34.3285475Z 2025-09-09T14:56:34.3285767Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:56:34.3286137Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:56:34.3286382Z [0., 0., 0.], 2025-09-09T14:56:34.3286601Z [0., 0., 0.]], 2025-09-09T14:56:34.3286753Z 2025-09-09T14:56:34.3286832Z [[0., 0., 0.], 2025-09-09T14:56:34.3287044Z [0., 0., 0.], 2025-09-09T14:56:34.3287249Z [0., 0., 0.]], 2025-09-09T14:56:34.3287386Z 2025-09-09T14:56:34.3287468Z [[0., 0., 0.], 2025-09-09T14:56:34.3287671Z [0., 0., 0.], 2025-09-09T14:56:34.3287913Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T14:56:34.3288219Z converted model pt2e: GraphModule( 2025-09-09T14:56:34.3288486Z (conv): Module() 2025-09-09T14:56:34.3288681Z ) 2025-09-09T14:56:34.3288787Z 2025-09-09T14:56:34.3288791Z 2025-09-09T14:56:34.3288795Z 2025-09-09T14:56:34.3288880Z def forward(self, x): 2025-09-09T14:56:34.3289167Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:56:34.3289535Z quantize_per_tensor_default = self._frozen_param0 2025-09-09T14:56:34.3290447Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.0014933162601664662, 0, -127, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:56:34.3291690Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:56:34.3292928Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:56:34.3294258Z conv2d = torch.ops.aten.conv2d.default(dequantize_per_tensor_default_1, dequantize_per_tensor_default); dequantize_per_tensor_default_1 = dequantize_per_tensor_default = None 2025-09-09T14:56:34.3295070Z relu = torch.ops.aten.relu.default(conv2d); conv2d = None 2025-09-09T14:56:34.3295897Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.006232379004359245, -128, -128, 127, torch.int8); relu = None 2025-09-09T14:56:34.3297233Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.006232379004359245, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:56:34.3298533Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:56:34.3298943Z 2025-09-09T14:56:34.3299218Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:56:34.3299593Z onverted model fx: GraphModule( 2025-09-09T14:56:34.3299854Z (conv): ConvReLU2d( 2025-09-09T14:56:34.3300215Z (0): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False) 2025-09-09T14:56:34.3300622Z (1): ReLU() 2025-09-09T14:56:34.3300812Z ) 2025-09-09T14:56:34.3300989Z ) 2025-09-09T14:56:34.3301084Z 2025-09-09T14:56:34.3301231Z 2025-09-09T14:56:34.3301235Z 2025-09-09T14:56:34.3301323Z def forward(self, x): 2025-09-09T14:56:34.3301937Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:56:34.3303271Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:56:34.3304267Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:56:34.3305120Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.006232379004359245, -128, -128, 127, torch.int8); conv = None 2025-09-09T14:56:34.3306404Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.006232379004359245, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:56:34.3307325Z return dequantize_per_tensor_default_1 2025-09-09T14:56:34.3307612Z 2025-09-09T14:56:34.3307886Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:56:34.3308255Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:56:34.3308487Z [0., 0., 0.], 2025-09-09T14:56:34.3308699Z [0., 0., 0.]], 2025-09-09T14:56:34.3308837Z 2025-09-09T14:56:34.3308912Z [[0., 0., 0.], 2025-09-09T14:56:34.3309123Z [0., 0., 0.], 2025-09-09T14:56:34.3309335Z [0., 0., 0.]], 2025-09-09T14:56:34.3309471Z 2025-09-09T14:56:34.3309547Z [[0., 0., 0.], 2025-09-09T14:56:34.3309753Z [0., 0., 0.], 2025-09-09T14:56:34.3309955Z [0., 0., 0.]]]]) 2025-09-09T14:56:34.3310194Z model pt2e: GraphModule( 2025-09-09T14:56:34.3310419Z (conv): Module() 2025-09-09T14:56:34.3310721Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:34.3311707Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0014, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:56:34.3312996Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1913, -0.1469, -0.1921]), max_val=tensor([0.1740, 0.1746, 0.1810])) 2025-09-09T14:56:34.3313641Z ) 2025-09-09T14:56:34.3313912Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:34.3314839Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:56:35.7633214Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:56:35.7633907Z ) 2025-09-09T14:56:35.7634268Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:35.7635440Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0080]), zero_point=tensor([-4], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:56:35.7636868Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.9872972965240479, max_val=1.0470484495162964) 2025-09-09T14:56:35.7637567Z ) 2025-09-09T14:56:35.7637780Z ) 2025-09-09T14:56:35.7637903Z 2025-09-09T14:56:35.7637915Z 2025-09-09T14:56:35.7637920Z 2025-09-09T14:56:35.7638025Z def forward(self, x): 2025-09-09T14:56:35.7638368Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:56:35.7638797Z conv_weight = self.conv.weight 2025-09-09T14:56:35.7639377Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T14:56:35.7640301Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:56:35.7641316Z conv2d = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T14:56:35.7642477Z activation_post_process_2 = self.activation_post_process_2(conv2d); conv2d = None 2025-09-09T14:56:35.7643167Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:56:35.7643640Z 2025-09-09T14:56:35.7643986Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:56:35.7644438Z model fx: GraphModule( 2025-09-09T14:56:35.7644822Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:35.7645987Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:56:35.7647350Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:56:35.7647988Z ) 2025-09-09T14:56:35.7648206Z (conv): Conv2d( 2025-09-09T14:56:35.7648501Z 3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False 2025-09-09T14:56:35.7648952Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:35.7650133Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0014, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:56:35.7651748Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1913, -0.1469, -0.1921]), max_val=tensor([0.1740, 0.1746, 0.1810])) 2025-09-09T14:56:35.7652551Z ) 2025-09-09T14:56:35.7652758Z ) 2025-09-09T14:56:35.7653098Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:35.7654263Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0080]), zero_point=tensor([-4], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:56:35.7655639Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.9872972965240479, max_val=1.0470484495162964) 2025-09-09T14:56:35.7656353Z ) 2025-09-09T14:56:35.7656565Z ) 2025-09-09T14:56:35.7656682Z 2025-09-09T14:56:35.7656687Z 2025-09-09T14:56:35.7656691Z 2025-09-09T14:56:35.7656809Z def forward(self, x): 2025-09-09T14:56:35.7657280Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:56:35.7657938Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:56:35.7658608Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:56:35.7659135Z return activation_post_process_1 2025-09-09T14:56:35.7659454Z 2025-09-09T14:56:35.7659793Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:56:35.7660246Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:56:35.7660541Z [0., 0., 0.], 2025-09-09T14:56:35.7660807Z [0., 0., 0.]], 2025-09-09T14:56:35.7660981Z 2025-09-09T14:56:35.7661076Z [[0., 0., 0.], 2025-09-09T14:56:35.7661332Z [0., 0., 0.], 2025-09-09T14:56:35.7661581Z [0., 0., 0.]], 2025-09-09T14:56:35.7661764Z 2025-09-09T14:56:35.7661857Z [[0., 0., 0.], 2025-09-09T14:56:35.7662106Z [0., 0., 0.], 2025-09-09T14:56:35.7662396Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T14:56:35.7662776Z converted model pt2e: GraphModule( 2025-09-09T14:56:35.7663093Z (conv): Module() 2025-09-09T14:56:35.7663334Z ) 2025-09-09T14:56:35.7663452Z 2025-09-09T14:56:35.7663457Z 2025-09-09T14:56:35.7663461Z 2025-09-09T14:56:35.7663564Z def forward(self, x): 2025-09-09T14:56:35.7664000Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:56:35.7664404Z _scale_0 = self._scale_0 2025-09-09T14:56:35.7664718Z _zero_point_0 = self._zero_point_0 2025-09-09T14:56:35.7665196Z quantize_per_channel_default = self._frozen_param0 2025-09-09T14:56:35.7666448Z dequantize_per_channel_default = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel_default, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel_default = _scale_0 = _zero_point_0 = None 2025-09-09T14:56:35.7668170Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:56:35.7669720Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:56:35.7671398Z conv2d = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_channel_default); dequantize_per_tensor_default = dequantize_per_channel_default = None 2025-09-09T14:56:35.7672882Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d, 0.007977825589478016, -4, -128, 127, torch.int8); conv2d = None 2025-09-09T14:56:35.7674506Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.007977825589478016, -4, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:56:35.7675862Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T14:56:35.7676274Z 2025-09-09T14:56:35.7676551Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:56:35.7676925Z onverted model fx: GraphModule( 2025-09-09T14:56:35.7677339Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False) 2025-09-09T14:56:35.7677754Z ) 2025-09-09T14:56:35.7677854Z 2025-09-09T14:56:35.7677858Z 2025-09-09T14:56:35.7677862Z 2025-09-09T14:56:35.7677946Z def forward(self, x): 2025-09-09T14:56:35.7678566Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:56:35.7679789Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:56:35.7680783Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:56:35.7681633Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.007977825589478016, -4, -128, 127, torch.int8); conv = None 2025-09-09T14:56:35.7682895Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.007977825589478016, -4, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:56:35.7683766Z return dequantize_per_tensor_default_1 2025-09-09T14:56:35.7684048Z 2025-09-09T14:56:35.7684324Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:56:35.7684697Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:56:35.7684935Z [0., 0., 0.], 2025-09-09T14:56:35.7685151Z [0., 0., 0.]], 2025-09-09T14:56:35.7685291Z 2025-09-09T14:56:35.7685372Z [[0., 0., 0.], 2025-09-09T14:56:35.7685581Z [0., 0., 0.], 2025-09-09T14:56:35.7685797Z [0., 0., 0.]], 2025-09-09T14:56:35.7685937Z 2025-09-09T14:56:35.7686016Z [[0., 0., 0.], 2025-09-09T14:56:35.7686227Z [0., 0., 0.], 2025-09-09T14:56:35.7686436Z [0., 0., 0.]]]]) 2025-09-09T14:56:35.7686675Z model pt2e: GraphModule( 2025-09-09T14:56:35.7686907Z (conv): Module() 2025-09-09T14:56:35.7687363Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:35.7688309Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:56:35.7689490Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.19212539494037628, max_val=0.18097467720508575) 2025-09-09T14:56:35.7690009Z ) 2025-09-09T14:56:35.7690290Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:35.7691215Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:56:35.7692293Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:56:35.7692794Z ) 2025-09-09T14:56:35.7693080Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:56:35.7693999Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0079]), zero_point=tensor([-5], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:56:35.7695094Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.9800506234169006, max_val=1.0470484495162964) 2025-09-09T14:56:35.7695605Z ) 2025-09-09T14:56:35.7695778Z ) 2025-09-09T14:56:35.7695921Z 2025-09-09T14:56:35.7695925Z 2025-09-09T14:56:35.7695929Z 2025-09-09T14:56:35.7696019Z def forward(self, x): 2025-09-09T14:56:35.7696310Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:56:35.7696643Z conv_weight = self.conv.weight 2025-09-09T14:56:35.7697132Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T14:56:35.7697904Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:56:35.7698707Z conv2d = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T14:56:35.7699531Z activation_post_process_2 = self.activation_post_process_2(conv2d); conv2d = None 2025-09-09T14:56:35.7700086Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T14:56:35.7700472Z 2025-09-09T14:57:50.8566730Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:57:50.8567129Z model fx: GraphModule( 2025-09-09T14:57:50.8567531Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:57:50.8568474Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:57:50.8569652Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:57:50.8570178Z ) 2025-09-09T14:57:50.8570368Z (conv): Conv2d( 2025-09-09T14:57:50.8570618Z 3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False 2025-09-09T14:57:50.8570987Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:57:50.8571953Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:57:50.8573074Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.19212539494037628, max_val=0.18097467720508575) 2025-09-09T14:57:50.8573600Z ) 2025-09-09T14:57:50.8573776Z ) 2025-09-09T14:57:50.8574061Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:57:50.8575304Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0079]), zero_point=tensor([-5], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:57:50.8576662Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.9800506234169006, max_val=1.0470484495162964) 2025-09-09T14:57:50.8577171Z ) 2025-09-09T14:57:50.8577353Z ) 2025-09-09T14:57:50.8577451Z 2025-09-09T14:57:50.8577455Z 2025-09-09T14:57:50.8577459Z 2025-09-09T14:57:50.8577559Z def forward(self, x): 2025-09-09T14:57:50.8577915Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:57:50.8578455Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:57:50.8579005Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:57:50.8579439Z return activation_post_process_1 2025-09-09T14:57:50.8579703Z 2025-09-09T14:57:50.8579986Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:57:50.8580354Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:57:50.8580591Z [0., 0., 0.], 2025-09-09T14:57:50.8580815Z [0., 0., 0.]], 2025-09-09T14:57:50.8580955Z 2025-09-09T14:57:50.8581033Z [[0., 0., 0.], 2025-09-09T14:57:50.8581243Z [0., 0., 0.], 2025-09-09T14:57:50.8581488Z [0., 0., 0.]], 2025-09-09T14:57:50.8581650Z 2025-09-09T14:57:50.8581725Z [[0., 0., 0.], 2025-09-09T14:57:50.8581935Z [0., 0., 0.], 2025-09-09T14:57:50.8582176Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T14:57:50.8582492Z converted model pt2e: GraphModule( 2025-09-09T14:57:50.8582753Z (conv): Module() 2025-09-09T14:57:50.8582953Z ) 2025-09-09T14:57:50.8583053Z 2025-09-09T14:57:50.8583057Z 2025-09-09T14:57:50.8583061Z 2025-09-09T14:57:50.8583145Z def forward(self, x): 2025-09-09T14:57:50.8583432Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:57:50.8583810Z quantize_per_tensor_default = self._frozen_param0 2025-09-09T14:57:50.8584772Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.0015127983642742038, 0, -127, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:57:50.8586031Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:57:50.8587272Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:57:50.8588601Z conv2d = torch.ops.aten.conv2d.default(dequantize_per_tensor_default_1, dequantize_per_tensor_default); dequantize_per_tensor_default_1 = dequantize_per_tensor_default = None 2025-09-09T14:57:50.8589784Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d, 0.007949408143758774, -5, -128, 127, torch.int8); conv2d = None 2025-09-09T14:57:50.8591102Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.007949408143758774, -5, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:57:50.8592111Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:57:50.8592526Z 2025-09-09T14:57:50.8592804Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:57:50.8593184Z onverted model fx: GraphModule( 2025-09-09T14:57:50.8593594Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False) 2025-09-09T14:57:50.8594009Z ) 2025-09-09T14:57:50.8594109Z 2025-09-09T14:57:50.8594113Z 2025-09-09T14:57:50.8594117Z 2025-09-09T14:57:50.8594204Z def forward(self, x): 2025-09-09T14:57:50.8594911Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:57:50.8596144Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:57:50.8597227Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:57:50.8598256Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.007949408143758774, -5, -128, 127, torch.int8); conv = None 2025-09-09T14:57:50.8599520Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.007949408143758774, -5, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:57:50.8600438Z return dequantize_per_tensor_default_1 2025-09-09T14:57:50.8600791Z 2025-09-09T14:57:50.8601136Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:57:50.8601613Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:57:50.8601920Z [0., 0., 0.], 2025-09-09T14:57:50.8602186Z [0., 0., 0.]], 2025-09-09T14:57:50.8602363Z 2025-09-09T14:57:50.8602466Z [[0., 0., 0.], 2025-09-09T14:57:50.8602727Z [0., 0., 0.], 2025-09-09T14:57:50.8602990Z [0., 0., 0.]], 2025-09-09T14:57:50.8603150Z 2025-09-09T14:57:50.8603227Z [[0., 0., 0.], 2025-09-09T14:57:50.8603441Z [0., 0., 0.], 2025-09-09T14:57:50.8603649Z [0., 0., 0.]]]]) 2025-09-09T14:57:50.8604085Z PASSED 2025-09-09T14:57:50.8604749Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_transpose_bn PASSED 2025-09-09T14:57:50.8605793Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_transpose_bn_relu PASSED 2025-09-09T14:57:50.8606751Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_inplace_add_relu model pt2e: GraphModule( 2025-09-09T14:57:50.8607367Z (conv): Module() 2025-09-09T14:57:50.8607678Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:57:50.8608639Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0011]), zero_point=tensor([127], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:57:50.8609820Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2877]), max_val=tensor([-0.2877])) 2025-09-09T14:57:50.8610408Z ) 2025-09-09T14:57:50.8610753Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:57:50.8611941Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0126]), zero_point=tensor([7], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:57:50.8613115Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.7008640766143799, max_val=1.5035617351531982) 2025-09-09T14:57:50.8613629Z ) 2025-09-09T14:57:50.8613913Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:57:50.8614839Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0036]), zero_point=tensor([43], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:57:50.8615987Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.6198297739028931, max_val=0.30200809240341187) 2025-09-09T14:57:50.8616509Z ) 2025-09-09T14:57:50.8616786Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:57:50.8617862Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0035]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:57:50.8618919Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.8897914886474609) 2025-09-09T14:57:50.8619502Z ) 2025-09-09T14:57:50.8619668Z ) 2025-09-09T14:57:50.8619772Z 2025-09-09T14:57:50.8619776Z 2025-09-09T14:57:50.8619780Z 2025-09-09T14:57:50.8619864Z def forward(self, x): 2025-09-09T14:57:50.8620160Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:57:50.8620497Z conv_weight = self.conv.weight 2025-09-09T14:57:50.8620961Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T14:57:50.8621431Z conv_bias = self.conv.bias 2025-09-09T14:57:50.8621813Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:57:50.8622602Z conv2d = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, conv_bias); activation_post_process_1 = conv_bias = None 2025-09-09T14:57:50.8623417Z activation_post_process_2 = self.activation_post_process_2(conv2d); conv2d = None 2025-09-09T14:57:50.8624227Z add_ = torch.ops.aten.add_.Tensor(activation_post_process_2, activation_post_process_0); activation_post_process_2 = activation_post_process_0 = None 2025-09-09T14:57:50.8624937Z relu_ = torch.ops.aten.relu_.default(add_); add_ = None 2025-09-09T14:57:50.8625425Z activation_post_process_3 = self.activation_post_process_3(relu_); relu_ = None 2025-09-09T14:57:50.8625977Z return pytree.tree_unflatten((activation_post_process_3,), self._out_spec) 2025-09-09T14:57:50.8626374Z 2025-09-09T14:57:50.8626666Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:57:50.8627029Z model fx: GraphModule( 2025-09-09T14:57:50.8627361Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:57:52.3585604Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0126]), zero_point=tensor([7], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:57:52.3587097Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.7008640766143799, max_val=1.5035617351531982) 2025-09-09T14:57:52.3587622Z ) 2025-09-09T14:57:52.3587810Z (conv): Conv2d( 2025-09-09T14:57:52.3588048Z 1, 1, kernel_size=(1, 1), stride=(1, 1) 2025-09-09T14:57:52.3588404Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:57:52.3589339Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0011]), zero_point=tensor([127], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:57:52.3590534Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2877]), max_val=tensor([-0.2877])) 2025-09-09T14:57:52.3591117Z ) 2025-09-09T14:57:52.3591302Z ) 2025-09-09T14:57:52.3591599Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:57:52.3592544Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0036]), zero_point=tensor([43], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:57:52.3593667Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.6198297739028931, max_val=0.30200809240341187) 2025-09-09T14:57:52.3594186Z ) 2025-09-09T14:57:52.3594392Z (relu): ReLU(inplace=True) 2025-09-09T14:57:52.3594748Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:57:52.3595691Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0035]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:57:52.3596997Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.8897914886474609) 2025-09-09T14:57:52.3597640Z ) 2025-09-09T14:57:52.3597824Z ) 2025-09-09T14:57:52.3598056Z 2025-09-09T14:57:52.3598060Z 2025-09-09T14:57:52.3598064Z 2025-09-09T14:57:52.3598160Z def forward(self, x): 2025-09-09T14:57:52.3598518Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:57:52.3598960Z conv = self.conv(activation_post_process_0) 2025-09-09T14:57:52.3599405Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:57:52.3600107Z add = activation_post_process_1 + activation_post_process_0; activation_post_process_1 = activation_post_process_0 = None 2025-09-09T14:57:52.3600676Z relu = self.relu(add); add = None 2025-09-09T14:57:52.3601103Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T14:57:52.3601552Z return activation_post_process_2 2025-09-09T14:57:52.3601824Z 2025-09-09T14:57:52.3602129Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:57:52.3602507Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:57:52.3602767Z [0., 0., 0.], 2025-09-09T14:57:52.3603018Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T14:57:52.3603340Z converted model pt2e: GraphModule( 2025-09-09T14:57:52.3603609Z (conv): Module() 2025-09-09T14:57:52.3603822Z ) 2025-09-09T14:57:52.3603927Z 2025-09-09T14:57:52.3603931Z 2025-09-09T14:57:52.3603935Z 2025-09-09T14:57:52.3604040Z def forward(self, x): 2025-09-09T14:57:52.3604330Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:57:52.3604676Z _scale_0 = self._scale_0 2025-09-09T14:57:52.3604939Z _zero_point_0 = self._zero_point_0 2025-09-09T14:57:52.3605279Z quantize_per_channel_default = self._frozen_param0 2025-09-09T14:57:52.3606291Z dequantize_per_channel_default = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel_default, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel_default = _scale_0 = _zero_point_0 = None 2025-09-09T14:57:52.3607258Z conv_bias = self.conv.bias 2025-09-09T14:57:52.3607898Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01256637554615736, 7, -128, 127, torch.int8); x = None 2025-09-09T14:57:52.3609010Z dequantize_per_tensor_default_4 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01256637554615736, 7, -128, 127, torch.int8) 2025-09-09T14:57:52.3610322Z dequantize_per_tensor_default_3 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01256637554615736, 7, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:57:52.3611759Z conv2d = torch.ops.aten.conv2d.default(dequantize_per_tensor_default_3, dequantize_per_channel_default, conv_bias); dequantize_per_tensor_default_3 = dequantize_per_channel_default = conv_bias = None 2025-09-09T14:57:52.3613027Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d, 0.003615050343796611, 43, -128, 127, torch.int8); conv2d = None 2025-09-09T14:57:52.3614328Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.003615050343796611, 43, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:57:52.3623203Z add_ = torch.ops.aten.add_.Tensor(dequantize_per_tensor_default_1, dequantize_per_tensor_default_4); dequantize_per_tensor_default_1 = dequantize_per_tensor_default_4 = None 2025-09-09T14:57:52.3624048Z relu_ = torch.ops.aten.relu_.default(add_); add_ = None 2025-09-09T14:57:52.3624827Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu_, 0.003489378374069929, -128, -128, 127, torch.int8); relu_ = None 2025-09-09T14:57:52.3626272Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.003489378374069929, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:57:52.3627386Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:57:52.3627808Z 2025-09-09T14:57:52.3628097Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:57:52.3628487Z onverted model fx: GraphModule( 2025-09-09T14:57:52.3628875Z (conv): QuantizedConv2d(Reference)(1, 1, kernel_size=(1, 1), stride=(1, 1)) 2025-09-09T14:57:52.3629283Z (relu): ReLU(inplace=True) 2025-09-09T14:57:52.3629519Z ) 2025-09-09T14:57:52.3629624Z 2025-09-09T14:57:52.3629628Z 2025-09-09T14:57:52.3629632Z 2025-09-09T14:57:52.3629719Z def forward(self, x): 2025-09-09T14:57:52.3630333Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01256637554615736, 7, -128, 127, torch.int8); x = None 2025-09-09T14:57:52.3631572Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01256637554615736, 7, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:57:52.3632522Z conv = self.conv(dequantize_per_tensor_default) 2025-09-09T14:57:52.3633261Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.003615050343796611, 43, -128, 127, torch.int8); conv = None 2025-09-09T14:57:52.3634523Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.003615050343796611, 43, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:57:52.3635728Z add = dequantize_per_tensor_default_1 + dequantize_per_tensor_default; dequantize_per_tensor_default_1 = dequantize_per_tensor_default = None 2025-09-09T14:57:52.3636357Z relu = self.relu(add); add = None 2025-09-09T14:57:52.3637059Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.003489378374069929, -128, -128, 127, torch.int8); relu = None 2025-09-09T14:57:52.3638350Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.003489378374069929, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:57:52.3639247Z return dequantize_per_tensor_default_2 2025-09-09T14:57:52.3639532Z 2025-09-09T14:57:52.3639812Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:57:52.3640190Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:57:52.3640442Z [0., 0., 0.], 2025-09-09T14:57:52.3640654Z [0., 0., 0.]]]]) 2025-09-09T14:57:52.3640895Z model pt2e: GraphModule( 2025-09-09T14:57:52.3641128Z (conv): Module() 2025-09-09T14:57:52.3641441Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:57:52.3642498Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0011]), zero_point=tensor([127], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:57:52.3643636Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.28767645359039307, max_val=-0.28767645359039307) 2025-09-09T14:57:52.3644156Z ) 2025-09-09T14:57:52.3644431Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:57:52.3645363Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0126]), zero_point=tensor([7], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:57:52.3646462Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.7008640766143799, max_val=1.5035617351531982) 2025-09-09T14:57:52.3646976Z ) 2025-09-09T14:57:52.3647365Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:57:52.3648306Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0036]), zero_point=tensor([43], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:57:52.3649481Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.6198297739028931, max_val=0.30200809240341187) 2025-09-09T14:57:52.3649988Z ) 2025-09-09T14:57:52.3650269Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:58:24.8257127Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0035]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:58:24.8258475Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.8897914886474609) 2025-09-09T14:58:24.8258958Z ) 2025-09-09T14:58:24.8259141Z ) 2025-09-09T14:58:24.8259259Z 2025-09-09T14:58:24.8259265Z 2025-09-09T14:58:24.8259268Z 2025-09-09T14:58:24.8259388Z def forward(self, x): 2025-09-09T14:58:24.8259746Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:58:24.8260105Z conv_weight = self.conv.weight 2025-09-09T14:58:24.8260571Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T14:58:24.8261051Z conv_bias = self.conv.bias 2025-09-09T14:58:24.8261428Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:58:24.8262219Z conv2d = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, conv_bias); activation_post_process_1 = conv_bias = None 2025-09-09T14:58:24.8263076Z activation_post_process_2 = self.activation_post_process_2(conv2d); conv2d = None 2025-09-09T14:58:24.8263872Z add_ = torch.ops.aten.add_.Tensor(activation_post_process_2, activation_post_process_0); activation_post_process_2 = activation_post_process_0 = None 2025-09-09T14:58:24.8264580Z relu_ = torch.ops.aten.relu_.default(add_); add_ = None 2025-09-09T14:58:24.8265050Z activation_post_process_3 = self.activation_post_process_3(relu_); relu_ = None 2025-09-09T14:58:24.8265605Z return pytree.tree_unflatten((activation_post_process_3,), self._out_spec) 2025-09-09T14:58:24.8265992Z 2025-09-09T14:58:24.8266276Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:58:24.8266642Z model fx: GraphModule( 2025-09-09T14:58:24.8266965Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:58:24.8267917Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0126]), zero_point=tensor([7], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:58:24.8269012Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.7008640766143799, max_val=1.5035617351531982) 2025-09-09T14:58:24.8269535Z ) 2025-09-09T14:58:24.8269723Z (conv): Conv2d( 2025-09-09T14:58:24.8269952Z 1, 1, kernel_size=(1, 1), stride=(1, 1) 2025-09-09T14:58:24.8270311Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:58:24.8271229Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0011]), zero_point=tensor([127], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:58:24.8272412Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.28767645359039307, max_val=-0.28767645359039307) 2025-09-09T14:58:24.8272936Z ) 2025-09-09T14:58:24.8273108Z ) 2025-09-09T14:58:24.8273390Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:58:24.8274538Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0036]), zero_point=tensor([43], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:58:24.8275642Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.6198297739028931, max_val=0.30200809240341187) 2025-09-09T14:58:24.8276292Z ) 2025-09-09T14:58:24.8276489Z (relu): ReLU(inplace=True) 2025-09-09T14:58:24.8276844Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:58:24.8277790Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0035]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:58:24.8278871Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.8897914886474609) 2025-09-09T14:58:24.8279336Z ) 2025-09-09T14:58:24.8279511Z ) 2025-09-09T14:58:24.8279611Z 2025-09-09T14:58:24.8279615Z 2025-09-09T14:58:24.8279620Z 2025-09-09T14:58:24.8279715Z def forward(self, x): 2025-09-09T14:58:24.8280067Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:58:24.8280502Z conv = self.conv(activation_post_process_0) 2025-09-09T14:58:24.8280940Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:58:24.8281639Z add = activation_post_process_1 + activation_post_process_0; activation_post_process_1 = activation_post_process_0 = None 2025-09-09T14:58:24.8282240Z relu = self.relu(add); add = None 2025-09-09T14:58:24.8282659Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T14:58:24.8283087Z return activation_post_process_2 2025-09-09T14:58:24.8283343Z 2025-09-09T14:58:24.8283622Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:58:24.8283985Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:58:24.8284227Z [0., 0., 0.], 2025-09-09T14:58:24.8284466Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T14:58:24.8284782Z converted model pt2e: GraphModule( 2025-09-09T14:58:24.8285044Z (conv): Module() 2025-09-09T14:58:24.8285242Z ) 2025-09-09T14:58:24.8285340Z 2025-09-09T14:58:24.8285349Z 2025-09-09T14:58:24.8285353Z 2025-09-09T14:58:24.8285444Z def forward(self, x): 2025-09-09T14:58:24.8285722Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:58:24.8286100Z quantize_per_tensor_default = self._frozen_param0 2025-09-09T14:58:24.8287014Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.002265168819576502, 0, -127, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:58:24.8287868Z conv_bias = self.conv.bias 2025-09-09T14:58:24.8288506Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01256637554615736, 7, -128, 127, torch.int8); x = None 2025-09-09T14:58:24.8289634Z dequantize_per_tensor_default_5 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.01256637554615736, 7, -128, 127, torch.int8) 2025-09-09T14:58:24.8290947Z dequantize_per_tensor_default_4 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.01256637554615736, 7, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:58:24.8292376Z conv2d = torch.ops.aten.conv2d.default(dequantize_per_tensor_default_4, dequantize_per_tensor_default, conv_bias); dequantize_per_tensor_default_4 = dequantize_per_tensor_default = conv_bias = None 2025-09-09T14:58:24.8293638Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d, 0.003615050343796611, 43, -128, 127, torch.int8); conv2d = None 2025-09-09T14:58:24.8294916Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.003615050343796611, 43, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:58:24.8296391Z add_ = torch.ops.aten.add_.Tensor(dequantize_per_tensor_default_2, dequantize_per_tensor_default_5); dequantize_per_tensor_default_2 = dequantize_per_tensor_default_5 = None 2025-09-09T14:58:24.8297262Z relu_ = torch.ops.aten.relu_.default(add_); add_ = None 2025-09-09T14:58:24.8298717Z quantize_per_tensor_default_3 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu_, 0.003489378374069929, -128, -128, 127, torch.int8); relu_ = None 2025-09-09T14:58:24.8300004Z dequantize_per_tensor_default_3 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_3, 0.003489378374069929, -128, -128, 127, torch.int8); quantize_per_tensor_default_3 = None 2025-09-09T14:58:24.8301024Z return pytree.tree_unflatten((dequantize_per_tensor_default_3,), self._out_spec) 2025-09-09T14:58:24.8301439Z 2025-09-09T14:58:24.8301754Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:58:24.8302164Z onverted model fx: GraphModule( 2025-09-09T14:58:24.8302541Z (conv): QuantizedConv2d(Reference)(1, 1, kernel_size=(1, 1), stride=(1, 1)) 2025-09-09T14:58:24.8302947Z (relu): ReLU(inplace=True) 2025-09-09T14:58:24.8303189Z ) 2025-09-09T14:58:24.8303286Z 2025-09-09T14:58:24.8303290Z 2025-09-09T14:58:24.8303295Z 2025-09-09T14:58:24.8303382Z def forward(self, x): 2025-09-09T14:58:24.8303996Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01256637554615736, 7, -128, 127, torch.int8); x = None 2025-09-09T14:58:24.8305211Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01256637554615736, 7, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:58:24.8306092Z conv = self.conv(dequantize_per_tensor_default) 2025-09-09T14:58:24.8306821Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.003615050343796611, 43, -128, 127, torch.int8); conv = None 2025-09-09T14:58:24.8308078Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.003615050343796611, 43, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:58:24.8309285Z add = dequantize_per_tensor_default_1 + dequantize_per_tensor_default; dequantize_per_tensor_default_1 = dequantize_per_tensor_default = None 2025-09-09T14:58:24.8309918Z relu = self.relu(add); add = None 2025-09-09T14:58:24.8310607Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.003489378374069929, -128, -128, 127, torch.int8); relu = None 2025-09-09T14:58:24.8311890Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.003489378374069929, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:58:24.8312834Z return dequantize_per_tensor_default_2 2025-09-09T14:58:24.8313117Z 2025-09-09T14:58:24.8313400Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:58:24.8313769Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:58:24.8314013Z [0., 0., 0.], 2025-09-09T14:58:24.8314221Z [0., 0., 0.]]]]) 2025-09-09T14:58:24.8314652Z PASSED 2025-09-09T14:58:24.8315357Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_per_channel_weight_custom_dtype PASSED 2025-09-09T14:58:45.6371435Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_preserve_source_fn_stack PASSED 2025-09-09T14:58:45.6373644Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_update_shared_qspec model pt2e: GraphModule( 2025-09-09T14:58:45.6374469Z (conv): Module() 2025-09-09T14:58:45.6374728Z (bn): Module() 2025-09-09T14:58:45.6375093Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:58:45.6376633Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:58:45.6378210Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:58:45.6378851Z ) 2025-09-09T14:58:45.6379197Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:58:45.6380430Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0014]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:58:45.6382068Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1640, -0.1903, -0.1739]), max_val=tensor([0.1851, 0.1825, 0.1577])) 2025-09-09T14:58:45.6382895Z ) 2025-09-09T14:58:45.6383233Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:58:45.6384402Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0163]), zero_point=tensor([6], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:58:45.6385786Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1757256984710693, max_val=1.9743094444274902) 2025-09-09T14:58:45.6386438Z ) 2025-09-09T14:58:45.6386776Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:58:45.6387938Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0163]), zero_point=tensor([6], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:58:45.6389369Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1757256984710693, max_val=1.9743094444274902) 2025-09-09T14:58:45.6390009Z ) 2025-09-09T14:58:45.6390222Z ) 2025-09-09T14:58:45.6390342Z 2025-09-09T14:58:45.6390347Z 2025-09-09T14:58:45.6390358Z 2025-09-09T14:58:45.6390471Z def forward(self, x): 2025-09-09T14:58:45.6390814Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:58:45.6391234Z conv_weight = self.conv.weight 2025-09-09T14:58:45.6391565Z conv_bias = self.conv.bias 2025-09-09T14:58:45.6391882Z bn_weight = self.bn.weight 2025-09-09T14:58:45.6392187Z bn_bias = self.bn.bias 2025-09-09T14:58:45.6392507Z bn_running_mean = self.bn.running_mean 2025-09-09T14:58:45.6392890Z bn_running_var = self.bn.running_var 2025-09-09T14:58:45.6393306Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:58:45.6393850Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:58:45.6394570Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:58:45.6395225Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:58:45.6395696Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:58:45.6396209Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:58:45.6396748Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T14:58:45.6397610Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:58:45.6398341Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:58:45.6399132Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:58:45.6400354Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:58:45.6401599Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T14:58:45.6402277Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T14:58:45.6403137Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T14:58:45.6403825Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:58:45.6404907Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:58:45.6406064Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T14:58:45.6406988Z hardtanh = torch.ops.aten.hardtanh.default(activation_post_process_2, -1.0, 1.0); activation_post_process_2 = None 2025-09-09T14:58:45.6407884Z activation_post_process_3 = self.activation_post_process_3(hardtanh); hardtanh = None 2025-09-09T14:58:45.6408598Z return pytree.tree_unflatten((activation_post_process_3,), self._out_spec) 2025-09-09T14:58:45.6409087Z 2025-09-09T14:58:45.6409438Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:58:45.6409896Z model fx: GraphModule( 2025-09-09T14:58:45.6410292Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:58:45.6411461Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:58:45.6412833Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:58:45.6413475Z ) 2025-09-09T14:58:45.6413692Z (conv): ConvBn2d( 2025-09-09T14:58:45.6413979Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T14:58:45.6414487Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:58:45.6415066Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:58:45.6416323Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0014]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T14:58:45.6417970Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1640, -0.1903, -0.1739]), max_val=tensor([0.1851, 0.1825, 0.1577])) 2025-09-09T14:58:45.6418838Z ) 2025-09-09T14:58:45.6419055Z ) 2025-09-09T14:58:45.6419396Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:58:45.6420571Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0163]), zero_point=tensor([6], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:58:45.6421959Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1757256984710693, max_val=1.9743094444274902) 2025-09-09T14:58:45.6422618Z ) 2025-09-09T14:58:45.6422884Z (hardtanh): Hardtanh(min_val=-1.0, max_val=1.0) 2025-09-09T14:58:45.6423370Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:58:45.6424532Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0163]), zero_point=tensor([6], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:58:45.6425948Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1757256984710693, max_val=1.9743094444274902) 2025-09-09T14:58:45.6426653Z ) 2025-09-09T14:58:45.6426864Z ) 2025-09-09T14:58:45.6426988Z 2025-09-09T14:58:45.6426995Z 2025-09-09T14:58:45.6427000Z 2025-09-09T14:58:45.6427102Z def forward(self, x): 2025-09-09T14:58:45.6427554Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:58:45.6428096Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:58:45.6428750Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:58:45.6429363Z hardtanh = self.hardtanh(activation_post_process_1); activation_post_process_1 = None 2025-09-09T14:58:45.6429974Z activation_post_process_2 = self.activation_post_process_2(hardtanh); hardtanh = None 2025-09-09T14:58:45.6430431Z return activation_post_process_2 2025-09-09T14:58:45.6430699Z 2025-09-09T14:58:45.6430981Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:58:45.6431364Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:58:45.6431606Z [0., 0., 0.], 2025-09-09T14:58:45.6431829Z [0., 0., 0.]], 2025-09-09T14:58:45.6431971Z 2025-09-09T14:58:45.6432051Z [[0., 0., 0.], 2025-09-09T14:58:45.6432269Z [0., 0., 0.], 2025-09-09T14:58:45.6432495Z [0., 0., 0.]], 2025-09-09T14:58:45.6432636Z 2025-09-09T14:58:45.6432714Z [[0., 0., 0.], 2025-09-09T14:58:45.6432930Z [0., 0., 0.], 2025-09-09T14:58:45.6433182Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T14:58:45.6433500Z converted model pt2e: GraphModule( 2025-09-09T14:58:45.6433765Z (conv): Module() 2025-09-09T14:58:45.6433974Z (bn): Module() 2025-09-09T14:58:45.6434168Z ) 2025-09-09T14:58:45.6434276Z 2025-09-09T14:58:45.6434280Z 2025-09-09T14:58:45.6434283Z 2025-09-09T14:58:45.6434371Z def forward(self, x): 2025-09-09T14:58:45.6434657Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:58:45.6434991Z conv_bias = self.conv.bias 2025-09-09T14:58:45.6435637Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:58:45.6436888Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:58:45.6437748Z _scale_0 = self._scale_0 2025-09-09T14:58:45.6438017Z _zero_point_0 = self._zero_point_0 2025-09-09T14:58:45.6438321Z quantize_per_channel = self._frozen_param0 2025-09-09T14:58:45.6439213Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T14:58:45.6440561Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T14:58:48.3328698Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_2, 0.016274645924568176, 6, -128, 127, torch.int8); conv2d_2 = None 2025-09-09T14:58:48.3330412Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.016274645924568176, 6, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:58:48.3331890Z hardtanh = torch.ops.aten.hardtanh.default(dequantize_per_tensor_default_1, -1.0, 1.0); dequantize_per_tensor_default_1 = None 2025-09-09T14:58:48.3333175Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(hardtanh, 0.016274645924568176, 6, -128, 127, torch.int8); hardtanh = None 2025-09-09T14:58:48.3334816Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.016274645924568176, 6, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:58:48.3336194Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T14:58:48.3336714Z 2025-09-09T14:58:48.3337250Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:58:48.3337727Z onverted model fx: GraphModule( 2025-09-09T14:58:48.3338202Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T14:58:48.3338904Z (hardtanh): Hardtanh(min_val=-1.0, max_val=1.0) 2025-09-09T14:58:48.3339269Z ) 2025-09-09T14:58:48.3339391Z 2025-09-09T14:58:48.3339397Z 2025-09-09T14:58:48.3339403Z 2025-09-09T14:58:48.3339513Z def forward(self, x): 2025-09-09T14:58:48.3340324Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:58:48.3341883Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:58:48.3343146Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:58:48.3344219Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.016274645924568176, 6, -128, 127, torch.int8); conv = None 2025-09-09T14:58:48.3345822Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.016274645924568176, 6, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:58:48.3347146Z hardtanh = self.hardtanh(dequantize_per_tensor_default_1); dequantize_per_tensor_default_1 = None 2025-09-09T14:58:48.3348295Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(hardtanh, 0.016274645924568176, 6, -128, 127, torch.int8); hardtanh = None 2025-09-09T14:58:48.3350027Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.016274645924568176, 6, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:58:48.3351091Z return dequantize_per_tensor_default_2 2025-09-09T14:58:48.3351368Z 2025-09-09T14:58:48.3351646Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:58:48.3352024Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:58:48.3352262Z [0., 0., 0.], 2025-09-09T14:58:48.3352482Z [0., 0., 0.]], 2025-09-09T14:58:48.3352624Z 2025-09-09T14:58:48.3352703Z [[0., 0., 0.], 2025-09-09T14:58:48.3352916Z [0., 0., 0.], 2025-09-09T14:58:48.3353121Z [0., 0., 0.]], 2025-09-09T14:58:48.3353266Z 2025-09-09T14:58:48.3353344Z [[0., 0., 0.], 2025-09-09T14:58:48.3353555Z [0., 0., 0.], 2025-09-09T14:58:48.3353763Z [0., 0., 0.]]]]) 2025-09-09T14:58:48.3354005Z model pt2e: GraphModule( 2025-09-09T14:58:48.3354235Z (conv): Module() 2025-09-09T14:58:48.3354446Z (bn): Module() 2025-09-09T14:58:48.3354747Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:58:48.3355692Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:58:48.3356808Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:58:48.3357321Z ) 2025-09-09T14:58:48.3357604Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:58:48.3358545Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:58:48.3359665Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.19029980897903442, max_val=0.18509264290332794) 2025-09-09T14:58:48.3360189Z ) 2025-09-09T14:58:48.3360465Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:58:48.3361492Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0163]), zero_point=tensor([6], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:58:48.3362679Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1751670837402344, max_val=1.979515790939331) 2025-09-09T14:58:48.3363207Z ) 2025-09-09T14:58:48.3363488Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:58:48.3364425Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0163]), zero_point=tensor([6], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:58:48.3365532Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1751670837402344, max_val=1.979515790939331) 2025-09-09T14:58:48.3366039Z ) 2025-09-09T14:58:48.3366212Z ) 2025-09-09T14:58:48.3366311Z 2025-09-09T14:58:48.3366321Z 2025-09-09T14:58:48.3366325Z 2025-09-09T14:58:48.3366418Z def forward(self, x): 2025-09-09T14:58:48.3366702Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:58:48.3367053Z conv_weight = self.conv.weight 2025-09-09T14:58:48.3367327Z conv_bias = self.conv.bias 2025-09-09T14:58:48.3367587Z bn_weight = self.bn.weight 2025-09-09T14:58:48.3367835Z bn_bias = self.bn.bias 2025-09-09T14:58:48.3368099Z bn_running_mean = self.bn.running_mean 2025-09-09T14:58:48.3368400Z bn_running_var = self.bn.running_var 2025-09-09T14:58:48.3368738Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T14:58:48.3369175Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:58:48.3369801Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T14:58:48.3370355Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T14:58:48.3370753Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T14:58:48.3371170Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T14:58:48.3371617Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T14:58:48.3372128Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T14:58:48.3372692Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T14:58:48.3373302Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T14:58:48.3374297Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T14:58:48.3375186Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T14:58:48.3375737Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T14:58:48.3376378Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T14:58:48.3376941Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T14:58:48.3377814Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T14:58:48.3378745Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T14:58:48.3379492Z hardtanh = torch.ops.aten.hardtanh.default(activation_post_process_2, -1.0, 1.0); activation_post_process_2 = None 2025-09-09T14:58:48.3380273Z activation_post_process_3 = self.activation_post_process_3(hardtanh); hardtanh = None 2025-09-09T14:58:48.3380846Z return pytree.tree_unflatten((activation_post_process_3,), self._out_spec) 2025-09-09T14:58:48.3381333Z 2025-09-09T14:58:48.3381617Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:58:48.3381988Z model fx: GraphModule( 2025-09-09T14:58:48.3382397Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:58:48.3383343Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:58:48.3384448Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T14:58:48.3384952Z ) 2025-09-09T14:58:48.3385140Z (conv): ConvBn2d( 2025-09-09T14:58:48.3385372Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T14:58:48.3385793Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T14:58:48.3386262Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:58:48.3387196Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T14:58:48.3388337Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.19029980897903442, max_val=0.18509264290332794) 2025-09-09T14:58:48.3388854Z ) 2025-09-09T14:58:48.3389032Z ) 2025-09-09T14:58:48.3389313Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:59:35.7482384Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0163]), zero_point=tensor([6], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:59:35.7494019Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1751670837402344, max_val=1.979515790939331) 2025-09-09T14:59:35.7494752Z ) 2025-09-09T14:59:35.7495136Z (hardtanh): Hardtanh(min_val=-1.0, max_val=1.0) 2025-09-09T14:59:35.7495659Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T14:59:35.7496713Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0163]), zero_point=tensor([6], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T14:59:35.7498124Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1751670837402344, max_val=1.979515790939331) 2025-09-09T14:59:35.7498641Z ) 2025-09-09T14:59:35.7498817Z ) 2025-09-09T14:59:35.7498917Z 2025-09-09T14:59:35.7498922Z 2025-09-09T14:59:35.7498925Z 2025-09-09T14:59:35.7499018Z def forward(self, x): 2025-09-09T14:59:35.7499389Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T14:59:35.7499942Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T14:59:35.7500499Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T14:59:35.7501165Z hardtanh = self.hardtanh(activation_post_process_1); activation_post_process_1 = None 2025-09-09T14:59:35.7501954Z activation_post_process_2 = self.activation_post_process_2(hardtanh); hardtanh = None 2025-09-09T14:59:35.7502561Z return activation_post_process_2 2025-09-09T14:59:35.7502898Z 2025-09-09T14:59:35.7503180Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:59:35.7503587Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:59:35.7503922Z [0., 0., 0.], 2025-09-09T14:59:35.7504170Z [0., 0., 0.]], 2025-09-09T14:59:35.7504370Z 2025-09-09T14:59:35.7504453Z [[0., 0., 0.], 2025-09-09T14:59:35.7504709Z [0., 0., 0.], 2025-09-09T14:59:35.7505047Z [0., 0., 0.]], 2025-09-09T14:59:35.7505283Z 2025-09-09T14:59:35.7505372Z [[0., 0., 0.], 2025-09-09T14:59:35.7505594Z [0., 0., 0.], 2025-09-09T14:59:35.7506125Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T14:59:35.7506460Z converted model pt2e: GraphModule( 2025-09-09T14:59:35.7506876Z (conv): Module() 2025-09-09T14:59:35.7507094Z (bn): Module() 2025-09-09T14:59:35.7507289Z ) 2025-09-09T14:59:35.7507407Z 2025-09-09T14:59:35.7507411Z 2025-09-09T14:59:35.7507415Z 2025-09-09T14:59:35.7507504Z def forward(self, x): 2025-09-09T14:59:35.7507803Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T14:59:35.7508142Z conv_bias = self.conv.bias 2025-09-09T14:59:35.7509010Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:59:35.7510496Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:59:35.7511539Z quantize_per_tensor = self._frozen_param0 2025-09-09T14:59:35.7512558Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.0014984237495809793, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T14:59:35.7513869Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T14:59:35.7515087Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_2, 0.0162928756326437, 6, -128, 127, torch.int8); conv2d_2 = None 2025-09-09T14:59:35.7516566Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.0162928756326437, 6, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:59:35.7517886Z hardtanh = torch.ops.aten.hardtanh.default(dequantize_per_tensor_default_2, -1.0, 1.0); dequantize_per_tensor_default_2 = None 2025-09-09T14:59:35.7519093Z quantize_per_tensor_default_3 = torch.ops.quantized_decomposed.quantize_per_tensor.default(hardtanh, 0.0162928756326437, 6, -128, 127, torch.int8); hardtanh = None 2025-09-09T14:59:35.7520562Z dequantize_per_tensor_default_3 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_3, 0.0162928756326437, 6, -128, 127, torch.int8); quantize_per_tensor_default_3 = None 2025-09-09T14:59:35.7521583Z return pytree.tree_unflatten((dequantize_per_tensor_default_3,), self._out_spec) 2025-09-09T14:59:35.7522010Z 2025-09-09T14:59:35.7522297Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:59:35.7522741Z onverted model fx: GraphModule( 2025-09-09T14:59:35.7523181Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T14:59:35.7523668Z (hardtanh): Hardtanh(min_val=-1.0, max_val=1.0) 2025-09-09T14:59:35.7524076Z ) 2025-09-09T14:59:35.7524214Z 2025-09-09T14:59:35.7524219Z 2025-09-09T14:59:35.7524223Z 2025-09-09T14:59:35.7524311Z def forward(self, x): 2025-09-09T14:59:35.7524946Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T14:59:35.7526449Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T14:59:35.7527601Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T14:59:35.7528465Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.0162928756326437, 6, -128, 127, torch.int8); conv = None 2025-09-09T14:59:35.7529914Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.0162928756326437, 6, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T14:59:35.7531231Z hardtanh = self.hardtanh(dequantize_per_tensor_default_1); dequantize_per_tensor_default_1 = None 2025-09-09T14:59:35.7532285Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(hardtanh, 0.0162928756326437, 6, -128, 127, torch.int8); hardtanh = None 2025-09-09T14:59:35.7533874Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.0162928756326437, 6, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T14:59:35.7534773Z return dequantize_per_tensor_default_2 2025-09-09T14:59:35.7535074Z 2025-09-09T14:59:35.7535399Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T14:59:35.7535773Z diff: tensor([[[[0., 0., 0.], 2025-09-09T14:59:35.7536098Z [0., 0., 0.], 2025-09-09T14:59:35.7536315Z [0., 0., 0.]], 2025-09-09T14:59:35.7536470Z 2025-09-09T14:59:35.7536553Z [[0., 0., 0.], 2025-09-09T14:59:35.7536771Z [0., 0., 0.], 2025-09-09T14:59:35.7536995Z [0., 0., 0.]], 2025-09-09T14:59:35.7537190Z 2025-09-09T14:59:35.7537308Z [[0., 0., 0.], 2025-09-09T14:59:35.7537539Z [0., 0., 0.], 2025-09-09T14:59:35.7537758Z [0., 0., 0.]]]]) 2025-09-09T14:59:35.7538262Z PASSED 2025-09-09T14:59:35.7539016Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQATModels::test_qat_mobilenet_v2 SKIPPED 2025-09-09T14:59:35.7540105Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQATModels::test_qat_resnet18 SKIPPED 2025-09-09T14:59:35.7541256Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizeMixQATAndPTQ::test_mixing_qat_ptq PASSED 2025-09-09T14:59:35.7542152Z test/quantization/pt2e/test_representation.py::TestPT2ERepresentation::test_add PASSED 2025-09-09T14:59:35.7543000Z test/quantization/pt2e/test_representation.py::TestPT2ERepresentation::test_add_relu PASSED 2025-09-09T14:59:35.7543855Z test/quantization/pt2e/test_representation.py::TestPT2ERepresentation::test_conv2d PASSED 2025-09-09T14:59:35.7544924Z test/quantization/pt2e/test_representation.py::TestPT2ERepresentation::test_dynamic_linear PASSED 2025-09-09T14:59:35.7546028Z test/quantization/pt2e/test_representation.py::TestPT2ERepresentation::test_maxpool2d PASSED 2025-09-09T14:59:35.7546988Z test/quantization/pt2e/test_representation.py::TestPT2ERepresentation::test_qdq PASSED 2025-09-09T14:59:35.7548010Z test/quantization/pt2e/test_representation.py::TestPT2ERepresentation::test_qdq_per_channel PASSED 2025-09-09T14:59:35.7548992Z test/quantization/pt2e/test_representation.py::TestPT2ERepresentation::test_static_linear PASSED 2025-09-09T14:59:35.7550374Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_False_M_1_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.7552285Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_False_M_1_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.7554249Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_False_M_1_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.7556232Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_False_M_1_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.7558032Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_False_M_32_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.7560078Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_False_M_32_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.7744481Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_False_M_32_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.7746384Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_False_M_32_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.7748297Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_True_M_1_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.7750329Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_True_M_1_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.7752033Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_True_M_1_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.7753712Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_True_M_1_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.7755529Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_True_M_32_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.7757535Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_True_M_32_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.7759447Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_True_M_32_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.7761147Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_True_M_32_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.7763061Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_False_M_1_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.7764952Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_False_M_1_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.7766919Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_False_M_1_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.7768591Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_False_M_1_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.7770767Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_False_M_32_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.7772854Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_False_M_32_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.7774736Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_False_M_32_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.7776714Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_False_M_32_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.7778780Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_True_M_1_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.7780671Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_True_M_1_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.7782347Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_True_M_1_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.7784269Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_True_M_1_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.7786155Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_True_M_32_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.7788086Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_True_M_32_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.7789782Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_True_M_32_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.7791772Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_True_M_32_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.7793675Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_False_M_1_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.7795525Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_False_M_1_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.7797519Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_False_M_1_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.7799411Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_False_M_1_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.7801463Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_False_M_32_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.7803332Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_False_M_32_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.7805344Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_False_M_32_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.7807325Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_False_M_32_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.7809251Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_True_M_1_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.7810929Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_True_M_1_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8002083Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_True_M_1_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8003895Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_True_M_1_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8005830Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_True_M_32_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8007840Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_True_M_32_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8009553Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_True_M_32_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8011231Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_True_M_32_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8013021Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_False_M_1_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8015037Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_False_M_1_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8017041Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_False_M_1_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8018742Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_False_M_1_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8020825Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_False_M_32_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8022909Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_False_M_32_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8024734Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_False_M_32_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8026691Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_False_M_32_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8028613Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_True_M_1_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8030623Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_True_M_1_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8032301Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_True_M_1_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8034160Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_True_M_1_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8036171Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_True_M_32_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8038102Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_True_M_32_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8039782Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_True_M_32_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8041793Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_True_M_32_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8043681Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_False_M_1_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8045551Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_False_M_1_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8047338Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_False_M_1_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8049237Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_False_M_1_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8051281Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_False_M_32_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8053118Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_False_M_32_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8054963Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_False_M_32_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8057118Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_False_M_32_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8059035Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_True_M_1_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8060721Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_True_M_1_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8062754Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_True_M_1_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8064645Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_True_M_1_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8066528Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_True_M_32_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8068329Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_True_M_32_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8272609Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_True_M_32_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8274358Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_True_M_32_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8276179Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_False_M_1_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8278091Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_False_M_1_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8280118Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_False_M_1_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8281848Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_False_M_1_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8283641Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_False_M_32_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8285720Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_False_M_32_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8287842Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_False_M_32_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8289650Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_False_M_32_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8291721Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_True_M_1_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8293634Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_True_M_1_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8295445Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_True_M_1_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8297507Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_True_M_1_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8299430Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_True_M_32_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8301486Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_True_M_32_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8303201Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_True_M_32_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8305014Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_True_M_32_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8307055Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_False_M_1_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8308975Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_False_M_1_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8310675Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_False_M_1_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8312695Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_False_M_1_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8314628Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_False_M_32_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8316447Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_False_M_32_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8318404Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_False_M_32_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8320347Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_False_M_32_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8322426Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_True_M_1_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8324114Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_True_M_1_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8326034Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_True_M_1_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8327945Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_True_M_1_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8329860Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_True_M_32_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8331552Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_True_M_32_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8333551Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_True_M_32_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8335503Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_True_M_32_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8337335Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_False_M_1_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8339103Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_False_M_1_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8340984Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_False_M_1_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8666632Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_False_M_1_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8668474Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_False_M_32_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8670471Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_False_M_32_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8672361Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_False_M_32_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8674195Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_False_M_32_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8676349Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_True_M_1_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8678179Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_True_M_1_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8680000Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_True_M_1_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8681789Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_True_M_1_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8683647Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_True_M_32_inplace_add_False_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8685636Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_True_M_32_inplace_add_False_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8687313Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_True_M_32_inplace_add_True_expand_a_scale_False SKIPPED 2025-09-09T14:59:35.8689085Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_True_M_32_inplace_add_True_expand_a_scale_True SKIPPED 2025-09-09T14:59:35.8690561Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_dynamic_qlinear_cpu SKIPPED 2025-09-09T14:59:35.8691752Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_dynamic_qlinear_input_dim_exceeds_2 SKIPPED 2025-09-09T14:59:35.8692918Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_dynamic_qlinear_qat_cpu SKIPPED 2025-09-09T14:59:35.8693883Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_linear_dynamic_fp16 SKIPPED 2025-09-09T14:59:35.8694847Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_linear_relu_dynamic_fp16 SKIPPED 2025-09-09T14:59:35.8695892Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qat_qconv2d SKIPPED 2025-09-09T14:59:35.8696791Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qat_qconv2d_add SKIPPED 2025-09-09T14:59:35.8698057Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qat_qconv2d_add_relu SKIPPED 2025-09-09T14:59:35.8699235Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qat_qconv2d_hardswish SKIPPED 2025-09-09T14:59:35.8700297Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qat_qconv2d_hardtanh SKIPPED 2025-09-09T14:59:35.8701435Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qat_qconv2d_relu SKIPPED 2025-09-09T14:59:35.8702362Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qat_qconv2d_relu6 SKIPPED 2025-09-09T14:59:35.8703272Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qat_qconv2d_silu SKIPPED 2025-09-09T14:59:35.8704342Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qcat SKIPPED 2025-09-09T14:59:35.8705407Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv1d_relu_cpu SKIPPED 2025-09-09T14:59:35.8706556Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_add_2 SKIPPED 2025-09-09T14:59:35.8707607Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_add_3 SKIPPED 2025-09-09T14:59:35.8708790Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_add_broadcast_shapes_cpu SKIPPED 2025-09-09T14:59:35.8709762Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_add_cpu SKIPPED 2025-09-09T14:59:35.8710730Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_add_int8_mixed_bf16 SKIPPED 2025-09-09T14:59:35.8711843Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_add_relu_cpu SKIPPED 2025-09-09T14:59:35.8713044Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_add_relu_int8_mixed_bf16 SKIPPED 2025-09-09T14:59:35.8714129Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_cpu SKIPPED 2025-09-09T14:59:35.8715293Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_dequant_promotion_cpu SKIPPED 2025-09-09T14:59:35.8716294Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_hardswish_cpu SKIPPED 2025-09-09T14:59:35.8717318Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_hardswish_int8_mixed_bf16_cpu SKIPPED 2025-09-09T14:59:35.8718450Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_hardtanh_cpu SKIPPED 2025-09-09T14:59:35.8719672Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_hardtanh_int8_mixed_bf16_cpu SKIPPED 2025-09-09T14:59:35.8720816Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_int8_mixed_bf16 SKIPPED 2025-09-09T14:59:35.8721975Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_relu6_cpu SKIPPED 2025-09-09T14:59:35.8722899Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_relu_cpu SKIPPED 2025-09-09T14:59:35.8723880Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_relu_int8_mixed_bf16_xpu SKIPPED 2025-09-09T14:59:35.8724974Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_silu_cpu SKIPPED 2025-09-09T14:59:35.8726193Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_silu_int8_mixed_bf16_cpu SKIPPED 2025-09-09T14:59:35.8727304Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_with_concat_cpu SKIPPED 2025-09-09T14:59:35.8728362Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qflatten SKIPPED 2025-09-09T14:59:35.8729568Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_cpu_use_relu_False_is_qat_False_is_dynamic_False SKIPPED 2025-09-09T14:59:35.8730801Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_cpu_use_relu_False_is_qat_False_is_dynamic_True SKIPPED 2025-09-09T14:59:35.8732144Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_cpu_use_relu_False_is_qat_True_is_dynamic_False SKIPPED 2025-09-09T14:59:35.8733570Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_cpu_use_relu_False_is_qat_True_is_dynamic_True SKIPPED 2025-09-09T14:59:35.8735104Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_cpu_use_relu_True_is_qat_False_is_dynamic_False SKIPPED 2025-09-09T14:59:35.8736580Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_cpu_use_relu_True_is_qat_False_is_dynamic_True SKIPPED 2025-09-09T14:59:35.8737899Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_cpu_use_relu_True_is_qat_True_is_dynamic_False SKIPPED 2025-09-09T14:59:37.2642369Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_cpu_use_relu_True_is_qat_True_is_dynamic_True SKIPPED 2025-09-09T14:59:37.2643724Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_int8_mixed_bf16_use_relu_False_is_qat_False_is_dynamic_False SKIPPED 2025-09-09T14:59:37.2645065Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_int8_mixed_bf16_use_relu_False_is_qat_False_is_dynamic_True SKIPPED 2025-09-09T14:59:37.2646413Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_int8_mixed_bf16_use_relu_False_is_qat_True_is_dynamic_False SKIPPED 2025-09-09T14:59:37.2647728Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_int8_mixed_bf16_use_relu_False_is_qat_True_is_dynamic_True SKIPPED 2025-09-09T14:59:37.2649029Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_int8_mixed_bf16_use_relu_True_is_qat_False_is_dynamic_False SKIPPED 2025-09-09T14:59:37.2650313Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_int8_mixed_bf16_use_relu_True_is_qat_False_is_dynamic_True SKIPPED 2025-09-09T14:59:37.2651606Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_int8_mixed_bf16_use_relu_True_is_qat_True_is_dynamic_False SKIPPED 2025-09-09T14:59:37.2652899Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_int8_mixed_bf16_use_relu_True_is_qat_True_is_dynamic_True SKIPPED 2025-09-09T14:59:37.2653966Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_cpu SKIPPED 2025-09-09T14:59:37.2654923Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_dequant_promotion_cpu SKIPPED 2025-09-09T14:59:37.2656096Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_dequant_promotion_cpu_input_dim_exceeds_2 SKIPPED 2025-09-09T14:59:37.2657222Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_dequant_promotion_dynamic_cpu SKIPPED 2025-09-09T14:59:37.2658320Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_dequant_promotion_int8_mixed_bf16 SKIPPED 2025-09-09T14:59:37.2659510Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_dequant_promotion_int8_mixed_bf16_input_dim_exceeds_2 SKIPPED 2025-09-09T14:59:37.2660589Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_gelu_cpu SKIPPED 2025-09-09T14:59:37.2661560Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_gelu_int8_mixed_bf16 SKIPPED 2025-09-09T14:59:37.2662558Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_input_dim_exceeds_2 SKIPPED 2025-09-09T14:59:37.2663635Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_input_dim_exceeds_2_and_not_contiguous SKIPPED 2025-09-09T14:59:37.2664686Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_int8_mixed_bf16 SKIPPED 2025-09-09T14:59:37.2665788Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_int8_mixed_bf16_input_dim_exceeds_2 SKIPPED 2025-09-09T14:59:37.2667381Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_int8_mixed_bf16_input_dim_exceeds_2_and_not_contiguous SKIPPED 2025-09-09T14:59:37.2668638Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_mul_cpu SKIPPED 2025-09-09T14:59:37.2669558Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_relu_cpu SKIPPED 2025-09-09T14:59:37.2670549Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_relu_input_dim_exceeds_2 SKIPPED 2025-09-09T14:59:37.2671576Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_relu_int8_mixed_bf16 SKIPPED 2025-09-09T14:59:37.2672666Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_relu_int8_mixed_bf16_input_dim_exceeds_2 SKIPPED 2025-09-09T14:59:37.2673670Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qmaxpool2d SKIPPED 2025-09-09T14:59:37.2674843Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_False_bfloat16_per_channel_quant_False_dynamic_False SKIPPED 2025-09-09T14:59:37.2676314Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_False_bfloat16_per_channel_quant_False_dynamic_True SKIPPED 2025-09-09T14:59:37.2677712Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_False_bfloat16_per_channel_quant_True_dynamic_False SKIPPED 2025-09-09T14:59:37.2679113Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_False_bfloat16_per_channel_quant_True_dynamic_True SKIPPED 2025-09-09T14:59:37.2680514Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_False_float32_per_channel_quant_False_dynamic_False SKIPPED 2025-09-09T14:59:37.2681902Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_False_float32_per_channel_quant_False_dynamic_True SKIPPED 2025-09-09T14:59:37.2683298Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_False_float32_per_channel_quant_True_dynamic_False SKIPPED 2025-09-09T14:59:37.2684686Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_False_float32_per_channel_quant_True_dynamic_True SKIPPED 2025-09-09T14:59:37.2686070Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_True_bfloat16_per_channel_quant_False_dynamic_False SKIPPED 2025-09-09T14:59:37.2687471Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_True_bfloat16_per_channel_quant_False_dynamic_True SKIPPED 2025-09-09T14:59:37.2688862Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_True_bfloat16_per_channel_quant_True_dynamic_False SKIPPED 2025-09-09T14:59:37.2690254Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_True_bfloat16_per_channel_quant_True_dynamic_True SKIPPED 2025-09-09T14:59:37.2691643Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_True_float32_per_channel_quant_False_dynamic_False SKIPPED 2025-09-09T14:59:37.2693021Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_True_float32_per_channel_quant_False_dynamic_True SKIPPED 2025-09-09T14:59:37.2694496Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_True_float32_per_channel_quant_True_dynamic_False SKIPPED 2025-09-09T14:59:37.2695998Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_True_float32_per_channel_quant_True_dynamic_True SKIPPED 2025-09-09T14:59:37.2697268Z test/quantization/pt2e/test_x86inductor_fusion.py::TestDynamicPatternMatcher::test_q_attention_block SKIPPED 2025-09-09T14:59:37.2698487Z test/quantization/pt2e/test_x86inductor_fusion.py::TestDynamicPatternMatcher::test_qat_bn_conv2d SKIPPED 2025-09-09T14:59:37.2699551Z test/quantization/pt2e/test_x86inductor_fusion.py::TestDynamicPatternMatcher::test_qconv2d_maxpool2d_linear_dynamic_cpu SKIPPED 2025-09-09T14:59:37.2700674Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_adaptive_avg_pool2d_recipe PASSED 2025-09-09T14:59:37.2701760Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_annotate_mul_tensor PASSED 2025-09-09T14:59:37.2702789Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_attention_block PASSED 2025-09-09T14:59:37.2703830Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_avg_pool2d_recipe PASSED 2025-09-09T14:59:37.2704929Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_cat_recipe PASSED 2025-09-09T14:59:37.2706200Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_cat_recipe_same_inputs PASSED 2025-09-09T14:59:37.2707357Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_cat_recipe_single_input PASSED 2025-09-09T14:59:37.2708363Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_conv2d PASSED 2025-09-09T14:59:37.2709328Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_conv2d_binary PASSED 2025-09-09T15:04:19.3201335Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_conv2d_binary2 PASSED 2025-09-09T15:04:19.3203779Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_conv2d_binary_unary PASSED 2025-09-09T15:04:19.3204889Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_conv2d_serials_binary_unary PASSED 2025-09-09T15:04:19.3205957Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_conv2d_unary PASSED 2025-09-09T15:04:19.3206974Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_dynamic_quant_linear PASSED 2025-09-09T15:04:19.3208030Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_filter_conv2d_recipe PASSED 2025-09-09T15:04:19.3209090Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_filter_linear_recipe PASSED 2025-09-09T15:04:19.3210172Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_filter_maxpool2d_recipe PASSED 2025-09-09T15:04:19.3211215Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_flatten_recipe PASSED 2025-09-09T15:04:19.3212225Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_flatten_recipe2 PASSED 2025-09-09T15:04:19.3213183Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear PASSED 2025-09-09T15:04:19.3214143Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_binary PASSED 2025-09-09T15:04:19.3215136Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_binary2 PASSED 2025-09-09T15:04:19.3216804Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_binary_dynamic PASSED 2025-09-09T15:04:19.3217887Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_binary_dynamic_qat PASSED 2025-09-09T15:04:19.3219133Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_binary_qat PASSED 2025-09-09T15:04:19.3220158Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_binary_unary PASSED 2025-09-09T15:04:19.3221219Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_binary_unary_dynamic PASSED 2025-09-09T15:04:19.3222323Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_binary_unary_dynamic_qat PASSED 2025-09-09T15:04:19.3223430Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_binary_unary_qat PASSED 2025-09-09T15:04:19.3224507Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_binary_unary_serials PASSED 2025-09-09T15:04:19.3225580Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_dynamic_fp16 PASSED 2025-09-09T15:04:19.3226584Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_unary PASSED 2025-09-09T15:04:19.3227585Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_unary_dynamic PASSED 2025-09-09T15:04:19.3228641Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_unary_dynamic_qat PASSED 2025-09-09T15:04:19.3229674Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_unary_qat PASSED 2025-09-09T15:04:19.3230679Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_lowering_to_x86 SKIPPED 2025-09-09T15:04:19.3231689Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_maxpool2d_recipe PASSED 2025-09-09T15:04:19.3232672Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_qat_conv2d PASSED 2025-09-09T15:04:19.3233659Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_qat_conv2d_binary PASSED 2025-09-09T15:04:19.3234677Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_qat_conv2d_binary2 PASSED 2025-09-09T15:04:19.3235714Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_qat_conv2d_binary_unary PASSED 2025-09-09T15:04:19.3236746Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_qat_conv2d_unary PASSED 2025-09-09T15:04:19.3237785Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_qat_dynamic_quant_linear PASSED 2025-09-09T15:04:19.3238903Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_set_module_name_and_module_type_case1 PASSED 2025-09-09T15:04:19.3240063Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_set_module_name_and_module_type_case2 PASSED 2025-09-09T15:04:19.3241268Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_set_module_name_and_module_type_with_mixed_configs PASSED 2025-09-09T15:04:19.3242422Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_set_module_name_qconfig PASSED 2025-09-09T15:04:19.3243562Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_set_module_name_qconfig_for_dynamic_quant PASSED 2025-09-09T15:04:19.3244845Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_set_module_name_qconfig_with_underscores PASSED 2025-09-09T15:04:19.3246045Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_set_module_name_with_mixed_configs PASSED 2025-09-09T15:04:19.3247175Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_bmm SKIPPED 2025-09-09T15:04:19.3248189Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_cat_granularity0_sizes0 SKIPPED 2025-09-09T15:04:19.3249281Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_cat_granularity0_sizes1 SKIPPED 2025-09-09T15:04:19.3250358Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_cat_granularity0_sizes2 SKIPPED 2025-09-09T15:04:19.3251448Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_cat_granularity1_sizes0 SKIPPED 2025-09-09T15:04:19.3252528Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_cat_granularity1_sizes1 SKIPPED 2025-09-09T15:04:19.3253620Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_cat_granularity1_sizes2 SKIPPED 2025-09-09T15:04:19.3254721Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_expected_gpu_kernel_fbgemm SKIPPED 2025-09-09T15:04:19.3256333Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:04:19.3258130Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:04:19.3259929Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:04:19.3261750Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:04:19.3263537Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:04:19.3265328Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:04:19.3267116Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:04:19.3268891Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:04:19.3479884Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:04:19.3482202Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:04:19.3484238Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:04:19.3486455Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:04:19.3488324Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:04:19.3490097Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:04:19.3492026Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:04:19.3493919Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:04:19.3495737Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:04:19.3497778Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:04:19.3499905Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:04:19.3501818Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:04:19.3503850Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:04:19.3505641Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:04:19.3507654Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:04:19.3509512Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:04:19.3511370Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:04:19.3513200Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:04:19.3515413Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:04:19.3517680Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:04:19.3519552Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:04:19.3521397Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:04:19.3523343Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:04:19.3525281Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:04:19.3527187Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:04:19.3529031Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:04:19.3530985Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:04:19.3532943Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:04:19.3534792Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:04:19.3536744Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:04:19.3538836Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:04:19.3540907Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:04:19.3542749Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:04:19.3544580Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:04:19.3546817Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:04:19.3548828Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:04:19.3757570Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:04:19.3759456Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:04:19.3761321Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:04:19.3763162Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:04:19.3764976Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:04:19.3766819Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:04:19.3768617Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:04:19.3770769Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:04:19.3772885Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:04:19.3774728Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:04:19.3776663Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:04:19.3778806Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:04:19.3780944Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:04:19.3783214Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:04:19.3785777Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:04:19.3787959Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:04:19.3789921Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:04:19.3791715Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:04:19.3793849Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:04:19.3795968Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:04:19.3798020Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:04:19.3799829Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:04:19.3801879Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:04:19.3803852Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:04:19.3805695Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:04:19.3807616Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:04:19.3809590Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:04:19.3811508Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:04:19.3813336Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:04:19.3815212Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:04:19.3817485Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:04:19.3819777Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:04:19.3821635Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:04:19.3823617Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:04:19.3825499Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:04:19.3827572Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:04:27.1119803Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:04:27.1122926Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:04:27.1125971Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:04:27.1129096Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:04:27.1132086Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:04:27.1135057Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:04:27.1138142Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:04:27.1141213Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:04:27.1144214Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:04:27.1147216Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:04:27.1150600Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:04:27.1155764Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:04:27.1158805Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:04:27.1161809Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:04:27.1164814Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:04:27.1167812Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:04:27.1170478Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_kernel_preference_numerical_equivalence_granularity0_sizes0 SKIPPED 2025-09-09T15:04:27.1172752Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_kernel_preference_numerical_equivalence_granularity0_sizes1 SKIPPED 2025-09-09T15:04:27.1174966Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_kernel_preference_numerical_equivalence_granularity1_sizes0 SKIPPED 2025-09-09T15:04:27.1177212Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_kernel_preference_numerical_equivalence_granularity1_sizes1 SKIPPED 2025-09-09T15:04:27.1179186Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_moe_weight_reshape_ops SKIPPED 2025-09-09T15:04:27.1181049Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_slice_and_copy_similar_to_vllm_granularity0 SKIPPED 2025-09-09T15:04:27.1183003Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_slice_and_copy_similar_to_vllm_granularity1 SKIPPED 2025-09-09T15:04:27.1184794Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_slice_granularity0 SKIPPED 2025-09-09T15:04:27.1186476Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_slice_granularity1 SKIPPED 2025-09-09T15:04:27.1188292Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_slice_preserves_aliasing_granularity0 SKIPPED 2025-09-09T15:04:27.1190244Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_slice_preserves_aliasing_granularity1 SKIPPED 2025-09-09T15:04:27.1192113Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_to_device_granularity0_sizes0 SKIPPED 2025-09-09T15:04:27.1193918Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_to_device_granularity0_sizes1 SKIPPED 2025-09-09T15:04:27.1195703Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_to_device_granularity0_sizes2 SKIPPED 2025-09-09T15:04:27.1197871Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_to_device_granularity1_sizes0 SKIPPED 2025-09-09T15:04:27.1199724Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_to_device_granularity1_sizes1 SKIPPED 2025-09-09T15:04:27.1201657Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_to_device_granularity1_sizes2 SKIPPED 2025-09-09T15:04:27.1203563Z test/quantization/quantize_/workflows/int4/test_int4_marlin_sparse_tensor.py::TestInt4MarlinSparseTensor::test_linear_config0_sizes0 SKIPPED 2025-09-09T15:04:27.1205532Z test/quantization/quantize_/workflows/int4/test_int4_marlin_sparse_tensor.py::TestInt4MarlinSparseTensor::test_linear_config0_sizes1 SKIPPED 2025-09-09T15:04:27.1207512Z test/quantization/quantize_/workflows/int4/test_int4_marlin_sparse_tensor.py::TestInt4MarlinSparseTensor::test_linear_config0_sizes2 SKIPPED 2025-09-09T15:04:27.1209504Z test/quantization/quantize_/workflows/int4/test_int4_marlin_sparse_tensor.py::TestInt4MarlinSparseTensor::test_module_path_config0 SKIPPED 2025-09-09T15:04:27.1211435Z test/quantization/quantize_/workflows/int4/test_int4_marlin_sparse_tensor.py::TestInt4MarlinSparseTensor::test_to_device_config0 SKIPPED 2025-09-09T15:04:27.1213372Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes0_bfloat16_group_size_128 PASSED 2025-09-09T15:04:27.1215304Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes0_bfloat16_group_size_32 PASSED 2025-09-09T15:04:27.1217327Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes0_bfloat16_group_size_64 PASSED 2025-09-09T15:04:27.1219313Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes0_float16_group_size_128 PASSED 2025-09-09T15:04:27.1221244Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes0_float16_group_size_32 PASSED 2025-09-09T15:04:27.1223166Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes0_float16_group_size_64 PASSED 2025-09-09T15:04:27.1225093Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes0_float32_group_size_128 PASSED 2025-09-09T15:04:27.1227018Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes0_float32_group_size_32 PASSED 2025-09-09T15:04:36.2159905Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes0_float32_group_size_64 PASSED 2025-09-09T15:04:36.2161194Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes1_bfloat16_group_size_128 PASSED 2025-09-09T15:04:36.2162464Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes1_bfloat16_group_size_32 PASSED 2025-09-09T15:04:36.2163702Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes1_bfloat16_group_size_64 PASSED 2025-09-09T15:04:36.2164933Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes1_float16_group_size_128 PASSED 2025-09-09T15:04:36.2166145Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes1_float16_group_size_32 PASSED 2025-09-09T15:04:36.2167364Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes1_float16_group_size_64 PASSED 2025-09-09T15:04:36.2168876Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes1_float32_group_size_128 PASSED 2025-09-09T15:04:36.2170096Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes1_float32_group_size_32 PASSED 2025-09-09T15:04:36.2171489Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes1_float32_group_size_64 PASSED 2025-09-09T15:04:36.2172744Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes2_bfloat16_group_size_128 PASSED 2025-09-09T15:04:36.2173967Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes2_bfloat16_group_size_32 PASSED 2025-09-09T15:04:36.2175187Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes2_bfloat16_group_size_64 PASSED 2025-09-09T15:04:36.2176505Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes2_float16_group_size_128 PASSED 2025-09-09T15:04:36.2177716Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes2_float16_group_size_32 PASSED 2025-09-09T15:04:36.2178937Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes2_float16_group_size_64 PASSED 2025-09-09T15:04:36.2180154Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes2_float32_group_size_128 PASSED 2025-09-09T15:04:36.2181373Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes2_float32_group_size_32 PASSED 2025-09-09T15:04:36.2182589Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes2_float32_group_size_64 PASSED 2025-09-09T15:04:36.2183762Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_module_path_bfloat16 PASSED 2025-09-09T15:04:36.2184880Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_module_path_float16 PASSED 2025-09-09T15:04:36.2185990Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_module_path_float32 PASSED 2025-09-09T15:04:36.2187184Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes0_bfloat16_group_size_128 SKIPPED 2025-09-09T15:04:36.2188456Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes0_bfloat16_group_size_32 SKIPPED 2025-09-09T15:04:36.2189716Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes0_bfloat16_group_size_64 SKIPPED 2025-09-09T15:04:36.2190985Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes0_float16_group_size_128 SKIPPED 2025-09-09T15:04:36.2192250Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes0_float16_group_size_32 SKIPPED 2025-09-09T15:04:36.2193498Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes0_float16_group_size_64 SKIPPED 2025-09-09T15:04:36.2194759Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes1_bfloat16_group_size_128 SKIPPED 2025-09-09T15:04:36.2196022Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes1_bfloat16_group_size_32 SKIPPED 2025-09-09T15:04:36.2197645Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes1_bfloat16_group_size_64 SKIPPED 2025-09-09T15:04:36.2198931Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes1_float16_group_size_128 SKIPPED 2025-09-09T15:04:36.2200298Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes1_float16_group_size_32 SKIPPED 2025-09-09T15:04:36.2201585Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes1_float16_group_size_64 SKIPPED 2025-09-09T15:04:36.2202870Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes2_bfloat16_group_size_128 SKIPPED 2025-09-09T15:04:36.2204129Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes2_bfloat16_group_size_32 SKIPPED 2025-09-09T15:04:36.2205397Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes2_bfloat16_group_size_64 SKIPPED 2025-09-09T15:04:36.2206663Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes2_float16_group_size_128 SKIPPED 2025-09-09T15:04:36.2207917Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes2_float16_group_size_32 SKIPPED 2025-09-09T15:04:36.2209171Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes2_float16_group_size_64 SKIPPED 2025-09-09T15:04:36.2210379Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_module_path_bfloat16 SKIPPED 2025-09-09T15:04:36.2211574Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_module_path_float16 SKIPPED 2025-09-09T15:04:36.2212754Z test/quantization/quantize_/workflows/int4/test_int4_preshuffled_tensor.py::TestInt4PreshuffledTensor::test_bmm_bmm_config0 SKIPPED 2025-09-09T15:04:36.2213944Z test/quantization/quantize_/workflows/int4/test_int4_preshuffled_tensor.py::TestInt4PreshuffledTensor::test_bmm_bmm_config1 SKIPPED 2025-09-09T15:04:36.2215112Z test/quantization/quantize_/workflows/int4/test_int4_preshuffled_tensor.py::TestInt4PreshuffledTensor::test_linear_config0 SKIPPED 2025-09-09T15:04:36.2216366Z test/quantization/quantize_/workflows/int4/test_int4_preshuffled_tensor.py::TestInt4PreshuffledTensor::test_linear_config1 SKIPPED 2025-09-09T15:04:36.2217555Z test/quantization/quantize_/workflows/int4/test_int4_preshuffled_tensor.py::TestInt4PreshuffledTensor::test_module_path_config0 SKIPPED 2025-09-09T15:04:36.2218777Z test/quantization/quantize_/workflows/int4/test_int4_preshuffled_tensor.py::TestInt4PreshuffledTensor::test_module_path_config1 SKIPPED 2025-09-09T15:04:36.2219984Z test/quantization/quantize_/workflows/int4/test_int4_preshuffled_tensor.py::TestInt4PreshuffledTensor::test_to_device_config0 SKIPPED 2025-09-09T15:04:36.2221179Z test/quantization/quantize_/workflows/int4/test_int4_preshuffled_tensor.py::TestInt4PreshuffledTensor::test_to_device_config1 SKIPPED 2025-09-09T15:04:36.2222286Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_activation_prescaling SKIPPED 2025-09-09T15:04:36.2223251Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_bmm SKIPPED 2025-09-09T15:04:36.2224155Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_cat_sizes0 SKIPPED 2025-09-09T15:04:36.2225089Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_cat_sizes1 SKIPPED 2025-09-09T15:04:36.2226127Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_cat_sizes2 SKIPPED 2025-09-09T15:04:36.2227044Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_linear SKIPPED 2025-09-09T15:04:36.2483519Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_moe_weight_reshape_ops SKIPPED 2025-09-09T15:04:36.2484681Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_slice SKIPPED 2025-09-09T15:04:36.2485892Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_slice_and_copy_similar_to_vllm SKIPPED 2025-09-09T15:04:36.2487189Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_slice_preserves_aliasing SKIPPED 2025-09-09T15:04:36.2488403Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_to_device_sizes0 SKIPPED 2025-09-09T15:04:36.2489586Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_to_device_sizes1 SKIPPED 2025-09-09T15:04:36.2490754Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_to_device_sizes2 SKIPPED 2025-09-09T15:04:36.2492162Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_cant_initialize_in_cpu SKIPPED 2025-09-09T15:04:36.2493822Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_different_group_sizes_group_size_128 SKIPPED 2025-09-09T15:04:36.2495530Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_different_group_sizes_group_size_32 SKIPPED 2025-09-09T15:04:36.2497529Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_different_group_sizes_group_size_64 SKIPPED 2025-09-09T15:04:36.2498854Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_error_conditions SKIPPED 2025-09-09T15:04:36.2500125Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_linear_sizes0_config0 SKIPPED 2025-09-09T15:04:36.2501428Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_linear_sizes0_config1 SKIPPED 2025-09-09T15:04:36.2502737Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_linear_sizes1_config0 SKIPPED 2025-09-09T15:04:36.2504012Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_linear_sizes1_config1 SKIPPED 2025-09-09T15:04:36.2505298Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_linear_sizes2_config0 SKIPPED 2025-09-09T15:04:36.2506571Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_linear_sizes2_config1 SKIPPED 2025-09-09T15:04:36.2507892Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_mm_int4wo_device_cuda_bfloat16 SKIPPED 2025-09-09T15:04:36.2509199Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_module_path_config0 SKIPPED 2025-09-09T15:04:36.2510458Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_module_path_config1 SKIPPED 2025-09-09T15:04:36.2511849Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_slice_and_copy_similar_to_vllm_config0 SKIPPED 2025-09-09T15:04:36.2513463Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_slice_and_copy_similar_to_vllm_config1 SKIPPED 2025-09-09T15:04:36.2514880Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_slice_config0 SKIPPED 2025-09-09T15:04:36.2516103Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_slice_config1 SKIPPED 2025-09-09T15:04:36.2517403Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_slice_preserves_aliasing_config0 SKIPPED 2025-09-09T15:04:36.2518760Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_slice_preserves_aliasing_config1 SKIPPED 2025-09-09T15:04:36.2520037Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_to_device SKIPPED 2025-09-09T15:04:36.2522089Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2524915Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2527710Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2530499Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2533337Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2536191Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2538976Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2541906Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2546103Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2549006Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2666519Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2669374Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2672171Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2674983Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2677835Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2680745Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2683656Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2686626Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2689533Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2692328Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2695183Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2698317Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2701159Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2703954Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2706740Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2709531Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2712383Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2715394Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2718232Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2721160Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2724006Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2842734Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2845593Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2848484Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2851333Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2854131Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2857012Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2859798Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2862843Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2865847Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2868687Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2871505Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2874289Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2877088Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2879932Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2882888Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2885743Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_kleidiai', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2888561Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_kleidiai', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2891547Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_kleidiai', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2894582Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_kleidiai', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.2897695Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.2900518Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3018329Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3021141Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3023993Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3026809Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3029598Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3032450Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3035480Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3038489Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3041350Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3044206Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3047009Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3049829Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3060783Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3063791Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3066648Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3069464Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3072424Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3075315Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3078167Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3081075Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3083972Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3199521Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3202399Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3205204Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3208062Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3210975Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3214017Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3217020Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3219804Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3222663Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3225517Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3228423Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3231268Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3234109Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3236909Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3239706Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3242638Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3245612Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3248447Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3251252Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3254091Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3256953Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3377285Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3380201Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3383073Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3385934Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3388924Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3392011Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3394844Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3397919Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3400765Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3403606Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3406497Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3409453Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3412395Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3415249Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3418279Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3421230Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3424170Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3427131Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3430031Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3432887Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3435735Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3554357Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3558015Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3560983Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3564085Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3567068Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3569920Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3572773Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3575674Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3578750Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3581648Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3584504Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3587347Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3590184Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3593233Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3596275Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3599468Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3602389Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3605253Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3608100Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3611006Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3613975Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3727288Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3730911Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3734667Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3737675Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3740578Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3743582Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3746442Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3749223Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3752040Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3754807Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3757576Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3760358Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3763200Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3766041Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3768860Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3771803Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3774625Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3777459Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3780225Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3783046Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3785868Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3889894Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3892932Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3895806Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3898862Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3901675Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3904516Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3907388Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3910200Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3912977Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3915746Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3918569Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3921512Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3924446Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3927369Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3930154Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3932978Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3935737Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3938596Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3941484Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.3944358Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.3947139Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4053267Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4056269Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4059207Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4062141Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4064954Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4067721Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4070478Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4073287Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4076096Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4078968Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4081786Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_kleidiai', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4084655Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_kleidiai', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4087582Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_kleidiai', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4090503Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_kleidiai', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4093410Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4096260Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4099173Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4101959Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4104743Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4107535Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4110314Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4221211Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4224232Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4227138Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4229971Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4232820Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4235599Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4238397Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4241238Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4244134Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4247000Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4249866Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4252752Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4255538Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4258463Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4261363Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4264238Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4267028Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4269804Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4272632Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4275470Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4278435Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4382310Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4385831Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4389328Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4392883Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4396446Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4399731Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4402618Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4405418Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4408195Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4411111Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4413950Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4417038Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4419868Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4422709Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4425477Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4428253Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4431082Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4434028Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4436887Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4439819Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4442646Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4539703Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4542618Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4545474Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4548306Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4551140Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4554091Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4557042Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4559932Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4562907Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4565752Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4568719Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4571670Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4574629Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4577728Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4580573Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4583452Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4586279Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4589163Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4592231Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4595110Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4598231Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4697034Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4700083Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4703030Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4705976Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4708870Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4711724Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4714570Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4718478Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4721383Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4724491Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4727382Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4730234Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4733125Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4736007Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4738901Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4741910Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4744800Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4747736Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4750639Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.4753533Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.4756425Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:04:36.7898829Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:04:36.7901450Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_export_compile_aoti SKIPPED 2025-09-09T15:04:36.7902876Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_moe_quant_intx SKIPPED 2025-09-09T15:04:36.7904682Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_serialization_{'packing_format': , 'compute_target': 'aten'} SKIPPED 2025-09-09T15:04:36.7906919Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_serialization_{'packing_format': , 'compute_target': 'torchao_auto'} SKIPPED 2025-09-09T15:04:36.7908783Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_embedding PASSED 2025-09-09T15:04:36.7910438Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_export_int8_dyn_act_intx_weight_config PASSED 2025-09-09T15:04:36.7912332Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_export_int8_dyn_act_intx_weight_config_with_unwrap PASSED 2025-09-09T15:04:36.7914137Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_export_intx_weight_only_config PASSED 2025-09-09T15:04:36.7916821Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:36.7920676Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:36.7924245Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:36.7927897Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:36.7931377Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:36.7934803Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:36.7938377Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:36.7941901Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:36.7945375Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:36.7948844Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:36.7952359Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:36.7955840Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:36.7959399Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:36.7963003Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:36.7966459Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:36.7969941Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:36.7973459Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:36.8872759Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:36.8876326Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:36.8879802Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:36.8883373Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:36.8886853Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:36.8890682Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:36.8894465Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:36.8898178Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:36.8901635Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:36.8905110Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:36.8908588Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:36.8912104Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:36.8915537Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:36.8918997Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:36.8922522Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:36.8926122Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:36.8929733Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:36.8933206Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:36.8936767Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:36.8940251Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:36.8943791Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:36.8947270Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:36.9846639Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:36.9850176Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:36.9853666Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:36.9857464Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:36.9860926Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:36.9864597Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:36.9868049Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:36.9871528Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:36.9875034Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:36.9878489Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:36.9881996Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:36.9885422Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:36.9888862Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:36.9892294Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:36.9895787Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:36.9899581Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:36.9903198Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:36.9906651Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:36.9910109Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:36.9913713Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:36.9917166Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:36.9920632Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.0841801Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.0845354Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.0849187Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.0852949Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.0856454Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.0859896Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.0863392Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.0866931Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.0870386Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.0873888Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.0877336Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.0880781Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.0884381Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.0887827Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.0891375Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.0894807Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.0898498Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.0902046Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.0905486Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.0908932Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.0912480Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.0915958Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.1849353Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.1853146Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.1856831Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.1860325Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.1863829Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.1867277Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.1870752Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.1874275Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.1877752Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.1881196Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.1884669Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.1888192Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.1891748Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.1895232Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.1909150Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.1912643Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.1916113Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.1919583Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.1923079Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.1926529Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.1929952Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.1933625Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.2820715Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.2824289Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.2827738Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.2831216Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.2834733Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.2838197Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.2841656Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.2845139Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.2848604Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.2852434Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.2856027Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.2859664Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.2863190Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.2866649Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.2870102Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.2873577Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.2877031Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.2880466Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.2883995Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.2887433Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.2890999Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.2894608Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.3795623Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.3799430Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.3802905Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.3806439Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.3809870Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.3813352Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.3816918Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.3820389Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.3824278Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.3827917Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.3831376Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.3834823Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.3838241Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.3841659Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.3845103Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.3848530Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.3852014Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.3855523Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.3859188Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.3862686Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.3866218Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.3869722Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.4755879Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.4759412Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.4762842Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.4766275Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.4769714Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.4773293Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.4776834Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.4780670Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.4784355Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.4787829Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.4791300Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.4794728Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.4798350Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.4801874Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.4805343Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.4808801Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.4812304Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.4815963Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.4819547Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.4823008Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.4826455Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.4829887Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.5723891Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.5727395Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.5730802Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.5734260Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.5737763Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.5741539Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.5745044Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.5748694Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.5752122Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.5755568Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.5759024Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.5762511Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.5765953Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.5769396Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.5772874Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.5776386Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.5779917Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.5783492Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.5786958Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.5790392Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.5793905Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.5797578Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.6716553Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.6720121Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.6723569Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.6727030Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.6730874Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.6734472Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.6737989Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.6741413Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.6744908Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.6748339Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.6751767Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.6755248Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.6758665Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.6762058Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.6765585Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.6769082Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.6772536Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.6776034Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.6779504Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.6782991Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.6786455Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.6789893Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.9598861Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.9601351Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_linear PASSED 2025-09-09T15:04:37.9603747Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.9607251Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.9610295Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.9613607Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.9616713Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.9619734Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.9622784Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.9625788Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.9628790Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.9631787Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.9634829Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.9637876Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.9640904Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.9644093Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.9647186Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.9650189Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.9653285Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.9656394Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.9659412Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.9662482Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.9665480Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:37.9668463Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:37.9671450Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:37.9674475Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.2396968Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.2400673Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.2403954Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.2406978Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.2410036Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.2413070Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.2416196Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.2419239Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.2422307Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.2425323Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.2428340Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.2431349Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.2434516Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.2437542Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.2440660Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.2443678Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.2446715Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.2449740Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.2452821Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.2456050Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.2459083Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.2462116Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.2465187Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.2468198Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.2471337Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.5123530Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.5126686Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.5129760Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.5132889Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.5135993Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.5139058Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.5142141Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.5145192Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.5148262Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.5151339Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.5154448Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.5157958Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.5161232Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.5164328Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.5167386Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.5170432Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.5173545Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.5176686Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.5179720Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.5182848Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.5185889Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.5188921Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.5191944Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.5195134Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.5198457Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.7890271Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.7893433Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.7896546Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.7899759Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.7902801Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.7905861Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.7908890Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.7911935Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.7915070Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.7918343Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.7921433Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.7924649Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.7927701Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.7930763Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.7933834Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.7936987Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.7940122Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.7943284Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.7946394Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.7949482Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.7952556Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.7955834Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:38.7958981Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:38.7962071Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:38.7965157Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.0672295Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.0675406Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.0678491Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.0681547Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.0684625Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.0687656Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.0690706Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.0693777Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.0697071Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.0700466Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.0703567Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.0706660Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.0709734Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.0712977Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.0716048Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.0719107Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.0722163Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.0725219Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.0728317Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.0731353Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.0734629Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.0738027Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.0741063Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.0744118Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.0747124Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.3432257Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.3435365Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.3438453Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.3441668Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.3444767Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.3447820Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.3451098Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.3454186Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.3457539Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.3460586Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.3463605Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.3466623Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.3469643Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.3472796Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.3475883Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.3478946Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.3482056Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.3485161Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.3488324Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.3491472Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.3494494Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.3497773Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.3500856Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.3503964Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.3507029Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.6219823Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.6222961Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.6226033Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.6229099Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.6232158Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.6235466Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.6238665Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.6241736Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.6244847Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.6247885Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.6250958Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.6254022Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.6257140Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.6260256Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.6263348Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.6266419Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.6269659Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.6272710Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.6275862Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.6278924Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.6281973Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.6285014Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.6288073Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.6291193Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.6294504Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.8951236Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.8954340Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.8957421Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.8960675Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.8963734Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.8966973Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.8970056Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.8973108Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.8976187Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.8979262Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.8982331Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.8985381Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.8988423Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.8991491Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.8994598Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.8998149Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.9001382Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.9004529Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.9007614Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.9010741Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.9013823Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.9025776Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:39.9028881Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:39.9031942Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:39.9035067Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:40.1208557Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:40.1210995Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:40.1213636Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:40.1216229Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:40.1218614Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:40.1220995Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:40.1223392Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:40.1225770Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:40.1228167Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:40.1230563Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:40.1232951Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:40.1235347Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:40.1237739Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:40.1240216Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:40.1242626Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:04:40.1245113Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:04:40.1247512Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:04:40.1249401Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_serialization_int8_dyn_act_intx_weight_config PASSED 2025-09-09T15:04:40.1250857Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_serialization_intx_weight_only_config PASSED 2025-09-09T15:04:40.1252139Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_slice PASSED 2025-09-09T15:04:40.1253340Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_slice_and_copy_ PASSED 2025-09-09T15:04:40.1254540Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_to_dtype PASSED 2025-09-09T15:04:40.1255591Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_concat_linear_cpu_x_dim_2_bias_False SKIPPED 2025-09-09T15:04:40.1256552Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_concat_linear_cpu_x_dim_2_bias_True SKIPPED 2025-09-09T15:04:40.1257455Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_concat_linear_cpu_x_dim_3_bias_False SKIPPED 2025-09-09T15:04:40.1258352Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_concat_linear_cpu_x_dim_3_bias_True SKIPPED 2025-09-09T15:04:40.1259326Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_2_bias_False_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:04:40.1260356Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_2_bias_False_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:04:40.1261366Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_2_bias_False_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:04:40.1262381Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_2_bias_False_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:04:40.1263390Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_2_bias_True_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:04:40.1264410Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_2_bias_True_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:04:40.1265415Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_2_bias_True_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:04:40.1266406Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_2_bias_True_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:04:40.1267424Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_3_bias_False_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:04:40.1268548Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_3_bias_False_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:04:40.1269555Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_3_bias_False_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:05:08.8483703Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_3_bias_False_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:05:08.8486741Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_3_bias_True_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:05:08.8487834Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_3_bias_True_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:05:08.8488838Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_3_bias_True_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:05:08.8489866Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_3_bias_True_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:05:08.8490869Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_2_bias_False_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:05:08.8491900Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_2_bias_False_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:05:08.8492950Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_2_bias_False_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:05:08.8493942Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_2_bias_False_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:05:08.8494941Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_2_bias_True_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:05:08.8496034Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_2_bias_True_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:05:08.8497030Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_2_bias_True_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:05:08.8498268Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_2_bias_True_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:05:08.8499265Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_3_bias_False_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:05:08.8500275Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_3_bias_False_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:05:08.8501272Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_3_bias_False_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:05:08.8502316Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_3_bias_False_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:05:08.8503324Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_3_bias_True_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:05:08.8504327Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_3_bias_True_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:05:08.8505326Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_3_bias_True_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:05:08.8506313Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_3_bias_True_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:05:08.8507314Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_2_bias_False_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:05:08.8508322Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_2_bias_False_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:05:08.8509700Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_2_bias_False_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:05:08.8510697Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_2_bias_False_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:05:08.8511852Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_2_bias_True_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:05:08.8512841Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_2_bias_True_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:05:08.8513838Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_2_bias_True_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:05:08.8514829Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_2_bias_True_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:05:08.8515829Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_3_bias_False_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:05:08.8516831Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_3_bias_False_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:05:08.8517835Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_3_bias_False_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:05:08.8518820Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_3_bias_False_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:05:08.8519813Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_3_bias_True_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:05:08.8520805Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_3_bias_True_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:05:08.8521808Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_3_bias_True_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:05:08.8522795Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_3_bias_True_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:05:08.8523658Z test/quantization/test_gptq.py::TestGPTQ::test_gptq_quantizer_int4_weight_only SKIPPED 2025-09-09T15:05:08.8524445Z test/quantization/test_gptq.py::TestMultiTensorFlow::test_multitensor_add_tensors PASSED 2025-09-09T15:05:08.8525276Z test/quantization/test_gptq.py::TestMultiTensorFlow::test_multitensor_inplace_operation PASSED 2025-09-09T15:05:08.8526088Z test/quantization/test_gptq.py::TestMultiTensorFlow::test_multitensor_pad_unpad PASSED 2025-09-09T15:05:08.8526913Z test/quantization/test_gptq.py::TestMultiTensorInputRecorder::test_gptq_with_input_recorder layers.0.attention.wqkv.weight 2025-09-09T15:05:08.8527531Z layers.0.attention.wo.weight 2025-09-09T15:05:08.8527810Z layers.0.feed_forward.w1.weight 2025-09-09T15:05:08.8528082Z layers.0.feed_forward.w3.weight 2025-09-09T15:05:08.8528361Z layers.0.feed_forward.w2.weight 2025-09-09T15:05:08.8528633Z layers.1.attention.wqkv.weight 2025-09-09T15:05:08.8528915Z layers.1.attention.wo.weight 2025-09-09T15:05:08.8529187Z layers.1.feed_forward.w1.weight 2025-09-09T15:05:08.8529455Z layers.1.feed_forward.w3.weight 2025-09-09T15:05:08.8529731Z layers.1.feed_forward.w2.weight 2025-09-09T15:05:08.8529980Z output.weight 2025-09-09T15:05:08.8530230Z PASSED 2025-09-09T15:05:08.8530812Z test/quantization/test_gptq.py::TestMultiTensorInputRecorder::test_multitensor_input_recorder PASSED 2025-09-09T15:05:08.8531910Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_aten SKIPPED 2025-09-09T15:05:08.8533146Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_kleidiai SKIPPED 2025-09-09T15:05:08.8535567Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8539008Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8542274Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8545537Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8614829Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8618292Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8621610Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8624935Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8628195Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8631608Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8635041Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8638406Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8641729Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8645053Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8648318Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8651582Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8654947Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8658373Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8661771Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8665155Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8668418Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8671670Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8675044Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8742423Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8745759Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8749026Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8752336Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8755732Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8759163Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8762594Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8765899Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8769167Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8772424Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8775681Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8779069Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8782441Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8785824Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8789170Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8792436Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8795690Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8799231Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8802656Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8870128Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8873472Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8876736Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8880141Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8883460Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8886945Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8890277Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8893657Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8897043Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8900561Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8903979Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8907404Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8910908Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8914227Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8917640Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8920946Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8924365Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8927794Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.8931150Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.8998264Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.9001585Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.9005074Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.9008430Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.9011984Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.9015370Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.9018759Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.9022114Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.9025404Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.9028756Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.9032170Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.9035609Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.9038997Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.9042351Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.9045640Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.9048996Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.9052457Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.9055804Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.9059205Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.9142601Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.9146955Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.9150403Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.9153873Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.9157225Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.9160536Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.9163898Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.9167197Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.9170539Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.9174004Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.9177536Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.9180918Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.9184220Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.9187515Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.9190872Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:05:08.9194281Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:05:08.9196544Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_export_QDQLayout SKIPPED 2025-09-09T15:05:08.9198263Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_export_compile_aoti_PackedLinearInt8DynamicActivationIntxWeightLayout SKIPPED 2025-09-09T15:05:08.9199991Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_export_dynamic_shape_PackedLinearInt8DynamicActivationIntxWeightLayout SKIPPED 2025-09-09T15:05:08.9201935Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 128, 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9204033Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 128, 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9206087Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 128, 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9326330Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 128, 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9328908Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 128, 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9331479Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 128, 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9334098Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 128, 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9336208Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 128, 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9338260Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 128, 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9340311Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 32, 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9342417Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 32, 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9344464Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 32, 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9346517Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 32, 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9348561Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 32, 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9350607Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 32, 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9352712Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 32, 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9354757Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 32, 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9356880Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 32, 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9358932Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 64, 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9361073Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 64, 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9363121Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 64, 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9365176Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 64, 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9367230Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 64, 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9369277Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 64, 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9371323Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 64, 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9373427Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 64, 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9375482Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 64, 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9377758Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynamicActivationInt4WeightConfig_{'group_size': 32, 'mapping_type': , 'act_mapping_type': } SKIPPED 2025-09-09T15:05:08.9380107Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynamicActivationInt4WeightConfig_{'group_size': 64, 'mapping_type': , 'act_mapping_type': } SKIPPED 2025-09-09T15:05:08.9382843Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9385888Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9388827Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9460619Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9463634Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9466568Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9469506Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9472493Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9475424Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9478371Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9481314Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9484384Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9487323Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9490354Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9493340Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9496359Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9499506Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9502476Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9505408Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9508354Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9511286Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9514350Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9517285Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9520324Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9598265Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9601212Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9604141Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9607067Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9610000Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9612972Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9615959Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9619024Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9621956Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9625023Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9627940Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9630853Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9633835Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9636780Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9639715Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9642652Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9645581Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9648599Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9651528Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9656252Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9659173Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9735668Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9738676Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9741612Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9744598Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9747523Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9750450Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9753575Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9757275Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9761083Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9764790Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9768496Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9772252Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9776041Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9779741Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9783484Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9787176Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9790959Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9793951Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9796871Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9800040Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9802998Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9876496Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9879435Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9882391Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9885328Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9888241Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9891293Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9894393Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9897637Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9900586Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9903579Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9906508Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9909437Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9912367Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9915311Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9918240Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9921322Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9924420Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9927351Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:08.9930283Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:08.9933264Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:08.9936231Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0011487Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0015251Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0018329Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0021251Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0024393Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0027429Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0030346Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0033346Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0036256Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0039184Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0042138Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0045058Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0047981Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0050909Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0053958Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0057035Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0059936Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0062925Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0065837Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0068754Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0071650Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0146510Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0149482Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0152487Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0155581Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0158622Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0161543Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0164540Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0167499Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0170438Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0173582Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0185377Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0188414Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0191431Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0194604Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0198055Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0201081Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0204138Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0207176Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0210183Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0213233Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0216321Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0310708Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0313797Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0316917Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0319978Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0322936Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0325879Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0328830Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0331797Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0334780Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0337793Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0340727Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0343712Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0346736Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0349748Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0352681Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0355629Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0358598Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0361550Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0364559Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0367512Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0370456Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0474992Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0478094Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0481166Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0484165Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0487106Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0490052Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0493158Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0496144Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0499285Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0502270Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0505215Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0508282Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0511332Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0514326Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0517257Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0520199Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0523193Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0526128Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0529060Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0531994Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0534975Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0634465Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0637525Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0640452Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0643376Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0646487Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0649413Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0652391Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0655312Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0658305Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0661258Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0664359Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0667395Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0670347Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0673295Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0676242Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0679202Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0682190Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0685152Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0688098Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0691052Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0694108Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0791511Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0794519Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0797611Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0800556Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0803542Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0806479Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0809423Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0812412Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0815362Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0818510Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0822369Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0826054Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0829741Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0833443Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0837131Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0840833Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0844527Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0848203Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0851903Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0855714Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0858803Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0956037Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0961873Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0964784Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0967727Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0970689Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0973696Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0976708Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0979644Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0982777Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0985840Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0988791Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.0991730Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.0994722Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.0997828Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1000775Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1003723Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1006660Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1009581Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1012713Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1015753Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1018756Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1100295Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1103310Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1106250Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1109195Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1112184Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1115121Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1118057Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1121119Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1124165Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1127508Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1131216Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1134965Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1138123Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1141051Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1144021Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1146949Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1149877Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1152874Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1155891Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1158844Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1161797Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1242423Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1246162Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1249091Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1252031Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1254969Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1258111Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1261049Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1264170Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1267104Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1270041Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1273020Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1275937Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1278859Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1281786Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1284768Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1287779Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1290712Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1293764Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1296751Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1299838Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1302762Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1379970Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1382968Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1385883Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1388810Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1391877Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1394799Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1398055Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1400980Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1403949Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1406866Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1409772Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1412723Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1415660Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1418689Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1421745Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1424734Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1427763Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1430684Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1433668Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1436594Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1439516Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1519858Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1522858Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1525786Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1528855Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1531780Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1536161Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1539085Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1542064Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1544985Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1547914Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1550845Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1553762Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1556693Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1559702Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1562673Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1565673Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1568589Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1571505Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1574440Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1577410Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1580335Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1662544Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1665481Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1668529Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1671444Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1674515Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1677421Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1680354Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1683355Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1686299Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1689255Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1692255Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1698735Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1701849Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1704847Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1707867Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1710812Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1713808Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1716754Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1719693Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1722660Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1725577Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1799146Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1802304Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1805245Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1808252Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1811203Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1814190Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1817189Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1820121Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1823049Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1825980Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1829012Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1832025Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1835067Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1838002Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1840928Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1843912Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1846834Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1849755Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1852682Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1855601Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1858612Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1938427Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1941446Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1944389Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1947332Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1950263Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1953236Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1956166Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1959099Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1962036Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1965083Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1968103Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1971072Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1974002Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1977007Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1979921Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1982895Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1985813Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1988731Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.1991662Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.1994720Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.1997937Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2081792Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2087501Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2092690Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2095631Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2098768Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2101703Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2104696Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2107624Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2110650Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2113726Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2116695Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2119612Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2122538Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2125458Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2128365Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2131302Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2134301Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2137299Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2140302Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2143361Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2146333Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2223452Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2227182Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2230878Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2234635Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2237568Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2240508Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2243494Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2246519Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2249556Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2252610Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2255534Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2258542Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2261486Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2264482Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2267411Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2270344Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2273274Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2276260Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2279261Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2282252Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2285206Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2364784Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2367740Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2370665Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2373636Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2376664Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2379586Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2382661Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2385696Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2388660Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2391611Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2394576Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2397685Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2400647Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2403641Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2406591Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2409538Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2412618Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2415665Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2418711Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2421674Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2424616Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2498527Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2501645Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2504640Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2507575Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2510512Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2513563Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2516646Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2519661Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2522662Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2525607Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2528546Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2531484Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2534476Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2537497Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2540483Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2543591Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2546720Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2549916Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2553030Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2556065Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2559144Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2634037Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2645134Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2648081Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2651011Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2654125Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2657254Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2660248Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2663183Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2666107Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2669023Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2671938Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2674902Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2677832Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2680763Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2683791Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2686778Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2689733Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2692649Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2695559Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2698727Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2701636Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2772517Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2775495Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2778518Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2781550Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2784624Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2787594Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2790510Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2793473Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2796379Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2799458Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2802417Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2805325Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2808232Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2811219Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2814296Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2817304Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2820205Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2823147Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2826054Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2829776Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2833542Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2912191Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2915151Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2918174Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2921205Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2924199Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2927123Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2930050Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2933030Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2936032Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2938955Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2941866Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2944828Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2947800Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2950781Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2953794Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2956723Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2959653Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2962578Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.2965499Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.2968418Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.2971340Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3054165Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3057245Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3060267Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3063294Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3066227Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3069135Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3072045Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3075007Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3077910Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3080811Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3083719Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3086699Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3089710Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3092744Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3095688Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3098909Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3101852Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3104836Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3107768Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3110708Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3113634Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3195233Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3198533Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3201537Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3204473Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3207406Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3210328Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3213313Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3216322Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3219246Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3222221Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3226354Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3229293Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3232273Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3235201Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3238122Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3241053Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3244035Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3246971Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3249896Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3252876Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3255985Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3337622Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3341460Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3345195Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3348891Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3352594Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3356280Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3359997Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3363779Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3367485Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3370589Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3373573Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3376588Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3379529Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3382505Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3385433Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3388357Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3391294Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3394265Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3397183Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3400419Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3403391Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3479057Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3482010Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3484999Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3487921Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3490848Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3493817Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3496805Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3499855Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3502979Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3505909Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3508897Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3511828Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3514794Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3517712Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3520620Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3523589Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3526507Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3529403Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3532501Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3535427Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3538437Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3618570Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3621528Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3624513Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3627440Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3630361Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3633344Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3636372Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3639388Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3642337Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3645344Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3648271Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3651202Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3654173Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3657149Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3660072Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3663040Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3666013Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3668987Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3671910Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3674928Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3677858Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3758738Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3761670Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3764629Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3767548Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3770471Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3773486Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3776599Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3779516Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3782535Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3785454Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3788352Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3791238Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3794184Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3797089Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3800137Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3803146Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3806215Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3809153Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3812177Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3815132Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3818133Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3900001Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3903018Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3905951Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3908887Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3911931Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3915021Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3918001Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3920926Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3923843Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3926763Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3929694Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3932662Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3935593Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3938602Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3941588Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3944644Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3947605Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3950526Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.3953455Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.3956391Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.3959314Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4039017Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4041958Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4044881Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4047884Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4050889Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4053906Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4056891Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4059807Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4062763Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4065697Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4068628Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4071546Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4074478Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4077483Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4080475Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4083495Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4086420Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4089349Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4092275Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4095252Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4098383Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4181122Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4186060Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4192024Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4195048Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4198234Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4201166Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4204128Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4207061Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4209979Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4212944Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4215925Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4218830Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4221831Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4224856Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4227827Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4230754Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4233706Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4236615Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4239532Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4242485Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4245393Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4320599Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4323663Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4326668Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4329665Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4332652Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4335576Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4338584Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4341509Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4344418Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4347349Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4350269Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4353284Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4356341Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4359313Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4362240Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4365201Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4368110Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4371018Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4373935Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4376908Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4379820Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4462904Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4465979Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4468959Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4471893Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4474861Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4477771Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4480681Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4483648Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4486558Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4489481Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4492464Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4495440Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4498668Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4501581Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4504520Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4509581Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4512590Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4515485Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4518429Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4521378Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4524406Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4603179Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4606183Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4609111Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4612038Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4615136Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4618165Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4621104Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4624091Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4627016Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4630008Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4633017Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4635969Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4646582Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4649661Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4652773Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4655782Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4658846Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4661829Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4664853Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4667880Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4670891Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4741156Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4744168Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4747096Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4750145Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4753145Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4756079Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4759017Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4761941Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4764935Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4767928Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4770914Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4773885Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4776911Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4779922Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4782921Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4785871Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4788822Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4791755Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4794750Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4797904Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4800902Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4880468Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4884266Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4887291Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4890242Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4893241Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4896333Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4899671Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4902740Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4905763Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4909723Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4912671Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4915623Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4918640Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4921589Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4924582Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4927526Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4930468Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4933503Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.4936575Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.4939572Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.4942499Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.5013720Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.5016824Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.5019749Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.5022723Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.5025639Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.5028551Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.5031568Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.5034576Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.5037588Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.5040541Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.5043535Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.5046526Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.5049476Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.5052470Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.5055412Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.5058431Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.5061432Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.5064411Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.5067408Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.5070348Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.5073329Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.5153155Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.5156127Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.5159050Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.5161996Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.5164981Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.5168007Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.5171004Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.5174038Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.5177021Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.5179959Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.5182992Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.5185922Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.5188851Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.5191775Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.5194750Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.5198008Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.5200935Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.5203952Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.5206871Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:05:09.5209779Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:05:09.5212763Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:05:09.5214773Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_moe_quant_intx SKIPPED 2025-09-09T15:15:32.0989355Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_serialization_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO)} SKIPPED 2025-09-09T15:15:32.0991348Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_serialization_{'layout': QDQLayout()} SKIPPED 2025-09-09T15:15:32.0992473Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_fp8dq_base_0_single_token SKIPPED 2025-09-09T15:15:32.0993367Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_fp8dq_base_1_multiple_tokens SKIPPED 2025-09-09T15:15:32.0994260Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_fp8dq_fake_dim_0_single_token SKIPPED 2025-09-09T15:15:32.0995155Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_fp8dq_fake_dim_1_multiple_tokens SKIPPED 2025-09-09T15:15:32.0996043Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_fp8wo_base_0_single_token SKIPPED 2025-09-09T15:15:32.0997228Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_fp8wo_base_1_multiple_tokens SKIPPED 2025-09-09T15:15:32.0998366Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_fp8wo_fake_dim_0_single_token SKIPPED 2025-09-09T15:15:32.0999261Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_fp8wo_fake_dim_1_multiple_tokens SKIPPED 2025-09-09T15:15:32.1000143Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int4wo_base_0_single_token SKIPPED 2025-09-09T15:15:32.1001166Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int4wo_base_1_multiple_tokens SKIPPED 2025-09-09T15:15:32.1002044Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int4wo_fake_dim_0_single_token PASSED 2025-09-09T15:15:32.1003050Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int4wo_fake_dim_1_multiple_tokens PASSED 2025-09-09T15:15:32.1003941Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int8dq_base_0_multiple_tokens PASSED 2025-09-09T15:15:32.1004820Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int8dq_fake_dim_0_multiple_tokens PASSED 2025-09-09T15:15:32.1005699Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int8wo_base_0_single_token PASSED 2025-09-09T15:15:32.1006557Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int8wo_base_1_multiple_tokens PASSED 2025-09-09T15:15:32.1007450Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int8wo_base_cpu_0_single_token PASSED 2025-09-09T15:15:32.1008351Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int8wo_base_cpu_1_multiple_tokens PASSED 2025-09-09T15:15:32.1009244Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int8wo_fake_dim_0_single_token PASSED 2025-09-09T15:15:32.1010144Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int8wo_fake_dim_1_multiple_tokens PASSED 2025-09-09T15:15:32.1011053Z test/quantization/test_observer.py::TestQuantFlow::test_block_size_calc_success PASSED 2025-09-09T15:15:32.1011969Z test/quantization/test_observer.py::TestQuantFlow::test_block_size_row_errors PASSED 2025-09-09T15:15:32.1012882Z test/quantization/test_observer.py::TestQuantFlow::test_fixed_qparams_observer PASSED 2025-09-09T15:15:32.1013937Z test/quantization/test_observer.py::TestQuantFlow::test_min_max_per_channel_affine PASSED 2025-09-09T15:15:32.1014738Z test/quantization/test_observer.py::TestQuantFlow::test_min_max_per_tensor_affine PASSED 2025-09-09T15:15:32.1015474Z test/quantization/test_observer.py::TestQuantFlow::test_mse_observer PASSED 2025-09-09T15:15:32.1016432Z test/quantization/test_observer.py::TestLinearObserver::test_linear_observer_tensor_observe_weight_False PASSED 2025-09-09T15:15:32.1017419Z test/quantization/test_observer.py::TestLinearObserver::test_linear_observer_tensor_observe_weight_True PASSED 2025-09-09T15:15:32.1018241Z test/quantization/test_qat.py::TestQAT::test_composable_qat_quantizer PASSED 2025-09-09T15:15:32.1018935Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_config_dtype PASSED 2025-09-09T15:15:32.1019709Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_config_dynamic_and_range_learning PASSED 2025-09-09T15:15:32.1020475Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_config_eps PASSED 2025-09-09T15:15:32.1021186Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_config_granularity PASSED 2025-09-09T15:15:32.1021971Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_config_granularity_error_cases PASSED 2025-09-09T15:15:32.1022768Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_config_mapping_type PASSED 2025-09-09T15:15:32.1023498Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_config_torch_intx PASSED 2025-09-09T15:15:32.1024307Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_per_channel_group PASSED 2025-09-09T15:15:32.1025006Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_per_token PASSED 2025-09-09T15:15:32.1025755Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_per_token_vs_convert_bfloat16 PASSED 2025-09-09T15:15:32.1026575Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_per_token_vs_convert_float16 PASSED 2025-09-09T15:15:32.1027433Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_per_token_vs_convert_float32 PASSED 2025-09-09T15:15:32.1028192Z test/quantization/test_qat.py::TestQAT::test_fake_quantized_embedding_4w PASSED 2025-09-09T15:15:32.1028885Z test/quantization/test_qat.py::TestQAT::test_fake_quantized_linear_4w PASSED 2025-09-09T15:15:32.1029606Z test/quantization/test_qat.py::TestQAT::test_fake_quantized_linear_8da4w PASSED 2025-09-09T15:15:32.1030394Z test/quantization/test_qat.py::TestQAT::test_fake_quantizer_range_learning_is_symmetric_False PASSED 2025-09-09T15:15:32.1031251Z test/quantization/test_qat.py::TestQAT::test_fake_quantizer_range_learning_is_symmetric_True PASSED 2025-09-09T15:15:32.1032001Z test/quantization/test_qat.py::TestQAT::test_fake_quantizer_repr PASSED 2025-09-09T15:15:32.1032654Z test/quantization/test_qat.py::TestQAT::test_fbgemm_fp8_primitives SKIPPED 2025-09-09T15:15:32.1033430Z test/quantization/test_qat.py::TestQAT::test_fbgemm_int4_preshuffled_primitives SKIPPED 2025-09-09T15:15:32.1034166Z test/quantization/test_qat.py::TestQAT::test_float8_fake_quantize_config PASSED 2025-09-09T15:15:32.1034887Z test/quantization/test_qat.py::TestQAT::test_float8_fake_quantize_granularity0 PASSED 2025-09-09T15:15:32.1035642Z test/quantization/test_qat.py::TestQAT::test_float8_fake_quantize_granularity1 PASSED 2025-09-09T15:15:32.1036333Z test/quantization/test_qat.py::TestQAT::test_infer_fp8_int4_config PASSED 2025-09-09T15:15:32.1037018Z test/quantization/test_qat.py::TestQAT::test_infer_int4_weight_only_config PASSED 2025-09-09T15:15:32.1037712Z test/quantization/test_qat.py::TestQAT::test_legacy_quantize_api_e2e PASSED 2025-09-09T15:15:32.1038513Z test/quantization/test_qat.py::TestQAT::test_qat_4w_embedding PASSED 2025-09-09T15:15:32.1039421Z test/quantization/test_qat.py::TestQAT::test_qat_4w_linear tensor(5.9366e-05, device='cuda:0', dtype=torch.bfloat16, 2025-09-09T15:15:32.1040101Z grad_fn=) 2025-09-09T15:15:32.1040455Z PASSED 2025-09-09T15:15:32.1041046Z test/quantization/test_qat.py::TestQAT::test_qat_4w_primitives tensor(0.0005, device='cuda:0', dtype=torch.bfloat16) 2025-09-09T15:15:32.1041755Z PASSED 2025-09-09T15:15:32.1042430Z test/quantization/test_qat.py::TestQAT::test_qat_4w_quantizer tensor(0.0038, device='cuda:0', dtype=torch.bfloat16, grad_fn=) 2025-09-09T15:15:32.1043220Z PASSED 2025-09-09T15:15:32.1043832Z test/quantization/test_qat.py::TestQAT::test_qat_4w_quantizer_gradients PASSED 2025-09-09T15:15:32.1044618Z test/quantization/test_qat.py::TestQAT::test_qat_8da4w_eps PASSED 2025-09-09T15:15:32.1045369Z test/quantization/test_qat.py::TestQAT::test_qat_8da4w_linear PASSED 2025-09-09T15:15:32.1046233Z test/quantization/test_qat.py::TestQAT::test_qat_8da4w_prepare_vs_convert_bfloat16 PASSED 2025-09-09T15:15:32.1047182Z test/quantization/test_qat.py::TestQAT::test_qat_8da4w_prepare_vs_convert_float16 PASSED 2025-09-09T15:15:32.1048165Z test/quantization/test_qat.py::TestQAT::test_qat_8da4w_prepare_vs_convert_float32 PASSED 2025-09-09T15:15:32.1049015Z test/quantization/test_qat.py::TestQAT::test_qat_8da4w_quantizer PASSED 2025-09-09T15:15:32.1049899Z test/quantization/test_qat.py::TestQAT::test_qat_8da4w_quantizer_disable_fake_quant PASSED 2025-09-09T15:15:32.1050974Z test/quantization/test_qat.py::TestQAT::test_qat_8da4w_quantizer_disable_fake_quant_backward PASSED 2025-09-09T15:15:32.1051933Z test/quantization/test_qat.py::TestQAT::test_qat_8da4w_quantizer_gradients PASSED 2025-09-09T15:15:32.1052906Z test/quantization/test_qat.py::TestQAT::test_qat_8da4w_quantizer_meta_weights PASSED 2025-09-09T15:15:32.1053804Z test/quantization/test_qat.py::TestQAT::test_qat_api_convert_no_quantization PASSED 2025-09-09T15:15:32.1054654Z test/quantization/test_qat.py::TestQAT::test_qat_api_deprecation PASSED 2025-09-09T15:15:32.1055472Z test/quantization/test_qat.py::TestQAT::test_qat_config_init PASSED 2025-09-09T15:15:32.1056352Z test/quantization/test_qat.py::TestQAT::test_qat_fp8a4w_quantizer PASSED 2025-09-09T15:15:32.1057174Z test/quantization/test_qat.py::TestQAT::test_qat_linear_bias PASSED 2025-09-09T15:15:32.1058032Z test/quantization/test_qat.py::TestQAT::test_qat_nvfp4_use_per_tensor_scale_False PASSED 2025-09-09T15:15:32.1058972Z test/quantization/test_qat.py::TestQAT::test_qat_nvfp4_use_per_tensor_scale_True PASSED 2025-09-09T15:15:46.6065075Z test/quantization/test_qat.py::TestQAT::test_qat_prototype_bc PASSED 2025-09-09T15:15:46.6066028Z test/quantization/test_qat.py::TestQAT::test_qat_range_learning_is_symmetric_False PASSED 2025-09-09T15:15:46.6066976Z test/quantization/test_qat.py::TestQAT::test_qat_range_learning_is_symmetric_True PASSED 2025-09-09T15:15:46.6067859Z test/quantization/test_qat.py::TestQAT::test_quantize_api_e2e PASSED 2025-09-09T15:15:46.6068696Z test/quantization/test_qat.py::TestQAT::test_quantize_api_errors PASSED 2025-09-09T15:15:46.6069567Z test/quantization/test_qat.py::TestQAT::test_quantize_api_fp8_fp8_granularity0 SKIPPED 2025-09-09T15:15:46.6070516Z test/quantization/test_qat.py::TestQAT::test_quantize_api_fp8_fp8_granularity1 SKIPPED 2025-09-09T15:15:46.6071383Z test/quantization/test_qat.py::TestQAT::test_quantize_api_fp8_int4 SKIPPED 2025-09-09T15:15:46.6072224Z test/quantization/test_qat.py::TestQAT::test_quantize_api_int4_version_1 SKIPPED 2025-09-09T15:15:46.6073078Z test/quantization/test_qat.py::TestQAT::test_quantize_api_int4_version_2 SKIPPED 2025-09-09T15:15:46.6073908Z test/quantization/test_qat.py::TestQAT::test_quantize_api_int8_int4 PASSED 2025-09-09T15:15:46.6075055Z test/quantization/test_qat.py::TestQAT::test_quantize_api_nvfp4 SKIPPED 2025-09-09T15:15:46.6075859Z test/quantization/test_qat.py::TestQAT::test_quantize_api_prepare PASSED 2025-09-09T15:15:46.6076652Z test/quantization/test_qat.py::TestQAT::test_replace_linear_8da4w PASSED 2025-09-09T15:15:46.6077496Z test/quantization/test_qat.py::TestQAT::test_replace_linear_int4 PASSED 2025-09-09T15:15:46.6078400Z test/quantization/test_quant_api.py::TestQuantFlow::test_8da4w_quantizer PASSED 2025-09-09T15:15:46.6079158Z test/quantization/test_quant_api.py::TestQuantFlow::test_8da4w_quantizer_linear_bias PASSED 2025-09-09T15:15:46.6079956Z test/quantization/test_quant_api.py::TestQuantFlow::test_dynamic_quant_gpu_singleline PASSED 2025-09-09T15:15:46.6080838Z test/quantization/test_quant_api.py::TestQuantFlow::test_dynamic_quant_gpu_unified_api_eager_mode_impl SKIPPED 2025-09-09T15:15:46.6081766Z test/quantization/test_quant_api.py::TestQuantFlow::test_dynamic_quant_gpu_unified_api_unified_impl SKIPPED 2025-09-09T15:15:46.6082572Z test/quantization/test_quant_api.py::TestQuantFlow::test_eval_wrapper SKIPPED 2025-09-09T15:15:46.6083299Z test/quantization/test_quant_api.py::TestQuantFlow::test_eval_wrapper_llama3 SKIPPED 2025-09-09T15:15:46.6084057Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4_wo_quant_save_load SKIPPED 2025-09-09T15:15:46.6084906Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_bfloat16_x_dim_2_use_hqq_False PASSED 2025-09-09T15:15:46.6085898Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_bfloat16_x_dim_2_use_hqq_True PASSED 2025-09-09T15:15:46.6086790Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_bfloat16_x_dim_3_use_hqq_False PASSED 2025-09-09T15:15:46.6087678Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_bfloat16_x_dim_3_use_hqq_True PASSED 2025-09-09T15:15:46.6088559Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_float16_x_dim_2_use_hqq_False PASSED 2025-09-09T15:15:46.6089528Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_float16_x_dim_2_use_hqq_True PASSED 2025-09-09T15:15:46.6090402Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_float16_x_dim_3_use_hqq_False PASSED 2025-09-09T15:15:46.6091360Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_float16_x_dim_3_use_hqq_True PASSED 2025-09-09T15:15:46.6092241Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_float32_x_dim_2_use_hqq_False PASSED 2025-09-09T15:15:46.6093113Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_float32_x_dim_2_use_hqq_True PASSED 2025-09-09T15:15:46.6093990Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_float32_x_dim_3_use_hqq_False PASSED 2025-09-09T15:15:46.6094863Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_float32_x_dim_3_use_hqq_True PASSED 2025-09-09T15:15:46.6095696Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cuda_serialization PASSED 2025-09-09T15:15:46.6096567Z test/quantization/test_quant_api.py::TestQuantFlow::test_int8_wo_quant_save_load PASSED 2025-09-09T15:15:46.6098246Z test/quantization/test_quant_api.py::TestQuantFlow::test_int8wo_quantized_model_to_device PASSED 2025-09-09T15:15:46.6099391Z test/quantization/test_quant_api.py::TestQuantFlow::test_module_fqn_to_config_default PASSED 2025-09-09T15:15:46.6100227Z test/quantization/test_quant_api.py::TestQuantFlow::test_module_fqn_to_config_embedding_linear PASSED 2025-09-09T15:15:46.6101076Z test/quantization/test_quant_api.py::TestQuantFlow::test_module_fqn_to_config_module_name PASSED 2025-09-09T15:15:46.6101880Z test/quantization/test_quant_api.py::TestQuantFlow::test_module_fqn_to_config_skip PASSED 2025-09-09T15:15:46.6102805Z test/quantization/test_quant_api.py::TestQuantFlow::test_quantized_model_streaming PASSED 2025-09-09T15:15:46.6103681Z test/quantization/test_quant_api.py::TestQuantFlow::test_quantized_tensor_subclass_8da4w_mapping_type0 PASSED 2025-09-09T15:15:46.6104625Z test/quantization/test_quant_api.py::TestQuantFlow::test_quantized_tensor_subclass_8da4w_mapping_type1 PASSED 2025-09-09T15:15:46.6105506Z test/quantization/test_quant_api.py::TestQuantFlow::test_quantized_tensor_subclass_int4 PASSED 2025-09-09T15:15:46.6106344Z test/quantization/test_quant_api.py::TestQuantFlow::test_quantized_tensor_subclass_int8_wo PASSED 2025-09-09T15:15:46.6107194Z test/quantization/test_quant_api.py::TestQuantFlow::test_quantized_tensor_subclass_save_load PASSED 2025-09-09T15:15:46.6108114Z test/quantization/test_quant_api.py::TestQuantFlow::test_quantized_tensor_subclass_save_load_map_location PASSED 2025-09-09T15:15:46.6108989Z test/quantization/test_quant_api.py::TestQuantFlow::test_quantizer_int4_weight_only SKIPPED 2025-09-09T15:15:46.6109803Z test/quantization/test_quant_api.py::TestQuantFlow::test_workflow_e2e_numerics_config0 PASSED 2025-09-09T15:15:46.6110618Z test/quantization/test_quant_api.py::TestQuantFlow::test_workflow_e2e_numerics_config1 PASSED 2025-09-09T15:15:46.6111427Z test/quantization/test_quant_api.py::TestQuantFlow::test_workflow_e2e_numerics_config10 PASSED 2025-09-09T15:15:46.6112247Z test/quantization/test_quant_api.py::TestQuantFlow::test_workflow_e2e_numerics_config2 PASSED 2025-09-09T15:15:46.6113131Z test/quantization/test_quant_api.py::TestQuantFlow::test_workflow_e2e_numerics_config3 PASSED 2025-09-09T15:15:46.6113945Z test/quantization/test_quant_api.py::TestQuantFlow::test_workflow_e2e_numerics_config4 PASSED 2025-09-09T15:15:46.6114759Z test/quantization/test_quant_api.py::TestQuantFlow::test_workflow_e2e_numerics_config5 PASSED 2025-09-09T15:15:46.6115567Z test/quantization/test_quant_api.py::TestQuantFlow::test_workflow_e2e_numerics_config6 PASSED 2025-09-09T15:15:46.6116454Z test/quantization/test_quant_api.py::TestQuantFlow::test_workflow_e2e_numerics_config7 PASSED 2025-09-09T15:15:46.6117259Z test/quantization/test_quant_api.py::TestQuantFlow::test_workflow_e2e_numerics_config8 PASSED 2025-09-09T15:15:46.6118182Z test/quantization/test_quant_api.py::TestQuantFlow::test_workflow_e2e_numerics_config9 PASSED 2025-09-09T15:15:46.6119046Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_choose_qparams_group_sym PASSED 2025-09-09T15:15:46.6120022Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_choose_qparams_group_sym_no_clipping_err PASSED 2025-09-09T15:15:46.6121006Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_choose_qparams_tensor_asym PASSED 2025-09-09T15:15:46.6121949Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_choose_qparams_tensor_asym_eps PASSED 2025-09-09T15:15:46.6122877Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_choose_qparams_tensor_sym PASSED 2025-09-09T15:15:46.6123795Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_choose_qparams_token_asym PASSED 2025-09-09T15:15:46.6124694Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_fake_quantize_affine PASSED 2025-09-09T15:15:46.6125602Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_fake_quantize_affine_cachemask PASSED 2025-09-09T15:15:46.6126549Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_get_group_qparams_symmetric PASSED 2025-09-09T15:15:46.6127502Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_get_group_qparams_symmetric_memory PASSED 2025-09-09T15:15:46.6128509Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_get_groupwise_affine_qparams PASSED 2025-09-09T15:15:46.6129624Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_groupwise_affine_dequantize_tensor_from_qparams PASSED 2025-09-09T15:15:46.6130703Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_groupwise_affine_quantize_tensor_from_qparams PASSED 2025-09-09T15:15:46.6131725Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_maybe_expand_scale_to_tensor_shape PASSED 2025-09-09T15:15:46.6132726Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_quantize_activation_per_token_abs_max PASSED 2025-09-09T15:15:46.6133757Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_quantize_activation_per_token_abs_max_dtype PASSED 2025-09-09T15:17:48.7137616Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_quantize_activation_per_token_abs_max_zero_input PASSED 2025-09-09T15:17:48.7139079Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_quantize_dequantize_channel_asym PASSED 2025-09-09T15:17:48.7140079Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_quantize_dequantize_channel_asym_4d PASSED 2025-09-09T15:17:48.7141169Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_quantize_dequantize_channel_asym_4d_multi_dim_reduction PASSED 2025-09-09T15:17:48.7142226Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_quantize_dequantize_group_sym PASSED 2025-09-09T15:17:48.7143479Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_quantize_dequantize_tensor_asym PASSED 2025-09-09T15:17:48.7144342Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_raises PASSED 2025-09-09T15:17:48.7145090Z test/sparsity/test_activation24.py::test_sparse24_sm90_sparsify_identity SKIPPED 2025-09-09T15:17:48.7145848Z test/sparsity/test_activation24.py::test_sparse24_sm90_sparsify_identity_scaled SKIPPED 2025-09-09T15:17:48.7146588Z test/sparsity/test_activation24.py::test_sparse24_sm90_sparsify_srelu SKIPPED 2025-09-09T15:17:48.7147440Z test/sparsity/test_activation24.py::test_srelu_fp8_semi_sparse_activation_linear SKIPPED 2025-09-09T15:17:48.7148205Z test/sparsity/test_activation24.py::test_sparse24_fp8_sm90_cutlass_gemm_eye SKIPPED 2025-09-09T15:17:48.7149767Z test/sparsity/test_activation24.py::test_sparse24_fp8_sm90_cutlass_gemm_random_tensor SKIPPED 2025-09-09T15:17:48.7150738Z test/sparsity/test_fast_sparse_training.py::TestRuntimeSemiStructuredSparsity::test_runtime_weight_sparsification SKIPPED 2025-09-09T15:17:48.7151871Z test/sparsity/test_fast_sparse_training.py::TestRuntimeSemiStructuredSparsity::test_runtime_weight_sparsification_compile SKIPPED 2025-09-09T15:17:48.7152803Z test/sparsity/test_marlin.py::SparseMarlin24::test_pack_unpack_equivalence PASSED 2025-09-09T15:17:48.7153583Z test/sparsity/test_marlin.py::SparseMarlin24::test_quant_sparse_marlin_layout_compile PASSED 2025-09-09T15:17:48.7154389Z test/sparsity/test_marlin.py::SparseMarlin24::test_quant_sparse_marlin_layout_eager PASSED 2025-09-09T15:17:48.7155148Z test/sparsity/test_sparse_api.py::TestSemiStructuredSparse::test_sparse SKIPPED 2025-09-09T15:17:48.7155961Z test/sparsity/test_sparse_api.py::TestQuantSemiSparse::test_quant_semi_sparse_compile_False SKIPPED 2025-09-09T15:17:48.7156813Z test/sparsity/test_sparse_api.py::TestQuantSemiSparse::test_sparse_marlin_compile_False PASSED 2025-09-09T15:17:48.7157648Z test/sparsity/test_sparse_api.py::TestQuantSemiSparse::test_sparse_marlin_compile_True PASSED 2025-09-09T15:17:48.7158513Z test/sparsity/test_sparse_api.py::TestBlockSparseWeight::test_sparse_compile_False_input_shape_1 PASSED 2025-09-09T15:17:48.7159434Z test/sparsity/test_sparse_api.py::TestBlockSparseWeight::test_sparse_compile_False_input_shape_1024 PASSED 2025-09-09T15:17:48.7160473Z test/sparsity/test_sparse_api.py::TestBlockSparseWeight::test_sparse_compile_True_input_shape_1 PASSED 2025-09-09T15:17:48.7161387Z test/sparsity/test_sparse_api.py::TestBlockSparseWeight::test_sparse_compile_True_input_shape_1024 PASSED 2025-09-09T15:17:48.7162269Z test/sparsity/test_sparse_api.py::TestQuantBlockSparseWeight::test_sparse_compile_False PASSED 2025-09-09T15:17:48.7163116Z test/sparsity/test_sparse_api.py::TestQuantBlockSparseWeight::test_sparse_compile_True PASSED 2025-09-09T15:17:48.7163868Z test/sparsity/test_supermask.py::TestSupermask::test_from_linear PASSED 2025-09-09T15:17:48.7164645Z test/sparsity/test_supermask.py::TestSupermask::test_supermask_sparsity_level_0_25_blocksize_2 PASSED 2025-09-09T15:17:48.7165521Z test/sparsity/test_supermask.py::TestSupermask::test_supermask_sparsity_level_0_25_blocksize_4 PASSED 2025-09-09T15:17:48.7166405Z test/sparsity/test_supermask.py::TestSupermask::test_supermask_sparsity_level_0_25_blocksize_8 PASSED 2025-09-09T15:17:48.7167301Z test/sparsity/test_supermask.py::TestSupermask::test_supermask_sparsity_level_0_5_blocksize_2 PASSED 2025-09-09T15:17:48.7168210Z test/sparsity/test_supermask.py::TestSupermask::test_supermask_sparsity_level_0_5_blocksize_4 PASSED 2025-09-09T15:17:48.7169089Z test/sparsity/test_supermask.py::TestSupermask::test_supermask_sparsity_level_0_5_blocksize_8 PASSED 2025-09-09T15:17:48.7169923Z test/sparsity/test_wanda.py::TestWandaSparsifier::test_one_layer_mlp_2x4 PASSED 2025-09-09T15:17:48.7170683Z test/sparsity/test_wanda.py::TestWandaSparsifier::test_one_layer_mlp_unstructured PASSED 2025-09-09T15:17:48.7171388Z test/sparsity/test_wanda.py::TestWandaSparsifier::test_prepare PASSED 2025-09-09T15:17:48.7172038Z test/sparsity/test_wanda.py::TestWandaSparsifier::test_squash_mask PASSED 2025-09-09T15:17:48.7172763Z test/sparsity/test_wanda.py::TestWandaSparsifier::test_two_layer_mlp_unstructured PASSED 2025-09-09T15:17:48.7173607Z test/sparsity/test_wanda.py::TestWandaSparsifier::test_two_layer_mlp_unstructured_custom_config PASSED 2025-09-09T15:17:48.7174610Z test/test_ao_models.py::TorchAOBasicTestCase::test_ao_inference_mode_device_cpu_batch_size_1_is_training_False PASSED 2025-09-09T15:17:48.7175629Z test/test_ao_models.py::TorchAOBasicTestCase::test_ao_inference_mode_device_cpu_batch_size_1_is_training_True PASSED 2025-09-09T15:17:48.7176695Z test/test_ao_models.py::TorchAOBasicTestCase::test_ao_inference_mode_device_cpu_batch_size_4_is_training_False PASSED 2025-09-09T15:17:48.7177707Z test/test_ao_models.py::TorchAOBasicTestCase::test_ao_inference_mode_device_cpu_batch_size_4_is_training_True PASSED 2025-09-09T15:17:48.7178682Z test/test_ao_models.py::TorchAOBasicTestCase::test_ao_inference_mode_device_cuda_batch_size_1_is_training_False PASSED 2025-09-09T15:17:48.7179658Z test/test_ao_models.py::TorchAOBasicTestCase::test_ao_inference_mode_device_cuda_batch_size_1_is_training_True PASSED 2025-09-09T15:17:48.7180628Z test/test_ao_models.py::TorchAOBasicTestCase::test_ao_inference_mode_device_cuda_batch_size_4_is_training_False PASSED 2025-09-09T15:17:48.7181600Z test/test_ao_models.py::TorchAOBasicTestCase::test_ao_inference_mode_device_cuda_batch_size_4_is_training_True PASSED 2025-09-09T15:17:48.7182502Z test/test_low_bit_optim.py::TestQuantize::test_bf16_stochastic_round_device_cpu_compile_False PASSED 2025-09-09T15:17:48.7183329Z test/test_low_bit_optim.py::TestQuantize::test_bf16_stochastic_round_device_cpu_compile_True PASSED 2025-09-09T15:17:48.7184161Z test/test_low_bit_optim.py::TestQuantize::test_bf16_stochastic_round_device_cuda_compile_False PASSED 2025-09-09T15:17:48.7184986Z test/test_low_bit_optim.py::TestQuantize::test_bf16_stochastic_round_device_cuda_compile_True PASSED 2025-09-09T15:17:48.7185855Z test/test_low_bit_optim.py::TestQuantize::test_quantize_4bit_with_qmap_compile_device_cpu PASSED 2025-09-09T15:17:48.7186664Z test/test_low_bit_optim.py::TestQuantize::test_quantize_4bit_with_qmap_compile_device_cuda PASSED 2025-09-09T15:17:48.7187475Z test/test_low_bit_optim.py::TestQuantize::test_quantize_4bit_with_qmap_correctness_device_cpu PASSED 2025-09-09T15:17:48.7188360Z test/test_low_bit_optim.py::TestQuantize::test_quantize_4bit_with_qmap_correctness_device_cuda PASSED 2025-09-09T15:17:48.7189173Z test/test_low_bit_optim.py::TestQuantize::test_quantize_8bit_with_qmap_compile_device_cpu PASSED 2025-09-09T15:17:48.7189976Z test/test_low_bit_optim.py::TestQuantize::test_quantize_8bit_with_qmap_compile_device_cuda PASSED 2025-09-09T15:17:48.7190797Z test/test_low_bit_optim.py::TestQuantize::test_quantize_8bit_with_qmap_correctness_device_cpu PASSED 2025-09-09T15:17:48.7191625Z test/test_low_bit_optim.py::TestQuantize::test_quantize_8bit_with_qmap_correctness_device_cuda PASSED 2025-09-09T15:17:48.7192441Z test/test_low_bit_optim.py::TestOptim::test_optim_4bit_correctness_optim_name_Adam4bit SKIPPED 2025-09-09T15:17:48.7193230Z test/test_low_bit_optim.py::TestOptim::test_optim_4bit_correctness_optim_name_AdamW4bit SKIPPED 2025-09-09T15:17:48.7194017Z test/test_low_bit_optim.py::TestOptim::test_optim_8bit_correctness_optim_name_Adam8bit SKIPPED 2025-09-09T15:17:48.7194807Z test/test_low_bit_optim.py::TestOptim::test_optim_8bit_correctness_optim_name_AdamW8bit SKIPPED 2025-09-09T15:17:48.7195627Z test/test_low_bit_optim.py::TestOptim::test_optim_bf16_stochastic_round_correctness PASSED 2025-09-09T15:17:48.7196476Z test/test_low_bit_optim.py::TestOptim::test_optim_cpu_offload_correctness_offload_grad_False_grad_accum_1 PASSED 2025-09-09T15:17:48.7197712Z test/test_low_bit_optim.py::TestOptim::test_optim_cpu_offload_correctness_offload_grad_False_grad_accum_2 PASSED 2025-09-09T15:17:48.7198634Z test/test_low_bit_optim.py::TestOptim::test_optim_cpu_offload_correctness_offload_grad_True_grad_accum_1 PASSED 2025-09-09T15:17:48.7199490Z test/test_low_bit_optim.py::TestOptim::test_optim_cpu_offload_save_load PASSED 2025-09-09T15:17:48.7200255Z test/test_low_bit_optim.py::TestOptim::test_optim_default_dtype_bf16_optim_name_Adam4bit_device_cpu PASSED 2025-09-09T15:17:48.7201211Z test/test_low_bit_optim.py::TestOptim::test_optim_default_dtype_bf16_optim_name_Adam4bit_device_cuda PASSED 2025-09-09T15:17:48.7202065Z test/test_low_bit_optim.py::TestOptim::test_optim_default_dtype_bf16_optim_name_Adam8bit_device_cpu PASSED 2025-09-09T15:17:48.7202922Z test/test_low_bit_optim.py::TestOptim::test_optim_default_dtype_bf16_optim_name_Adam8bit_device_cuda PASSED 2025-09-09T15:23:06.1551825Z test/test_low_bit_optim.py::TestOptim::test_optim_default_dtype_bf16_optim_name_AdamFp8_device_cpu PASSED 2025-09-09T15:23:06.1552778Z test/test_low_bit_optim.py::TestOptim::test_optim_default_dtype_bf16_optim_name_AdamFp8_device_cuda SKIPPED 2025-09-09T15:23:06.1553646Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_Adam4bit_bfloat16_device_cpu PASSED 2025-09-09T15:23:06.1554483Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_Adam4bit_bfloat16_device_cuda PASSED 2025-09-09T15:23:06.1555321Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_Adam4bit_float32_device_cpu PASSED 2025-09-09T15:23:06.1556151Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_Adam4bit_float32_device_cuda PASSED 2025-09-09T15:23:06.1556970Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_Adam8bit_bfloat16_device_cpu PASSED 2025-09-09T15:23:06.1557808Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_Adam8bit_bfloat16_device_cuda PASSED 2025-09-09T15:23:06.1558628Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_Adam8bit_float32_device_cpu PASSED 2025-09-09T15:23:06.1559985Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_Adam8bit_float32_device_cuda PASSED 2025-09-09T15:23:06.1560828Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamFp8_bfloat16_device_cpu PASSED 2025-09-09T15:23:06.1561657Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamFp8_bfloat16_device_cuda SKIPPED 2025-09-09T15:23:06.1562488Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamFp8_float32_device_cpu PASSED 2025-09-09T15:23:06.1563313Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamFp8_float32_device_cuda SKIPPED 2025-09-09T15:23:06.1564153Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamW4bit_bfloat16_device_cpu PASSED 2025-09-09T15:23:06.1564993Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamW4bit_bfloat16_device_cuda PASSED 2025-09-09T15:23:06.1565832Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamW4bit_float32_device_cpu PASSED 2025-09-09T15:23:06.1566663Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamW4bit_float32_device_cuda PASSED 2025-09-09T15:23:06.1567487Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamW8bit_bfloat16_device_cpu PASSED 2025-09-09T15:23:06.1568327Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamW8bit_bfloat16_device_cuda PASSED 2025-09-09T15:23:06.1569278Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamW8bit_float32_device_cpu PASSED 2025-09-09T15:23:06.1570149Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamW8bit_float32_device_cuda PASSED 2025-09-09T15:23:06.1570976Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamWFp8_bfloat16_device_cpu PASSED 2025-09-09T15:23:06.1571813Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamWFp8_bfloat16_device_cuda SKIPPED 2025-09-09T15:23:06.1572768Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamWFp8_float32_device_cpu PASSED 2025-09-09T15:23:06.1573602Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamWFp8_float32_device_cuda SKIPPED 2025-09-09T15:23:06.1574528Z test/test_low_bit_optim.py::TestOptim::test_param_groups_optim_name_Adam4bit_device_cpu PASSED 2025-09-09T15:23:06.1575316Z test/test_low_bit_optim.py::TestOptim::test_param_groups_optim_name_Adam4bit_device_cuda PASSED 2025-09-09T15:23:06.1576220Z test/test_low_bit_optim.py::TestOptim::test_param_groups_optim_name_Adam8bit_device_cpu PASSED 2025-09-09T15:23:06.1576996Z test/test_low_bit_optim.py::TestOptim::test_param_groups_optim_name_Adam8bit_device_cuda PASSED 2025-09-09T15:23:06.1577769Z test/test_low_bit_optim.py::TestOptim::test_param_groups_optim_name_AdamFp8_device_cpu PASSED 2025-09-09T15:23:06.1578544Z test/test_low_bit_optim.py::TestOptim::test_param_groups_optim_name_AdamFp8_device_cuda SKIPPED 2025-09-09T15:23:06.1579327Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass0_shape0_device_cpu PASSED 2025-09-09T15:23:06.1580153Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass0_shape0_device_cuda PASSED 2025-09-09T15:23:06.1580924Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass0_shape1_device_cpu PASSED 2025-09-09T15:23:06.1581702Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass0_shape1_device_cuda PASSED 2025-09-09T15:23:06.1582468Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass1_shape0_device_cpu PASSED 2025-09-09T15:23:06.1583239Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass1_shape0_device_cuda PASSED 2025-09-09T15:23:06.1584075Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass1_shape1_device_cpu PASSED 2025-09-09T15:23:06.1584848Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass1_shape1_device_cuda PASSED 2025-09-09T15:23:06.1585627Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass2_shape0_device_cpu PASSED 2025-09-09T15:23:06.1586402Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass2_shape0_device_cuda SKIPPED 2025-09-09T15:23:06.1587179Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass2_shape1_device_cpu PASSED 2025-09-09T15:23:06.1587958Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass2_shape1_device_cuda SKIPPED 2025-09-09T15:23:06.1588551Z test/test_low_bit_optim.py::TestFSDP2::test_fsdp2 dist init r=0, world=2 2025-09-09T15:23:06.1588938Z dist init r=1, world=2 2025-09-09T15:23:06.1589196Z PASSED 2025-09-09T15:23:06.1589545Z test/test_low_bit_optim.py::TestFSDP2::test_uneven_shard dist init r=0, world=2 2025-09-09T15:23:06.1589945Z dist init r=1, world=2 2025-09-09T15:23:06.1590256Z PASSED 2025-09-09T15:23:06.1590771Z test/test_model_architecture.py::TestModels::test_ln_linear_activation_model_0_cpu PASSED 2025-09-09T15:23:06.1591541Z test/test_model_architecture.py::TestModels::test_ln_linear_activation_model_1_cuda PASSED 2025-09-09T15:23:06.1592276Z test/test_model_architecture.py::TestModels::test_toy_linear_model_0_cpu PASSED 2025-09-09T15:23:06.1593019Z test/test_model_architecture.py::TestModels::test_toy_linear_model_1_cuda PASSED 2025-09-09T15:23:06.1593727Z test/test_model_architecture.py::TestModels::test_transformer_block_0_cpu PASSED 2025-09-09T15:23:06.1594442Z test/test_model_architecture.py::TestModels::test_transformer_block_1_cuda PASSED 2025-09-09T15:23:06.1595293Z test/test_ops.py::TestOps::test_quant_llm_linear_correctness_BS_1_OC_2048_IC_4096_splitK_5_ebits_2_mbits_2_bfloat16 PASSED 2025-09-09T15:23:06.1596250Z test/test_ops.py::TestOps::test_quant_llm_linear_correctness_BS_1_OC_2048_IC_4096_splitK_5_ebits_2_mbits_2_float16 PASSED 2025-09-09T15:23:06.1597235Z test/test_ops.py::TestOps::test_quant_llm_linear_correctness_BS_1_OC_2048_IC_4096_splitK_5_ebits_3_mbits_2_bfloat16 PASSED 2025-09-09T15:23:06.1598442Z test/test_ops.py::TestOps::test_quant_llm_linear_correctness_BS_1_OC_2048_IC_4096_splitK_5_ebits_3_mbits_2_float16 PASSED 2025-09-09T15:23:06.1599473Z test/test_ops.py::TestOps::test_quant_llm_linear_correctness_BS_2_OC_8192_IC_8192_splitK_6_ebits_2_mbits_2_bfloat16 PASSED 2025-09-09T15:23:06.1600412Z test/test_ops.py::TestOps::test_quant_llm_linear_correctness_BS_2_OC_8192_IC_8192_splitK_6_ebits_2_mbits_2_float16 PASSED 2025-09-09T15:23:06.1601363Z test/test_ops.py::TestOps::test_quant_llm_linear_correctness_BS_2_OC_8192_IC_8192_splitK_6_ebits_3_mbits_2_bfloat16 PASSED 2025-09-09T15:23:06.1602300Z test/test_ops.py::TestOps::test_quant_llm_linear_correctness_BS_2_OC_8192_IC_8192_splitK_6_ebits_3_mbits_2_float16 PASSED 2025-09-09T15:23:06.1603090Z test/test_ops.py::TestOps::test_quant_llm_linear_ebits_2_mbits_2_bfloat16 PASSED 2025-09-09T15:23:06.1603758Z test/test_ops.py::TestOps::test_quant_llm_linear_ebits_2_mbits_2_float16 PASSED 2025-09-09T15:23:06.1604419Z test/test_ops.py::TestOps::test_quant_llm_linear_ebits_3_mbits_2_bfloat16 PASSED 2025-09-09T15:23:06.1605088Z test/test_ops.py::TestOps::test_quant_llm_linear_ebits_3_mbits_2_float16 PASSED 2025-09-09T15:23:06.1605970Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_32_bfloat16 SKIPPED 2025-09-09T15:23:06.1607013Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_32_float32 SKIPPED 2025-09-09T15:23:06.1608133Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T15:23:06.1609195Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_64_bfloat16 SKIPPED 2025-09-09T15:23:06.1610283Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_64_float32 SKIPPED 2025-09-09T15:23:06.1611333Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T15:23:06.1612382Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_32_bfloat16 SKIPPED 2025-09-09T15:23:06.1823978Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_32_float32 SKIPPED 2025-09-09T15:23:06.1825056Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T15:23:06.1826101Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_64_bfloat16 SKIPPED 2025-09-09T15:23:06.1827132Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_64_float32 SKIPPED 2025-09-09T15:23:06.1828289Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T15:23:06.1829323Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_32_bfloat16 SKIPPED 2025-09-09T15:23:06.1830379Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_32_float32 SKIPPED 2025-09-09T15:23:06.1831439Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T15:23:06.1832561Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_64_bfloat16 SKIPPED 2025-09-09T15:23:06.1833648Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_64_float32 SKIPPED 2025-09-09T15:23:06.1834685Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T15:23:06.1835729Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_32_bfloat16 SKIPPED 2025-09-09T15:23:06.1836761Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_32_float32 SKIPPED 2025-09-09T15:23:06.1837799Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T15:23:06.1838843Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_64_bfloat16 SKIPPED 2025-09-09T15:23:06.1839880Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_64_float32 SKIPPED 2025-09-09T15:23:06.1840919Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T15:23:06.1841954Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_32_bfloat16 SKIPPED 2025-09-09T15:23:06.1843047Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_32_float32 SKIPPED 2025-09-09T15:23:06.1844088Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T15:23:06.1845133Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_64_bfloat16 SKIPPED 2025-09-09T15:23:06.1846157Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_64_float32 SKIPPED 2025-09-09T15:23:06.1847199Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T15:23:06.1848233Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_32_bfloat16 SKIPPED 2025-09-09T15:23:06.1849261Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_32_float32 SKIPPED 2025-09-09T15:23:06.1850353Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T15:23:06.1851390Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_64_bfloat16 SKIPPED 2025-09-09T15:23:06.1852506Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_64_float32 SKIPPED 2025-09-09T15:23:06.1853536Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T15:23:06.1854565Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_32_bfloat16 SKIPPED 2025-09-09T15:23:06.1855675Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_32_float32 SKIPPED 2025-09-09T15:23:06.1856804Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T15:23:06.1857875Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_64_bfloat16 SKIPPED 2025-09-09T15:23:06.1858896Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_64_float32 SKIPPED 2025-09-09T15:23:06.1859926Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T15:23:06.1861005Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_32_bfloat16 SKIPPED 2025-09-09T15:23:06.1862034Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_32_float32 SKIPPED 2025-09-09T15:23:06.1863061Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T15:23:06.1864095Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_64_bfloat16 SKIPPED 2025-09-09T15:23:06.1865112Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_64_float32 SKIPPED 2025-09-09T15:23:06.1866141Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T15:23:06.1867222Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_32_bfloat16 SKIPPED 2025-09-09T15:23:06.1868237Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_32_float32 SKIPPED 2025-09-09T15:23:06.1869255Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T15:23:06.1870298Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_64_bfloat16 SKIPPED 2025-09-09T15:23:06.1871303Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_64_float32 SKIPPED 2025-09-09T15:23:06.1872328Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T15:23:06.1873358Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_32_bfloat16 SKIPPED 2025-09-09T15:23:06.1874367Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_32_float32 SKIPPED 2025-09-09T15:23:06.1875388Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T15:23:06.1876464Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_64_bfloat16 SKIPPED 2025-09-09T15:23:06.1877473Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_64_float32 SKIPPED 2025-09-09T15:23:06.1878503Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T15:23:06.1879578Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_32_bfloat16 SKIPPED 2025-09-09T15:23:06.8898782Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_32_float32 SKIPPED 2025-09-09T15:23:06.8900243Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T15:23:06.8901345Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_64_bfloat16 SKIPPED 2025-09-09T15:23:06.8902383Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_64_float32 SKIPPED 2025-09-09T15:23:06.8903428Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T15:23:06.8904471Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_32_bfloat16 SKIPPED 2025-09-09T15:23:06.8905501Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_32_float32 SKIPPED 2025-09-09T15:23:06.8906549Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T15:23:06.8907587Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_64_bfloat16 SKIPPED 2025-09-09T15:23:06.8908624Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_64_float32 SKIPPED 2025-09-09T15:23:06.8909759Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T15:23:06.8910853Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_32_bfloat16 SKIPPED 2025-09-09T15:23:06.8911883Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_32_float32 SKIPPED 2025-09-09T15:23:06.8912913Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T15:23:06.8913950Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_64_bfloat16 SKIPPED 2025-09-09T15:23:06.8914979Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_64_float32 SKIPPED 2025-09-09T15:23:06.8916014Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T15:23:06.8917059Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_32_bfloat16 SKIPPED 2025-09-09T15:23:06.8918094Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_32_float32 SKIPPED 2025-09-09T15:23:06.8919203Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T15:23:06.8920250Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_64_bfloat16 SKIPPED 2025-09-09T15:23:06.8921283Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_64_float32 SKIPPED 2025-09-09T15:23:06.8922415Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T15:23:06.8923455Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_32_bfloat16 SKIPPED 2025-09-09T15:23:06.8924514Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_32_float32 SKIPPED 2025-09-09T15:23:06.8925557Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T15:23:06.8926602Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_64_bfloat16 SKIPPED 2025-09-09T15:23:06.8927622Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_64_float32 SKIPPED 2025-09-09T15:23:06.8928669Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T15:23:06.8929713Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_32_bfloat16 SKIPPED 2025-09-09T15:23:06.8930787Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_32_float32 SKIPPED 2025-09-09T15:23:06.8931850Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T15:23:06.8932894Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_64_bfloat16 SKIPPED 2025-09-09T15:23:06.8933962Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_64_float32 SKIPPED 2025-09-09T15:23:06.8935014Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T15:23:06.8936023Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_4096x4096-tiles_2] PASSED 2025-09-09T15:23:06.8936852Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_4096x4096-tiles_4] PASSED 2025-09-09T15:23:06.8937677Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_4096x4096-tiles_8] PASSED 2025-09-09T15:23:06.8938489Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_4096x11008-tiles_2] PASSED 2025-09-09T15:23:06.8939313Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_4096x11008-tiles_4] PASSED 2025-09-09T15:23:06.8940128Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_4096x11008-tiles_8] PASSED 2025-09-09T15:23:06.8941006Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_11008x4096-tiles_2] PASSED 2025-09-09T15:23:06.8941837Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_11008x4096-tiles_4] PASSED 2025-09-09T15:23:06.8942651Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_11008x4096-tiles_8] PASSED 2025-09-09T15:23:06.8943525Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_4096x14336-tiles_2] PASSED 2025-09-09T15:23:06.8944338Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_4096x14336-tiles_4] PASSED 2025-09-09T15:23:06.8945166Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_4096x14336-tiles_8] PASSED 2025-09-09T15:23:06.8945995Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_14336x4096-tiles_2] PASSED 2025-09-09T15:23:06.8946813Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_14336x4096-tiles_4] PASSED 2025-09-09T15:23:06.8947686Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_14336x4096-tiles_8] PASSED 2025-09-09T15:23:06.8948499Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_4096x4096-tiles_2] PASSED 2025-09-09T15:23:06.8949246Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_4096x4096-tiles_4] PASSED 2025-09-09T15:23:06.8949994Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_4096x4096-tiles_8] PASSED 2025-09-09T15:23:06.8950731Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_4096x11008-tiles_2] PASSED 2025-09-09T15:23:06.8951478Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_4096x11008-tiles_4] PASSED 2025-09-09T15:23:06.8952218Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_4096x11008-tiles_8] PASSED 2025-09-09T15:23:06.8952960Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_11008x4096-tiles_2] PASSED 2025-09-09T15:23:06.8953704Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_11008x4096-tiles_4] PASSED 2025-09-09T15:23:06.8954441Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_11008x4096-tiles_8] PASSED 2025-09-09T15:23:06.8955180Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_4096x14336-tiles_2] PASSED 2025-09-09T15:23:06.8955913Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_4096x14336-tiles_4] PASSED 2025-09-09T15:23:06.8956651Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_4096x14336-tiles_8] PASSED 2025-09-09T15:23:06.8957391Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_14336x4096-tiles_2] PASSED 2025-09-09T15:23:10.7895624Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_14336x4096-tiles_4] PASSED 2025-09-09T15:23:10.7897115Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_14336x4096-tiles_8] PASSED 2025-09-09T15:23:10.7899026Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-2-32] PASSED 2025-09-09T15:23:10.7900570Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-2-64] PASSED 2025-09-09T15:23:10.7902171Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-2-128] PASSED 2025-09-09T15:23:10.7903706Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-2-256] PASSED 2025-09-09T15:23:10.7905256Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-4-32] PASSED 2025-09-09T15:23:10.7906802Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-4-64] PASSED 2025-09-09T15:23:10.7908241Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-4-128] PASSED 2025-09-09T15:23:10.7909661Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-4-256] PASSED 2025-09-09T15:23:10.7911064Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-8-32] PASSED 2025-09-09T15:23:10.7912709Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-8-64] PASSED 2025-09-09T15:23:10.7914117Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-8-128] PASSED 2025-09-09T15:23:10.7915529Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-8-256] PASSED 2025-09-09T15:23:10.7916942Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-2-32] PASSED 2025-09-09T15:23:10.7918468Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-2-64] PASSED 2025-09-09T15:23:10.7919894Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-2-128] PASSED 2025-09-09T15:23:10.7921426Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-2-256] PASSED 2025-09-09T15:23:10.7922849Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-4-32] PASSED 2025-09-09T15:23:10.7924262Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-4-64] PASSED 2025-09-09T15:23:10.7925668Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-4-128] PASSED 2025-09-09T15:23:10.7927092Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-4-256] PASSED 2025-09-09T15:23:10.7928513Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-8-32] PASSED 2025-09-09T15:23:10.7929924Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-8-64] PASSED 2025-09-09T15:23:10.7931348Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-8-128] PASSED 2025-09-09T15:23:10.7932808Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-8-256] PASSED 2025-09-09T15:23:10.7934235Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-2-32] PASSED 2025-09-09T15:23:10.7935644Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-2-64] PASSED 2025-09-09T15:23:10.7937243Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-2-128] PASSED 2025-09-09T15:23:10.7938673Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-2-256] PASSED 2025-09-09T15:23:10.7940083Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-4-32] PASSED 2025-09-09T15:23:10.7941515Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-4-64] PASSED 2025-09-09T15:23:10.7942965Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-4-128] PASSED 2025-09-09T15:23:10.7944372Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-4-256] PASSED 2025-09-09T15:23:10.7945793Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-8-32] PASSED 2025-09-09T15:23:10.7947207Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-8-64] PASSED 2025-09-09T15:23:10.7948617Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-8-128] PASSED 2025-09-09T15:23:10.7950052Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-8-256] PASSED 2025-09-09T15:23:10.7951523Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-2-32] PASSED 2025-09-09T15:23:10.7952940Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-2-64] PASSED 2025-09-09T15:23:10.7954347Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-2-128] PASSED 2025-09-09T15:23:10.7955777Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-2-256] PASSED 2025-09-09T15:23:10.7957256Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-4-32] PASSED 2025-09-09T15:23:10.7958668Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-4-64] PASSED 2025-09-09T15:23:10.7960132Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-4-128] PASSED 2025-09-09T15:23:10.7961560Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-4-256] PASSED 2025-09-09T15:23:10.7962966Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-8-32] PASSED 2025-09-09T15:23:10.7964372Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-8-64] PASSED 2025-09-09T15:23:10.7965787Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-8-128] PASSED 2025-09-09T15:23:10.7967210Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-8-256] PASSED 2025-09-09T15:23:10.7968625Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-2-32] PASSED 2025-09-09T15:23:10.7970038Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-2-64] PASSED 2025-09-09T15:23:10.7971458Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-2-128] PASSED 2025-09-09T15:23:10.7972874Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-2-256] PASSED 2025-09-09T15:23:10.7974301Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-4-32] PASSED 2025-09-09T15:23:10.7975834Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-4-64] PASSED 2025-09-09T15:23:10.7977261Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-4-128] PASSED 2025-09-09T15:23:10.7978698Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-4-256] PASSED 2025-09-09T15:23:10.7980103Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-8-32] PASSED 2025-09-09T15:23:10.7981512Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-8-64] PASSED 2025-09-09T15:23:10.7982921Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-8-128] PASSED 2025-09-09T15:23:10.7984342Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-8-256] PASSED 2025-09-09T15:23:10.7985800Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-2-32] PASSED 2025-09-09T15:23:10.7987276Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-2-64] PASSED 2025-09-09T15:23:10.7988749Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-2-128] PASSED 2025-09-09T15:23:10.7990224Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-2-256] PASSED 2025-09-09T15:23:10.7991809Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-4-32] PASSED 2025-09-09T15:23:15.3469182Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-4-64] PASSED 2025-09-09T15:23:15.3470611Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-4-128] PASSED 2025-09-09T15:23:15.3472139Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-4-256] PASSED 2025-09-09T15:23:15.3473079Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-8-32] PASSED 2025-09-09T15:23:15.3474159Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-8-64] PASSED 2025-09-09T15:23:15.3475097Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-8-128] PASSED 2025-09-09T15:23:15.3476038Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-8-256] PASSED 2025-09-09T15:23:15.3476978Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-2-32] PASSED 2025-09-09T15:23:15.3477916Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-2-64] PASSED 2025-09-09T15:23:15.3478874Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-2-128] PASSED 2025-09-09T15:23:15.3479816Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-2-256] PASSED 2025-09-09T15:23:15.3480768Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-4-32] PASSED 2025-09-09T15:23:15.3481711Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-4-64] PASSED 2025-09-09T15:23:15.3482657Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-4-128] PASSED 2025-09-09T15:23:15.3483609Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-4-256] PASSED 2025-09-09T15:23:15.3484654Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-8-32] PASSED 2025-09-09T15:23:15.3485602Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-8-64] PASSED 2025-09-09T15:23:15.3486554Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-8-128] PASSED 2025-09-09T15:23:15.3487494Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-8-256] PASSED 2025-09-09T15:23:15.3488439Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-2-32] PASSED 2025-09-09T15:23:15.3489381Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-2-64] PASSED 2025-09-09T15:23:15.3490320Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-2-128] PASSED 2025-09-09T15:23:15.3491274Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-2-256] PASSED 2025-09-09T15:23:15.3492211Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-4-32] PASSED 2025-09-09T15:23:15.3493204Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-4-64] PASSED 2025-09-09T15:23:15.3494250Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-4-128] PASSED 2025-09-09T15:23:15.3495193Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-4-256] PASSED 2025-09-09T15:23:15.3496233Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-8-32] PASSED 2025-09-09T15:23:15.3497174Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-8-64] PASSED 2025-09-09T15:23:15.3498481Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-8-128] PASSED 2025-09-09T15:23:15.3499488Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-8-256] PASSED 2025-09-09T15:23:15.3500428Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-2-32] PASSED 2025-09-09T15:23:15.3501371Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-2-64] PASSED 2025-09-09T15:23:15.3502323Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-2-128] PASSED 2025-09-09T15:23:15.3503309Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-2-256] PASSED 2025-09-09T15:23:15.3504273Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-4-32] PASSED 2025-09-09T15:23:15.3505219Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-4-64] PASSED 2025-09-09T15:23:15.3506170Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-4-128] PASSED 2025-09-09T15:23:15.3507122Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-4-256] PASSED 2025-09-09T15:23:15.3508063Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-8-32] PASSED 2025-09-09T15:23:15.3509006Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-8-64] PASSED 2025-09-09T15:23:15.3510014Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-8-128] PASSED 2025-09-09T15:23:15.3510971Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-8-256] PASSED 2025-09-09T15:23:15.3511921Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-2-32] PASSED 2025-09-09T15:23:15.3512862Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-2-64] PASSED 2025-09-09T15:23:15.3513812Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-2-128] PASSED 2025-09-09T15:23:15.3514764Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-2-256] PASSED 2025-09-09T15:23:15.3515710Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-4-32] PASSED 2025-09-09T15:23:15.3516656Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-4-64] PASSED 2025-09-09T15:23:15.3517598Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-4-128] PASSED 2025-09-09T15:23:15.3518552Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-4-256] PASSED 2025-09-09T15:23:15.3519578Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-8-32] PASSED 2025-09-09T15:23:15.3520515Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-8-64] PASSED 2025-09-09T15:23:15.3521462Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-8-128] PASSED 2025-09-09T15:23:15.3522410Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-8-256] PASSED 2025-09-09T15:23:15.3523343Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-2-32] PASSED 2025-09-09T15:23:15.3524071Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-2-64] PASSED 2025-09-09T15:23:15.3524838Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-2-128] PASSED 2025-09-09T15:23:15.3525571Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-2-256] PASSED 2025-09-09T15:23:15.3526291Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-4-32] PASSED 2025-09-09T15:23:15.3527014Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-4-64] PASSED 2025-09-09T15:23:15.3527738Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-4-128] PASSED 2025-09-09T15:23:15.3528465Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-4-256] PASSED 2025-09-09T15:23:15.3529195Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-8-32] PASSED 2025-09-09T15:23:15.3529910Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-8-64] PASSED 2025-09-09T15:23:15.3530637Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-8-128] PASSED 2025-09-09T15:23:15.3531369Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-8-256] PASSED 2025-09-09T15:23:19.4164209Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-2-32] PASSED 2025-09-09T15:23:19.4165363Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-2-64] PASSED 2025-09-09T15:23:19.4166363Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-2-128] PASSED 2025-09-09T15:23:19.4167568Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-2-256] PASSED 2025-09-09T15:23:19.4168303Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-4-32] PASSED 2025-09-09T15:23:19.4169011Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-4-64] PASSED 2025-09-09T15:23:19.4169732Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-4-128] PASSED 2025-09-09T15:23:19.4170462Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-4-256] PASSED 2025-09-09T15:23:19.4171174Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-8-32] PASSED 2025-09-09T15:23:19.4171885Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-8-64] PASSED 2025-09-09T15:23:19.4172597Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-8-128] PASSED 2025-09-09T15:23:19.4173321Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-8-256] PASSED 2025-09-09T15:23:19.4174060Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-2-32] PASSED 2025-09-09T15:23:19.4174814Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-2-64] PASSED 2025-09-09T15:23:19.4175538Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-2-128] PASSED 2025-09-09T15:23:19.4176338Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-2-256] PASSED 2025-09-09T15:23:19.4177152Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-4-32] PASSED 2025-09-09T15:23:19.4177864Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-4-64] PASSED 2025-09-09T15:23:19.4178574Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-4-128] PASSED 2025-09-09T15:23:19.4179302Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-4-256] PASSED 2025-09-09T15:23:19.4180107Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-8-32] PASSED 2025-09-09T15:23:19.4190041Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-8-64] PASSED 2025-09-09T15:23:19.4190939Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-8-128] PASSED 2025-09-09T15:23:19.4191670Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-8-256] PASSED 2025-09-09T15:23:19.4192394Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-2-32] PASSED 2025-09-09T15:23:19.4193106Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-2-64] PASSED 2025-09-09T15:23:19.4193828Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-2-128] PASSED 2025-09-09T15:23:19.4194605Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-2-256] PASSED 2025-09-09T15:23:19.4195325Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-4-32] PASSED 2025-09-09T15:23:19.4196041Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-4-64] PASSED 2025-09-09T15:23:19.4196762Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-4-128] PASSED 2025-09-09T15:23:19.4197853Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-4-256] PASSED 2025-09-09T15:23:19.4198585Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-8-32] PASSED 2025-09-09T15:23:19.4199302Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-8-64] PASSED 2025-09-09T15:23:19.4200029Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-8-128] PASSED 2025-09-09T15:23:19.4200846Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-8-256] PASSED 2025-09-09T15:23:19.4201565Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-2-32] PASSED 2025-09-09T15:23:19.4202282Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-2-64] PASSED 2025-09-09T15:23:19.4202998Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-2-128] PASSED 2025-09-09T15:23:19.4203725Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-2-256] PASSED 2025-09-09T15:23:19.4204505Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-4-32] PASSED 2025-09-09T15:23:19.4205213Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-4-64] PASSED 2025-09-09T15:23:19.4205944Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-4-128] PASSED 2025-09-09T15:23:19.4206670Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-4-256] PASSED 2025-09-09T15:23:19.4207394Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-8-32] PASSED 2025-09-09T15:23:19.4208111Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-8-64] PASSED 2025-09-09T15:23:19.4208831Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-8-128] PASSED 2025-09-09T15:23:19.4209565Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-8-256] PASSED 2025-09-09T15:23:19.4210269Z test/test_ops.py::test_marlin_24[1-128-512-4--1-(1, 1, 1)] PASSED 2025-09-09T15:23:19.4210817Z test/test_ops.py::test_marlin_24[1-128-512-4--1-(1, 4, 8)] PASSED 2025-09-09T15:23:19.4211353Z test/test_ops.py::test_marlin_24[1-128-512-4--1-(1, 7, 5)] PASSED 2025-09-09T15:23:19.4211911Z test/test_ops.py::test_marlin_24[1-128-512-4--1-(13, 17, 67)] PASSED 2025-09-09T15:23:19.4212481Z test/test_ops.py::test_marlin_24[1-128-512-4--1-(26, 37, 13)] PASSED 2025-09-09T15:23:19.4213099Z test/test_ops.py::test_marlin_24[1-128-512-4--1-(67, 13, 11)] PASSED 2025-09-09T15:23:19.4213662Z test/test_ops.py::test_marlin_24[1-128-512-4-128-(1, 1, 1)] PASSED 2025-09-09T15:23:19.4214211Z test/test_ops.py::test_marlin_24[1-128-512-4-128-(1, 4, 8)] PASSED 2025-09-09T15:23:19.4214828Z test/test_ops.py::test_marlin_24[1-128-512-4-128-(1, 7, 5)] PASSED 2025-09-09T15:23:19.4215406Z test/test_ops.py::test_marlin_24[1-128-512-4-128-(13, 17, 67)] PASSED 2025-09-09T15:23:19.4216086Z test/test_ops.py::test_marlin_24[1-128-512-4-128-(26, 37, 13)] PASSED 2025-09-09T15:23:19.4216655Z test/test_ops.py::test_marlin_24[1-128-512-4-128-(67, 13, 11)] PASSED 2025-09-09T15:23:19.4217200Z test/test_ops.py::test_marlin_24[1-128-512-8--1-(1, 1, 1)] PASSED 2025-09-09T15:23:19.4217741Z test/test_ops.py::test_marlin_24[1-128-512-8--1-(1, 4, 8)] PASSED 2025-09-09T15:23:19.4218281Z test/test_ops.py::test_marlin_24[1-128-512-8--1-(1, 7, 5)] PASSED 2025-09-09T15:23:19.4218836Z test/test_ops.py::test_marlin_24[1-128-512-8--1-(13, 17, 67)] PASSED 2025-09-09T15:23:19.4219399Z test/test_ops.py::test_marlin_24[1-128-512-8--1-(26, 37, 13)] PASSED 2025-09-09T15:23:19.4219959Z test/test_ops.py::test_marlin_24[1-128-512-8--1-(67, 13, 11)] PASSED 2025-09-09T15:23:19.4220523Z test/test_ops.py::test_marlin_24[1-128-512-8-128-(1, 1, 1)] PASSED 2025-09-09T15:23:19.4221072Z test/test_ops.py::test_marlin_24[1-128-512-8-128-(1, 4, 8)] PASSED 2025-09-09T15:23:19.4221627Z test/test_ops.py::test_marlin_24[1-128-512-8-128-(1, 7, 5)] PASSED 2025-09-09T15:23:19.4222175Z test/test_ops.py::test_marlin_24[1-128-512-8-128-(13, 17, 67)] PASSED 2025-09-09T15:23:19.4222742Z test/test_ops.py::test_marlin_24[1-128-512-8-128-(26, 37, 13)] PASSED 2025-09-09T15:23:19.4223381Z test/test_ops.py::test_marlin_24[1-128-512-8-128-(67, 13, 11)] PASSED 2025-09-09T15:23:19.4223925Z test/test_ops.py::test_marlin_24[4-128-512-4--1-(1, 1, 1)] PASSED 2025-09-09T15:23:19.4224528Z test/test_ops.py::test_marlin_24[4-128-512-4--1-(1, 4, 8)] PASSED 2025-09-09T15:23:19.4225059Z test/test_ops.py::test_marlin_24[4-128-512-4--1-(1, 7, 5)] PASSED 2025-09-09T15:23:19.4225608Z test/test_ops.py::test_marlin_24[4-128-512-4--1-(13, 17, 67)] PASSED 2025-09-09T15:23:19.4226167Z test/test_ops.py::test_marlin_24[4-128-512-4--1-(26, 37, 13)] PASSED 2025-09-09T15:23:19.4226726Z test/test_ops.py::test_marlin_24[4-128-512-4--1-(67, 13, 11)] PASSED 2025-09-09T15:23:19.4227275Z test/test_ops.py::test_marlin_24[4-128-512-4-128-(1, 1, 1)] PASSED 2025-09-09T15:23:19.4227813Z test/test_ops.py::test_marlin_24[4-128-512-4-128-(1, 4, 8)] PASSED 2025-09-09T15:23:19.4228364Z test/test_ops.py::test_marlin_24[4-128-512-4-128-(1, 7, 5)] PASSED 2025-09-09T15:23:19.4228915Z test/test_ops.py::test_marlin_24[4-128-512-4-128-(13, 17, 67)] PASSED 2025-09-09T15:23:19.4229482Z test/test_ops.py::test_marlin_24[4-128-512-4-128-(26, 37, 13)] PASSED 2025-09-09T15:23:19.4230044Z test/test_ops.py::test_marlin_24[4-128-512-4-128-(67, 13, 11)] PASSED 2025-09-09T15:23:19.4230586Z test/test_ops.py::test_marlin_24[4-128-512-8--1-(1, 1, 1)] PASSED 2025-09-09T15:23:19.4231127Z test/test_ops.py::test_marlin_24[4-128-512-8--1-(1, 4, 8)] PASSED 2025-09-09T15:23:19.4231716Z test/test_ops.py::test_marlin_24[4-128-512-8--1-(1, 7, 5)] PASSED 2025-09-09T15:23:19.4232272Z test/test_ops.py::test_marlin_24[4-128-512-8--1-(13, 17, 67)] PASSED 2025-09-09T15:23:19.4232825Z test/test_ops.py::test_marlin_24[4-128-512-8--1-(26, 37, 13)] PASSED 2025-09-09T15:23:19.4233386Z test/test_ops.py::test_marlin_24[4-128-512-8--1-(67, 13, 11)] PASSED 2025-09-09T15:23:19.4233945Z test/test_ops.py::test_marlin_24[4-128-512-8-128-(1, 1, 1)] PASSED 2025-09-09T15:23:19.4234542Z test/test_ops.py::test_marlin_24[4-128-512-8-128-(1, 4, 8)] PASSED 2025-09-09T15:23:19.4235135Z test/test_ops.py::test_marlin_24[4-128-512-8-128-(1, 7, 5)] PASSED 2025-09-09T15:23:22.2070388Z test/test_ops.py::test_marlin_24[4-128-512-8-128-(13, 17, 67)] PASSED 2025-09-09T15:23:22.2071680Z test/test_ops.py::test_marlin_24[4-128-512-8-128-(26, 37, 13)] PASSED 2025-09-09T15:23:22.2072441Z test/test_ops.py::test_marlin_24[4-128-512-8-128-(67, 13, 11)] PASSED 2025-09-09T15:23:22.2073093Z test/test_ops.py::test_marlin_24[8-128-512-4--1-(1, 1, 1)] PASSED 2025-09-09T15:23:22.2073644Z test/test_ops.py::test_marlin_24[8-128-512-4--1-(1, 4, 8)] PASSED 2025-09-09T15:23:22.2074185Z test/test_ops.py::test_marlin_24[8-128-512-4--1-(1, 7, 5)] PASSED 2025-09-09T15:23:22.2074751Z test/test_ops.py::test_marlin_24[8-128-512-4--1-(13, 17, 67)] PASSED 2025-09-09T15:23:22.2075370Z test/test_ops.py::test_marlin_24[8-128-512-4--1-(26, 37, 13)] PASSED 2025-09-09T15:23:22.2075948Z test/test_ops.py::test_marlin_24[8-128-512-4--1-(67, 13, 11)] PASSED 2025-09-09T15:23:22.2076532Z test/test_ops.py::test_marlin_24[8-128-512-4-128-(1, 1, 1)] PASSED 2025-09-09T15:23:22.2077089Z test/test_ops.py::test_marlin_24[8-128-512-4-128-(1, 4, 8)] PASSED 2025-09-09T15:23:22.2077667Z test/test_ops.py::test_marlin_24[8-128-512-4-128-(1, 7, 5)] PASSED 2025-09-09T15:23:22.2078242Z test/test_ops.py::test_marlin_24[8-128-512-4-128-(13, 17, 67)] PASSED 2025-09-09T15:23:22.2078827Z test/test_ops.py::test_marlin_24[8-128-512-4-128-(26, 37, 13)] PASSED 2025-09-09T15:23:22.2079408Z test/test_ops.py::test_marlin_24[8-128-512-4-128-(67, 13, 11)] PASSED 2025-09-09T15:23:22.2079968Z test/test_ops.py::test_marlin_24[8-128-512-8--1-(1, 1, 1)] PASSED 2025-09-09T15:23:22.2080519Z test/test_ops.py::test_marlin_24[8-128-512-8--1-(1, 4, 8)] PASSED 2025-09-09T15:23:22.2081197Z test/test_ops.py::test_marlin_24[8-128-512-8--1-(1, 7, 5)] PASSED 2025-09-09T15:23:22.2081768Z test/test_ops.py::test_marlin_24[8-128-512-8--1-(13, 17, 67)] PASSED 2025-09-09T15:23:22.2082337Z test/test_ops.py::test_marlin_24[8-128-512-8--1-(26, 37, 13)] PASSED 2025-09-09T15:23:22.2082910Z test/test_ops.py::test_marlin_24[8-128-512-8--1-(67, 13, 11)] PASSED 2025-09-09T15:23:22.2083483Z test/test_ops.py::test_marlin_24[8-128-512-8-128-(1, 1, 1)] PASSED 2025-09-09T15:23:22.2084043Z test/test_ops.py::test_marlin_24[8-128-512-8-128-(1, 4, 8)] PASSED 2025-09-09T15:23:22.2084604Z test/test_ops.py::test_marlin_24[8-128-512-8-128-(1, 7, 5)] PASSED 2025-09-09T15:23:22.2085195Z test/test_ops.py::test_marlin_24[8-128-512-8-128-(13, 17, 67)] PASSED 2025-09-09T15:23:22.2085801Z test/test_ops.py::test_marlin_24[8-128-512-8-128-(26, 37, 13)] PASSED 2025-09-09T15:23:22.2086378Z test/test_ops.py::test_marlin_24[8-128-512-8-128-(67, 13, 11)] PASSED 2025-09-09T15:23:22.2086954Z test/test_ops.py::test_marlin_24[16-128-512-4--1-(1, 1, 1)] PASSED 2025-09-09T15:23:22.2087525Z test/test_ops.py::test_marlin_24[16-128-512-4--1-(1, 4, 8)] PASSED 2025-09-09T15:23:22.2088077Z test/test_ops.py::test_marlin_24[16-128-512-4--1-(1, 7, 5)] PASSED 2025-09-09T15:23:22.2088651Z test/test_ops.py::test_marlin_24[16-128-512-4--1-(13, 17, 67)] PASSED 2025-09-09T15:23:22.2089236Z test/test_ops.py::test_marlin_24[16-128-512-4--1-(26, 37, 13)] PASSED 2025-09-09T15:23:22.2089918Z test/test_ops.py::test_marlin_24[16-128-512-4--1-(67, 13, 11)] PASSED 2025-09-09T15:23:22.2090492Z test/test_ops.py::test_marlin_24[16-128-512-4-128-(1, 1, 1)] PASSED 2025-09-09T15:23:22.2091054Z test/test_ops.py::test_marlin_24[16-128-512-4-128-(1, 4, 8)] PASSED 2025-09-09T15:23:22.2091624Z test/test_ops.py::test_marlin_24[16-128-512-4-128-(1, 7, 5)] PASSED 2025-09-09T15:23:22.2092203Z test/test_ops.py::test_marlin_24[16-128-512-4-128-(13, 17, 67)] PASSED 2025-09-09T15:23:22.2092793Z test/test_ops.py::test_marlin_24[16-128-512-4-128-(26, 37, 13)] PASSED 2025-09-09T15:23:22.2093470Z test/test_ops.py::test_marlin_24[16-128-512-4-128-(67, 13, 11)] PASSED 2025-09-09T15:23:22.2094041Z test/test_ops.py::test_marlin_24[16-128-512-8--1-(1, 1, 1)] PASSED 2025-09-09T15:23:22.2094646Z test/test_ops.py::test_marlin_24[16-128-512-8--1-(1, 4, 8)] PASSED 2025-09-09T15:23:22.2095207Z test/test_ops.py::test_marlin_24[16-128-512-8--1-(1, 7, 5)] PASSED 2025-09-09T15:23:22.2095892Z test/test_ops.py::test_marlin_24[16-128-512-8--1-(13, 17, 67)] PASSED 2025-09-09T15:23:22.2096468Z test/test_ops.py::test_marlin_24[16-128-512-8--1-(26, 37, 13)] PASSED 2025-09-09T15:23:22.2097050Z test/test_ops.py::test_marlin_24[16-128-512-8--1-(67, 13, 11)] PASSED 2025-09-09T15:23:22.2098034Z test/test_ops.py::test_marlin_24[16-128-512-8-128-(1, 1, 1)] PASSED 2025-09-09T15:23:22.2098603Z test/test_ops.py::test_marlin_24[16-128-512-8-128-(1, 4, 8)] PASSED 2025-09-09T15:23:22.2099170Z test/test_ops.py::test_marlin_24[16-128-512-8-128-(1, 7, 5)] PASSED 2025-09-09T15:23:22.2099739Z test/test_ops.py::test_marlin_24[16-128-512-8-128-(13, 17, 67)] PASSED 2025-09-09T15:23:22.2100328Z test/test_ops.py::test_marlin_24[16-128-512-8-128-(26, 37, 13)] PASSED 2025-09-09T15:23:22.2100903Z test/test_ops.py::test_marlin_24[16-128-512-8-128-(67, 13, 11)] PASSED 2025-09-09T15:23:22.2101471Z test/test_ops.py::test_marlin_24[32-128-512-4--1-(1, 1, 1)] PASSED 2025-09-09T15:23:22.2102030Z test/test_ops.py::test_marlin_24[32-128-512-4--1-(1, 4, 8)] PASSED 2025-09-09T15:23:22.2102578Z test/test_ops.py::test_marlin_24[32-128-512-4--1-(1, 7, 5)] PASSED 2025-09-09T15:23:22.2103143Z test/test_ops.py::test_marlin_24[32-128-512-4--1-(13, 17, 67)] PASSED 2025-09-09T15:23:22.2103825Z test/test_ops.py::test_marlin_24[32-128-512-4--1-(26, 37, 13)] PASSED 2025-09-09T15:23:22.2104404Z test/test_ops.py::test_marlin_24[32-128-512-4--1-(67, 13, 11)] PASSED 2025-09-09T15:23:22.2104999Z test/test_ops.py::test_marlin_24[32-128-512-4-128-(1, 1, 1)] PASSED 2025-09-09T15:23:22.2105592Z test/test_ops.py::test_marlin_24[32-128-512-4-128-(1, 4, 8)] PASSED 2025-09-09T15:23:22.2106161Z test/test_ops.py::test_marlin_24[32-128-512-4-128-(1, 7, 5)] PASSED 2025-09-09T15:23:22.2106738Z test/test_ops.py::test_marlin_24[32-128-512-4-128-(13, 17, 67)] PASSED 2025-09-09T15:23:22.2107325Z test/test_ops.py::test_marlin_24[32-128-512-4-128-(26, 37, 13)] PASSED 2025-09-09T15:23:22.2107902Z test/test_ops.py::test_marlin_24[32-128-512-4-128-(67, 13, 11)] PASSED 2025-09-09T15:23:22.2108475Z test/test_ops.py::test_marlin_24[32-128-512-8--1-(1, 1, 1)] PASSED 2025-09-09T15:23:22.2109025Z test/test_ops.py::test_marlin_24[32-128-512-8--1-(1, 4, 8)] PASSED 2025-09-09T15:23:22.2109591Z test/test_ops.py::test_marlin_24[32-128-512-8--1-(1, 7, 5)] PASSED 2025-09-09T15:23:22.2110165Z test/test_ops.py::test_marlin_24[32-128-512-8--1-(13, 17, 67)] PASSED 2025-09-09T15:23:22.2110735Z test/test_ops.py::test_marlin_24[32-128-512-8--1-(26, 37, 13)] PASSED 2025-09-09T15:23:22.2111313Z test/test_ops.py::test_marlin_24[32-128-512-8--1-(67, 13, 11)] PASSED 2025-09-09T15:23:22.2111884Z test/test_ops.py::test_marlin_24[32-128-512-8-128-(1, 1, 1)] PASSED 2025-09-09T15:23:22.2112572Z test/test_ops.py::test_marlin_24[32-128-512-8-128-(1, 4, 8)] PASSED 2025-09-09T15:23:22.2113140Z test/test_ops.py::test_marlin_24[32-128-512-8-128-(1, 7, 5)] PASSED 2025-09-09T15:23:22.2113712Z test/test_ops.py::test_marlin_24[32-128-512-8-128-(13, 17, 67)] PASSED 2025-09-09T15:23:22.2114298Z test/test_ops.py::test_marlin_24[32-128-512-8-128-(26, 37, 13)] PASSED 2025-09-09T15:23:22.2114877Z test/test_ops.py::test_marlin_24[32-128-512-8-128-(67, 13, 11)] PASSED 2025-09-09T15:23:22.2115459Z test/test_ops.py::test_marlin_24[64-128-512-4--1-(1, 1, 1)] PASSED 2025-09-09T15:23:22.2116081Z test/test_ops.py::test_marlin_24[64-128-512-4--1-(1, 4, 8)] PASSED 2025-09-09T15:23:22.2116642Z test/test_ops.py::test_marlin_24[64-128-512-4--1-(1, 7, 5)] PASSED 2025-09-09T15:23:22.2117277Z test/test_ops.py::test_marlin_24[64-128-512-4--1-(13, 17, 67)] PASSED 2025-09-09T15:23:22.2117853Z test/test_ops.py::test_marlin_24[64-128-512-4--1-(26, 37, 13)] PASSED 2025-09-09T15:23:22.2118436Z test/test_ops.py::test_marlin_24[64-128-512-4--1-(67, 13, 11)] PASSED 2025-09-09T15:23:22.2119007Z test/test_ops.py::test_marlin_24[64-128-512-4-128-(1, 1, 1)] PASSED 2025-09-09T15:23:22.2119579Z test/test_ops.py::test_marlin_24[64-128-512-4-128-(1, 4, 8)] PASSED 2025-09-09T15:23:22.2120141Z test/test_ops.py::test_marlin_24[64-128-512-4-128-(1, 7, 5)] PASSED 2025-09-09T15:23:22.2120727Z test/test_ops.py::test_marlin_24[64-128-512-4-128-(13, 17, 67)] PASSED 2025-09-09T15:23:22.2121315Z test/test_ops.py::test_marlin_24[64-128-512-4-128-(26, 37, 13)] PASSED 2025-09-09T15:23:22.2121895Z test/test_ops.py::test_marlin_24[64-128-512-4-128-(67, 13, 11)] PASSED 2025-09-09T15:23:22.2122467Z test/test_ops.py::test_marlin_24[64-128-512-8--1-(1, 1, 1)] PASSED 2025-09-09T15:23:22.2123022Z test/test_ops.py::test_marlin_24[64-128-512-8--1-(1, 4, 8)] PASSED 2025-09-09T15:23:22.2123583Z test/test_ops.py::test_marlin_24[64-128-512-8--1-(1, 7, 5)] PASSED 2025-09-09T15:23:22.2124155Z test/test_ops.py::test_marlin_24[64-128-512-8--1-(13, 17, 67)] PASSED 2025-09-09T15:23:22.2124730Z test/test_ops.py::test_marlin_24[64-128-512-8--1-(26, 37, 13)] PASSED 2025-09-09T15:23:22.2125366Z test/test_ops.py::test_marlin_24[64-128-512-8--1-(67, 13, 11)] PASSED 2025-09-09T15:23:22.2125934Z test/test_ops.py::test_marlin_24[64-128-512-8-128-(1, 1, 1)] PASSED 2025-09-09T15:23:22.2126563Z test/test_ops.py::test_marlin_24[64-128-512-8-128-(1, 4, 8)] PASSED 2025-09-09T15:23:22.2127128Z test/test_ops.py::test_marlin_24[64-128-512-8-128-(1, 7, 5)] PASSED 2025-09-09T15:23:22.2127709Z test/test_ops.py::test_marlin_24[64-128-512-8-128-(13, 17, 67)] PASSED 2025-09-09T15:23:22.2128305Z test/test_ops.py::test_marlin_24[64-128-512-8-128-(26, 37, 13)] PASSED 2025-09-09T15:23:22.2128880Z test/test_ops.py::test_marlin_24[64-128-512-8-128-(67, 13, 11)] PASSED 2025-09-09T15:23:22.2129478Z test/test_ops.py::test_marlin_qqq[1-128-64-4--1-(1, 1, 1)] SKIPPED (...) 2025-09-09T15:23:22.2130081Z test/test_ops.py::test_marlin_qqq[1-128-64-4--1-(1, 4, 8)] SKIPPED (...) 2025-09-09T15:23:22.2130683Z test/test_ops.py::test_marlin_qqq[1-128-64-4--1-(1, 7, 5)] SKIPPED (...) 2025-09-09T15:23:22.2131271Z test/test_ops.py::test_marlin_qqq[1-128-64-4--1-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.2131858Z test/test_ops.py::test_marlin_qqq[1-128-64-4--1-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.2132442Z test/test_ops.py::test_marlin_qqq[1-128-64-4--1-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.2556869Z test/test_ops.py::test_marlin_qqq[1-128-64-4-128-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.2557764Z test/test_ops.py::test_marlin_qqq[1-128-64-4-128-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.2558554Z test/test_ops.py::test_marlin_qqq[1-128-64-4-128-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.2559277Z test/test_ops.py::test_marlin_qqq[1-128-64-4-128-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.2559882Z test/test_ops.py::test_marlin_qqq[1-128-64-4-128-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.2560477Z test/test_ops.py::test_marlin_qqq[1-128-64-4-128-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.2561062Z test/test_ops.py::test_marlin_qqq[1-128-128-4--1-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.2561634Z test/test_ops.py::test_marlin_qqq[1-128-128-4--1-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.2562211Z test/test_ops.py::test_marlin_qqq[1-128-128-4--1-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.2562866Z test/test_ops.py::test_marlin_qqq[1-128-128-4--1-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.2563453Z test/test_ops.py::test_marlin_qqq[1-128-128-4--1-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.2564099Z test/test_ops.py::test_marlin_qqq[1-128-128-4--1-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.2564687Z test/test_ops.py::test_marlin_qqq[1-128-128-4-128-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.2565300Z test/test_ops.py::test_marlin_qqq[1-128-128-4-128-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.2565894Z test/test_ops.py::test_marlin_qqq[1-128-128-4-128-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.2566480Z test/test_ops.py::test_marlin_qqq[1-128-128-4-128-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.2567081Z test/test_ops.py::test_marlin_qqq[1-128-128-4-128-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.2567675Z test/test_ops.py::test_marlin_qqq[1-128-128-4-128-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.2568263Z test/test_ops.py::test_marlin_qqq[1-128-256-4--1-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.2568836Z test/test_ops.py::test_marlin_qqq[1-128-256-4--1-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.2569400Z test/test_ops.py::test_marlin_qqq[1-128-256-4--1-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.2569982Z test/test_ops.py::test_marlin_qqq[1-128-256-4--1-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.2570570Z test/test_ops.py::test_marlin_qqq[1-128-256-4--1-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.2571158Z test/test_ops.py::test_marlin_qqq[1-128-256-4--1-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.2571733Z test/test_ops.py::test_marlin_qqq[1-128-256-4-128-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.2572315Z test/test_ops.py::test_marlin_qqq[1-128-256-4-128-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.2572896Z test/test_ops.py::test_marlin_qqq[1-128-256-4-128-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.2573551Z test/test_ops.py::test_marlin_qqq[1-128-256-4-128-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.2574151Z test/test_ops.py::test_marlin_qqq[1-128-256-4-128-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.2574741Z test/test_ops.py::test_marlin_qqq[1-128-256-4-128-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.2575345Z test/test_ops.py::test_marlin_qqq[4-128-64-4--1-(1, 1, 1)] SKIPPED (...) 2025-09-09T15:23:22.2576022Z test/test_ops.py::test_marlin_qqq[4-128-64-4--1-(1, 4, 8)] SKIPPED (...) 2025-09-09T15:23:22.2576623Z test/test_ops.py::test_marlin_qqq[4-128-64-4--1-(1, 7, 5)] SKIPPED (...) 2025-09-09T15:23:22.2577216Z test/test_ops.py::test_marlin_qqq[4-128-64-4--1-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.2577799Z test/test_ops.py::test_marlin_qqq[4-128-64-4--1-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.2578386Z test/test_ops.py::test_marlin_qqq[4-128-64-4--1-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.2578959Z test/test_ops.py::test_marlin_qqq[4-128-64-4-128-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.2579536Z test/test_ops.py::test_marlin_qqq[4-128-64-4-128-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.2580108Z test/test_ops.py::test_marlin_qqq[4-128-64-4-128-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.2580684Z test/test_ops.py::test_marlin_qqq[4-128-64-4-128-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.2581282Z test/test_ops.py::test_marlin_qqq[4-128-64-4-128-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.2581917Z test/test_ops.py::test_marlin_qqq[4-128-64-4-128-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.2582499Z test/test_ops.py::test_marlin_qqq[4-128-128-4--1-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.2583063Z test/test_ops.py::test_marlin_qqq[4-128-128-4--1-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.2583630Z test/test_ops.py::test_marlin_qqq[4-128-128-4--1-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.2584212Z test/test_ops.py::test_marlin_qqq[4-128-128-4--1-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.2584798Z test/test_ops.py::test_marlin_qqq[4-128-128-4--1-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.2585490Z test/test_ops.py::test_marlin_qqq[4-128-128-4--1-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.2586069Z test/test_ops.py::test_marlin_qqq[4-128-128-4-128-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.2586694Z test/test_ops.py::test_marlin_qqq[4-128-128-4-128-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.2587272Z test/test_ops.py::test_marlin_qqq[4-128-128-4-128-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.2587864Z test/test_ops.py::test_marlin_qqq[4-128-128-4-128-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.2588463Z test/test_ops.py::test_marlin_qqq[4-128-128-4-128-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.2589055Z test/test_ops.py::test_marlin_qqq[4-128-128-4-128-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.2589640Z test/test_ops.py::test_marlin_qqq[4-128-256-4--1-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.2590210Z test/test_ops.py::test_marlin_qqq[4-128-256-4--1-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.2590785Z test/test_ops.py::test_marlin_qqq[4-128-256-4--1-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.2591367Z test/test_ops.py::test_marlin_qqq[4-128-256-4--1-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.2591954Z test/test_ops.py::test_marlin_qqq[4-128-256-4--1-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.2592546Z test/test_ops.py::test_marlin_qqq[4-128-256-4--1-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.2593129Z test/test_ops.py::test_marlin_qqq[4-128-256-4-128-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.2593711Z test/test_ops.py::test_marlin_qqq[4-128-256-4-128-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.2594290Z test/test_ops.py::test_marlin_qqq[4-128-256-4-128-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.2594877Z test/test_ops.py::test_marlin_qqq[4-128-256-4-128-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.2595582Z test/test_ops.py::test_marlin_qqq[4-128-256-4-128-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.2596174Z test/test_ops.py::test_marlin_qqq[4-128-256-4-128-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.2596772Z test/test_ops.py::test_marlin_qqq[8-128-64-4--1-(1, 1, 1)] SKIPPED (...) 2025-09-09T15:23:22.2597535Z test/test_ops.py::test_marlin_qqq[8-128-64-4--1-(1, 4, 8)] SKIPPED (...) 2025-09-09T15:23:22.2598146Z test/test_ops.py::test_marlin_qqq[8-128-64-4--1-(1, 7, 5)] SKIPPED (...) 2025-09-09T15:23:22.2598743Z test/test_ops.py::test_marlin_qqq[8-128-64-4--1-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.2599323Z test/test_ops.py::test_marlin_qqq[8-128-64-4--1-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.2599908Z test/test_ops.py::test_marlin_qqq[8-128-64-4--1-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.2600493Z test/test_ops.py::test_marlin_qqq[8-128-64-4-128-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.2601064Z test/test_ops.py::test_marlin_qqq[8-128-64-4-128-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.2601637Z test/test_ops.py::test_marlin_qqq[8-128-64-4-128-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.2602221Z test/test_ops.py::test_marlin_qqq[8-128-64-4-128-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.2602816Z test/test_ops.py::test_marlin_qqq[8-128-64-4-128-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.2603404Z test/test_ops.py::test_marlin_qqq[8-128-64-4-128-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.2603984Z test/test_ops.py::test_marlin_qqq[8-128-128-4--1-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.2604666Z test/test_ops.py::test_marlin_qqq[8-128-128-4--1-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.2605240Z test/test_ops.py::test_marlin_qqq[8-128-128-4--1-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.2605827Z test/test_ops.py::test_marlin_qqq[8-128-128-4--1-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.2606414Z test/test_ops.py::test_marlin_qqq[8-128-128-4--1-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.2607001Z test/test_ops.py::test_marlin_qqq[8-128-128-4--1-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.2607573Z test/test_ops.py::test_marlin_qqq[8-128-128-4-128-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.2608224Z test/test_ops.py::test_marlin_qqq[8-128-128-4-128-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.2608806Z test/test_ops.py::test_marlin_qqq[8-128-128-4-128-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.2609453Z test/test_ops.py::test_marlin_qqq[8-128-128-4-128-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.2610056Z test/test_ops.py::test_marlin_qqq[8-128-128-4-128-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.2610642Z test/test_ops.py::test_marlin_qqq[8-128-128-4-128-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.2611225Z test/test_ops.py::test_marlin_qqq[8-128-256-4--1-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.2611792Z test/test_ops.py::test_marlin_qqq[8-128-256-4--1-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.2612363Z test/test_ops.py::test_marlin_qqq[8-128-256-4--1-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.2612947Z test/test_ops.py::test_marlin_qqq[8-128-256-4--1-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.2613534Z test/test_ops.py::test_marlin_qqq[8-128-256-4--1-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.2614121Z test/test_ops.py::test_marlin_qqq[8-128-256-4--1-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.2614704Z test/test_ops.py::test_marlin_qqq[8-128-256-4-128-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.2615334Z test/test_ops.py::test_marlin_qqq[8-128-256-4-128-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.2616080Z test/test_ops.py::test_marlin_qqq[8-128-256-4-128-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.2616674Z test/test_ops.py::test_marlin_qqq[8-128-256-4-128-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.2617272Z test/test_ops.py::test_marlin_qqq[8-128-256-4-128-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.2617958Z test/test_ops.py::test_marlin_qqq[8-128-256-4-128-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.2618681Z test/test_ops.py::test_marlin_qqq[16-128-64-4--1-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.3033145Z test/test_ops.py::test_marlin_qqq[16-128-64-4--1-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.3033897Z test/test_ops.py::test_marlin_qqq[16-128-64-4--1-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.3034571Z test/test_ops.py::test_marlin_qqq[16-128-64-4--1-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.3035191Z test/test_ops.py::test_marlin_qqq[16-128-64-4--1-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.3035973Z test/test_ops.py::test_marlin_qqq[16-128-64-4--1-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.3036718Z test/test_ops.py::test_marlin_qqq[16-128-64-4-128-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.3037310Z test/test_ops.py::test_marlin_qqq[16-128-64-4-128-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.3037971Z test/test_ops.py::test_marlin_qqq[16-128-64-4-128-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.3038619Z test/test_ops.py::test_marlin_qqq[16-128-64-4-128-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.3039254Z test/test_ops.py::test_marlin_qqq[16-128-64-4-128-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.3039946Z test/test_ops.py::test_marlin_qqq[16-128-64-4-128-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.3040609Z test/test_ops.py::test_marlin_qqq[16-128-128-4--1-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.3041251Z test/test_ops.py::test_marlin_qqq[16-128-128-4--1-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.3041879Z test/test_ops.py::test_marlin_qqq[16-128-128-4--1-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.3042837Z test/test_ops.py::test_marlin_qqq[16-128-128-4--1-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.3043530Z test/test_ops.py::test_marlin_qqq[16-128-128-4--1-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.3044194Z test/test_ops.py::test_marlin_qqq[16-128-128-4--1-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.3044829Z test/test_ops.py::test_marlin_qqq[16-128-128-4-128-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.3045455Z test/test_ops.py::test_marlin_qqq[16-128-128-4-128-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.3046165Z test/test_ops.py::test_marlin_qqq[16-128-128-4-128-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.3046965Z test/test_ops.py::test_marlin_qqq[16-128-128-4-128-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.3047673Z test/test_ops.py::test_marlin_qqq[16-128-128-4-128-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.3048469Z test/test_ops.py::test_marlin_qqq[16-128-128-4-128-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.3049076Z test/test_ops.py::test_marlin_qqq[16-128-256-4--1-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.3049742Z test/test_ops.py::test_marlin_qqq[16-128-256-4--1-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.3050395Z test/test_ops.py::test_marlin_qqq[16-128-256-4--1-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.3051007Z test/test_ops.py::test_marlin_qqq[16-128-256-4--1-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.3051654Z test/test_ops.py::test_marlin_qqq[16-128-256-4--1-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.3052329Z test/test_ops.py::test_marlin_qqq[16-128-256-4--1-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.3053014Z test/test_ops.py::test_marlin_qqq[16-128-256-4-128-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.3053628Z test/test_ops.py::test_marlin_qqq[16-128-256-4-128-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.3054285Z test/test_ops.py::test_marlin_qqq[16-128-256-4-128-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.3054970Z test/test_ops.py::test_marlin_qqq[16-128-256-4-128-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.3055656Z test/test_ops.py::test_marlin_qqq[16-128-256-4-128-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.3056365Z test/test_ops.py::test_marlin_qqq[16-128-256-4-128-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.3056996Z test/test_ops.py::test_marlin_qqq[32-128-64-4--1-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.3057658Z test/test_ops.py::test_marlin_qqq[32-128-64-4--1-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.3058441Z test/test_ops.py::test_marlin_qqq[32-128-64-4--1-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.3059115Z test/test_ops.py::test_marlin_qqq[32-128-64-4--1-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.3059734Z test/test_ops.py::test_marlin_qqq[32-128-64-4--1-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.3060410Z test/test_ops.py::test_marlin_qqq[32-128-64-4--1-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.3061077Z test/test_ops.py::test_marlin_qqq[32-128-64-4-128-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.3061698Z test/test_ops.py::test_marlin_qqq[32-128-64-4-128-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.3062329Z test/test_ops.py::test_marlin_qqq[32-128-64-4-128-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.3062999Z test/test_ops.py::test_marlin_qqq[32-128-64-4-128-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.3063671Z test/test_ops.py::test_marlin_qqq[32-128-64-4-128-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.3064289Z test/test_ops.py::test_marlin_qqq[32-128-64-4-128-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.3064931Z test/test_ops.py::test_marlin_qqq[32-128-128-4--1-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.3065623Z test/test_ops.py::test_marlin_qqq[32-128-128-4--1-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.3066308Z test/test_ops.py::test_marlin_qqq[32-128-128-4--1-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.3066909Z test/test_ops.py::test_marlin_qqq[32-128-128-4--1-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.3067570Z test/test_ops.py::test_marlin_qqq[32-128-128-4--1-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.3068330Z test/test_ops.py::test_marlin_qqq[32-128-128-4--1-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.3068995Z test/test_ops.py::test_marlin_qqq[32-128-128-4-128-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.3069604Z test/test_ops.py::test_marlin_qqq[32-128-128-4-128-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.3070256Z test/test_ops.py::test_marlin_qqq[32-128-128-4-128-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.3070938Z test/test_ops.py::test_marlin_qqq[32-128-128-4-128-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.3071687Z test/test_ops.py::test_marlin_qqq[32-128-128-4-128-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.3072350Z test/test_ops.py::test_marlin_qqq[32-128-128-4-128-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.3073022Z test/test_ops.py::test_marlin_qqq[32-128-256-4--1-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.3073697Z test/test_ops.py::test_marlin_qqq[32-128-256-4--1-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.3074369Z test/test_ops.py::test_marlin_qqq[32-128-256-4--1-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.3074983Z test/test_ops.py::test_marlin_qqq[32-128-256-4--1-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.3075665Z test/test_ops.py::test_marlin_qqq[32-128-256-4--1-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.3076357Z test/test_ops.py::test_marlin_qqq[32-128-256-4--1-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.3077027Z test/test_ops.py::test_marlin_qqq[32-128-256-4-128-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.3077641Z test/test_ops.py::test_marlin_qqq[32-128-256-4-128-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.3078287Z test/test_ops.py::test_marlin_qqq[32-128-256-4-128-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.3078971Z test/test_ops.py::test_marlin_qqq[32-128-256-4-128-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.3079659Z test/test_ops.py::test_marlin_qqq[32-128-256-4-128-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.3080301Z test/test_ops.py::test_marlin_qqq[32-128-256-4-128-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.3080934Z test/test_ops.py::test_marlin_qqq[64-128-64-4--1-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.3081591Z test/test_ops.py::test_marlin_qqq[64-128-64-4--1-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.3082232Z test/test_ops.py::test_marlin_qqq[64-128-64-4--1-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.3082820Z test/test_ops.py::test_marlin_qqq[64-128-64-4--1-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.3083562Z test/test_ops.py::test_marlin_qqq[64-128-64-4--1-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.3084153Z test/test_ops.py::test_marlin_qqq[64-128-64-4--1-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.3084821Z test/test_ops.py::test_marlin_qqq[64-128-64-4-128-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.3085405Z test/test_ops.py::test_marlin_qqq[64-128-64-4-128-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.3085976Z test/test_ops.py::test_marlin_qqq[64-128-64-4-128-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.3086560Z test/test_ops.py::test_marlin_qqq[64-128-64-4-128-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.3087238Z test/test_ops.py::test_marlin_qqq[64-128-64-4-128-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.3087832Z test/test_ops.py::test_marlin_qqq[64-128-64-4-128-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.3088411Z test/test_ops.py::test_marlin_qqq[64-128-128-4--1-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.3088995Z test/test_ops.py::test_marlin_qqq[64-128-128-4--1-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.3089581Z test/test_ops.py::test_marlin_qqq[64-128-128-4--1-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.3090172Z test/test_ops.py::test_marlin_qqq[64-128-128-4--1-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.3090773Z test/test_ops.py::test_marlin_qqq[64-128-128-4--1-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.3091443Z test/test_ops.py::test_marlin_qqq[64-128-128-4--1-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.3093579Z test/test_ops.py::test_marlin_qqq[64-128-128-4-128-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.3094159Z test/test_ops.py::test_marlin_qqq[64-128-128-4-128-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.3094755Z test/test_ops.py::test_marlin_qqq[64-128-128-4-128-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.3095340Z test/test_ops.py::test_marlin_qqq[64-128-128-4-128-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.3096047Z test/test_ops.py::test_marlin_qqq[64-128-128-4-128-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.3096655Z test/test_ops.py::test_marlin_qqq[64-128-128-4-128-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.3106978Z test/test_ops.py::test_marlin_qqq[64-128-256-4--1-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.3107673Z test/test_ops.py::test_marlin_qqq[64-128-256-4--1-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.3108332Z test/test_ops.py::test_marlin_qqq[64-128-256-4--1-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.3108922Z test/test_ops.py::test_marlin_qqq[64-128-256-4--1-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.3109508Z test/test_ops.py::test_marlin_qqq[64-128-256-4--1-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.3110098Z test/test_ops.py::test_marlin_qqq[64-128-256-4--1-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.3110686Z test/test_ops.py::test_marlin_qqq[64-128-256-4-128-(1, 1, 1)] SKIPPED 2025-09-09T15:23:22.3476893Z test/test_ops.py::test_marlin_qqq[64-128-256-4-128-(1, 4, 8)] SKIPPED 2025-09-09T15:23:22.3477584Z test/test_ops.py::test_marlin_qqq[64-128-256-4-128-(1, 7, 5)] SKIPPED 2025-09-09T15:23:22.3478186Z test/test_ops.py::test_marlin_qqq[64-128-256-4-128-(13, 17, 67)] SKIPPED 2025-09-09T15:23:22.3478791Z test/test_ops.py::test_marlin_qqq[64-128-256-4-128-(26, 37, 13)] SKIPPED 2025-09-09T15:23:22.3479407Z test/test_ops.py::test_marlin_qqq[64-128-256-4-128-(67, 13, 11)] SKIPPED 2025-09-09T15:23:22.3479985Z test/test_ops.py::test_swizzle_mm SKIPPED (ROCm not available) 2025-09-09T15:23:22.3480603Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1-1-torch.int64] SKIPPED 2025-09-09T15:23:22.3481255Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1-1-torch.int32] SKIPPED 2025-09-09T15:23:22.3481905Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1-128-torch.int64] SKIPPED 2025-09-09T15:23:22.3482571Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1-128-torch.int32] SKIPPED 2025-09-09T15:23:22.3483338Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1-512-torch.int64] SKIPPED 2025-09-09T15:23:22.3484010Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1-512-torch.int32] SKIPPED 2025-09-09T15:23:22.3484658Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-2-1-torch.int64] SKIPPED 2025-09-09T15:23:22.3485310Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-2-1-torch.int32] SKIPPED 2025-09-09T15:23:22.3486022Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-2-128-torch.int64] SKIPPED 2025-09-09T15:23:22.3486676Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-2-128-torch.int32] SKIPPED 2025-09-09T15:23:22.3487340Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-2-512-torch.int64] SKIPPED 2025-09-09T15:23:22.3487995Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-2-512-torch.int32] SKIPPED 2025-09-09T15:23:22.3488660Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-128-1-torch.int64] SKIPPED 2025-09-09T15:23:22.3489315Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-128-1-torch.int32] SKIPPED 2025-09-09T15:23:22.3489976Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-128-128-torch.int64] SKIPPED 2025-09-09T15:23:22.3490645Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-128-128-torch.int32] SKIPPED 2025-09-09T15:23:22.3491306Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-128-512-torch.int64] SKIPPED 2025-09-09T15:23:22.3491975Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-128-512-torch.int32] SKIPPED 2025-09-09T15:23:22.3492715Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1024-1-torch.int64] SKIPPED 2025-09-09T15:23:22.3493365Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1024-1-torch.int32] SKIPPED 2025-09-09T15:23:22.3494037Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1024-128-torch.int64] SKIPPED 2025-09-09T15:23:22.3494700Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1024-128-torch.int32] SKIPPED 2025-09-09T15:23:22.3495373Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1024-512-torch.int64] SKIPPED 2025-09-09T15:23:22.3496186Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1024-512-torch.int32] SKIPPED 2025-09-09T15:23:22.3496958Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1-1-torch.int64] SKIPPED 2025-09-09T15:23:22.3497853Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1-1-torch.int32] SKIPPED 2025-09-09T15:23:22.3498493Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1-128-torch.int64] SKIPPED 2025-09-09T15:23:22.3499143Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1-128-torch.int32] SKIPPED 2025-09-09T15:23:22.3499787Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1-512-torch.int64] SKIPPED 2025-09-09T15:23:22.3500434Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1-512-torch.int32] SKIPPED 2025-09-09T15:23:22.3501078Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-2-1-torch.int64] SKIPPED 2025-09-09T15:23:22.3501718Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-2-1-torch.int32] SKIPPED 2025-09-09T15:23:22.3502374Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-2-128-torch.int64] SKIPPED 2025-09-09T15:23:22.3503019Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-2-128-torch.int32] SKIPPED 2025-09-09T15:23:22.3503671Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-2-512-torch.int64] SKIPPED 2025-09-09T15:23:22.3504316Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-2-512-torch.int32] SKIPPED 2025-09-09T15:23:22.3504967Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-128-1-torch.int64] SKIPPED 2025-09-09T15:23:22.3505623Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-128-1-torch.int32] SKIPPED 2025-09-09T15:23:22.3506284Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-128-128-torch.int64] SKIPPED 2025-09-09T15:23:22.3506949Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-128-128-torch.int32] SKIPPED 2025-09-09T15:23:22.3507681Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-128-512-torch.int64] SKIPPED 2025-09-09T15:23:22.3508350Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-128-512-torch.int32] SKIPPED 2025-09-09T15:23:22.3509008Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1024-1-torch.int64] SKIPPED 2025-09-09T15:23:22.3509666Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1024-1-torch.int32] SKIPPED 2025-09-09T15:23:22.3510333Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1024-128-torch.int64] SKIPPED 2025-09-09T15:23:22.3511011Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1024-128-torch.int32] SKIPPED 2025-09-09T15:23:22.3511689Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1024-512-torch.int64] SKIPPED 2025-09-09T15:23:22.3512355Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1024-512-torch.int32] SKIPPED 2025-09-09T15:23:22.3513019Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1-1-torch.int64] SKIPPED 2025-09-09T15:23:22.3513665Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1-1-torch.int32] SKIPPED 2025-09-09T15:23:22.3514309Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1-128-torch.int64] SKIPPED 2025-09-09T15:23:22.3514983Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1-128-torch.int32] SKIPPED 2025-09-09T15:23:22.3515666Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1-512-torch.int64] SKIPPED 2025-09-09T15:23:22.3516390Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1-512-torch.int32] SKIPPED 2025-09-09T15:23:22.3517036Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-2-1-torch.int64] SKIPPED 2025-09-09T15:23:22.3517673Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-2-1-torch.int32] SKIPPED 2025-09-09T15:23:22.3518322Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-2-128-torch.int64] SKIPPED 2025-09-09T15:23:22.3518977Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-2-128-torch.int32] SKIPPED 2025-09-09T15:23:22.3519638Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-2-512-torch.int64] SKIPPED 2025-09-09T15:23:22.3520352Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-2-512-torch.int32] SKIPPED 2025-09-09T15:23:22.3520996Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-128-1-torch.int64] SKIPPED 2025-09-09T15:23:22.3521707Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-128-1-torch.int32] SKIPPED 2025-09-09T15:23:22.3522371Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-128-128-torch.int64] SKIPPED 2025-09-09T15:23:22.3523048Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-128-128-torch.int32] SKIPPED 2025-09-09T15:23:22.3523715Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-128-512-torch.int64] SKIPPED 2025-09-09T15:23:22.3524392Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-128-512-torch.int32] SKIPPED 2025-09-09T15:23:22.3525066Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1024-1-torch.int64] SKIPPED 2025-09-09T15:23:22.3525777Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1024-1-torch.int32] SKIPPED 2025-09-09T15:23:22.3526460Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1024-128-torch.int64] SKIPPED 2025-09-09T15:23:22.3527141Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1024-128-torch.int32] SKIPPED 2025-09-09T15:23:22.3527827Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1024-512-torch.int64] SKIPPED 2025-09-09T15:23:22.3528510Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1024-512-torch.int32] SKIPPED 2025-09-09T15:23:22.3529170Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1-1-torch.int64] SKIPPED 2025-09-09T15:23:22.3529824Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1-1-torch.int32] SKIPPED 2025-09-09T15:23:22.3530477Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1-128-torch.int64] SKIPPED 2025-09-09T15:23:22.3531194Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1-128-torch.int32] SKIPPED 2025-09-09T15:23:22.3531854Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1-512-torch.int64] SKIPPED 2025-09-09T15:23:22.3532513Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1-512-torch.int32] SKIPPED 2025-09-09T15:23:22.3533167Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-2-1-torch.int64] SKIPPED 2025-09-09T15:23:22.3533814Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-2-1-torch.int32] SKIPPED 2025-09-09T15:23:22.3534466Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-2-128-torch.int64] SKIPPED 2025-09-09T15:23:22.3535125Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-2-128-torch.int32] SKIPPED 2025-09-09T15:23:22.3535834Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-2-512-torch.int64] SKIPPED 2025-09-09T15:23:22.3536506Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-2-512-torch.int32] SKIPPED 2025-09-09T15:23:22.3537168Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-128-1-torch.int64] SKIPPED 2025-09-09T15:23:22.3537835Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-128-1-torch.int32] SKIPPED 2025-09-09T15:23:22.3538498Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-128-128-torch.int64] SKIPPED 2025-09-09T15:23:22.3539183Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-128-128-torch.int32] SKIPPED 2025-09-09T15:23:22.3539851Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-128-512-torch.int64] SKIPPED 2025-09-09T15:23:22.3776803Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-128-512-torch.int32] SKIPPED 2025-09-09T15:23:22.3777645Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1024-1-torch.int64] SKIPPED 2025-09-09T15:23:22.3778415Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1024-1-torch.int32] SKIPPED 2025-09-09T15:23:22.3779203Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1024-128-torch.int64] SKIPPED 2025-09-09T15:23:22.3779986Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1024-128-torch.int32] SKIPPED 2025-09-09T15:23:22.3780885Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1024-512-torch.int64] SKIPPED 2025-09-09T15:23:22.3781697Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1024-512-torch.int32] SKIPPED 2025-09-09T15:23:22.3782754Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype0-1-size_mnk0-False] SKIPPED 2025-09-09T15:23:22.3783878Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype1-1-size_mnk1-True] SKIPPED 2025-09-09T15:23:22.3785039Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype2-1-size_mnk2-False] SKIPPED 2025-09-09T15:23:22.3786150Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype3-1-size_mnk3-True] SKIPPED 2025-09-09T15:23:22.3787261Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype4-1-size_mnk4-False] SKIPPED 2025-09-09T15:23:22.3788375Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype5-1-size_mnk5-True] SKIPPED 2025-09-09T15:23:22.3789552Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype6-1-size_mnk6-False] SKIPPED 2025-09-09T15:23:22.3790661Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype7-1-size_mnk7-True] SKIPPED 2025-09-09T15:23:22.3791744Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype8-1-size_mnk8-False] SKIPPED 2025-09-09T15:23:22.3792844Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype9-1-size_mnk9-True] SKIPPED 2025-09-09T15:23:22.3794107Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype10-1-size_mnk10-False] SKIPPED 2025-09-09T15:23:22.3795244Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype11-1-size_mnk11-True] SKIPPED 2025-09-09T15:23:22.3796280Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype12-4-size_mnk12-False] SKIPPED 2025-09-09T15:23:22.3797582Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype13-4-size_mnk13-True] SKIPPED 2025-09-09T15:23:22.3798618Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype14-4-size_mnk14-False] SKIPPED 2025-09-09T15:23:22.3799643Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype15-4-size_mnk15-True] SKIPPED 2025-09-09T15:23:22.3800672Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype16-4-size_mnk16-False] SKIPPED 2025-09-09T15:23:22.3801799Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype17-4-size_mnk17-True] SKIPPED 2025-09-09T15:23:22.3802835Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype18-4-size_mnk18-False] SKIPPED 2025-09-09T15:23:22.3803852Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype19-4-size_mnk19-True] SKIPPED 2025-09-09T15:23:22.3804989Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype20-4-size_mnk20-False] SKIPPED 2025-09-09T15:23:22.3806049Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype21-4-size_mnk21-True] SKIPPED 2025-09-09T15:23:22.3807065Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype22-4-size_mnk22-False] SKIPPED 2025-09-09T15:23:22.3808191Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype23-4-size_mnk23-True] SKIPPED 2025-09-09T15:23:22.3809289Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype24-8-size_mnk24-False] SKIPPED 2025-09-09T15:23:22.3810312Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype25-8-size_mnk25-True] SKIPPED 2025-09-09T15:23:22.3811383Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype26-8-size_mnk26-False] SKIPPED 2025-09-09T15:23:22.3812388Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype27-8-size_mnk27-True] SKIPPED 2025-09-09T15:23:22.3813401Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype28-8-size_mnk28-False] SKIPPED 2025-09-09T15:23:22.3814407Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype29-8-size_mnk29-True] SKIPPED 2025-09-09T15:23:22.3815429Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype30-8-size_mnk30-False] SKIPPED 2025-09-09T15:23:22.3816505Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype31-8-size_mnk31-True] SKIPPED 2025-09-09T15:23:22.3817524Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype32-8-size_mnk32-False] SKIPPED 2025-09-09T15:23:22.3818539Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype33-8-size_mnk33-True] SKIPPED 2025-09-09T15:23:22.3819548Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype34-8-size_mnk34-False] SKIPPED 2025-09-09T15:23:22.3820564Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype35-8-size_mnk35-True] SKIPPED 2025-09-09T15:23:22.3821778Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype36-16-size_mnk36-False] SKIPPED 2025-09-09T15:23:22.3822836Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype37-16-size_mnk37-True] SKIPPED 2025-09-09T15:23:22.3823874Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype38-16-size_mnk38-False] SKIPPED 2025-09-09T15:23:22.3824908Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype39-16-size_mnk39-True] SKIPPED 2025-09-09T15:23:22.3825931Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype40-16-size_mnk40-False] SKIPPED 2025-09-09T15:23:22.3826958Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype41-16-size_mnk41-True] SKIPPED 2025-09-09T15:23:22.3827978Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype42-16-size_mnk42-False] SKIPPED 2025-09-09T15:23:22.3829017Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype43-16-size_mnk43-True] SKIPPED 2025-09-09T15:23:22.3830046Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype44-16-size_mnk44-False] SKIPPED 2025-09-09T15:23:22.3831071Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype45-16-size_mnk45-True] SKIPPED 2025-09-09T15:23:22.3832158Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype46-16-size_mnk46-False] SKIPPED 2025-09-09T15:23:22.3833194Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype47-16-size_mnk47-True] SKIPPED 2025-09-09T15:23:22.3834221Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype48-32-size_mnk48-False] SKIPPED 2025-09-09T15:23:22.3835309Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype49-32-size_mnk49-True] SKIPPED 2025-09-09T15:23:22.3836328Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype50-32-size_mnk50-False] SKIPPED 2025-09-09T15:23:22.3837398Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype51-32-size_mnk51-True] SKIPPED 2025-09-09T15:23:22.3838444Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype52-32-size_mnk52-False] SKIPPED 2025-09-09T15:23:22.3839466Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype53-32-size_mnk53-True] SKIPPED 2025-09-09T15:23:22.3840511Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype54-32-size_mnk54-False] SKIPPED 2025-09-09T15:23:22.3841537Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype55-32-size_mnk55-True] SKIPPED 2025-09-09T15:23:22.3842570Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype56-32-size_mnk56-False] SKIPPED 2025-09-09T15:23:22.4063564Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype57-32-size_mnk57-True] SKIPPED 2025-09-09T15:23:22.4064854Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype58-32-size_mnk58-False] SKIPPED 2025-09-09T15:23:22.4066104Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype59-32-size_mnk59-True] SKIPPED 2025-09-09T15:23:22.4067269Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype60-64-size_mnk60-False] SKIPPED 2025-09-09T15:23:22.4068399Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype61-64-size_mnk61-True] SKIPPED 2025-09-09T15:23:22.4069452Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype62-64-size_mnk62-False] SKIPPED 2025-09-09T15:23:22.4070494Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype63-64-size_mnk63-True] SKIPPED 2025-09-09T15:23:22.4071524Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype64-64-size_mnk64-False] SKIPPED 2025-09-09T15:23:22.4072563Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype65-64-size_mnk65-True] SKIPPED 2025-09-09T15:23:22.4073603Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype66-64-size_mnk66-False] SKIPPED 2025-09-09T15:23:22.4074639Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype67-64-size_mnk67-True] SKIPPED 2025-09-09T15:23:22.4075685Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype68-64-size_mnk68-False] SKIPPED 2025-09-09T15:23:22.4076831Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype69-64-size_mnk69-True] SKIPPED 2025-09-09T15:23:22.4077873Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype70-64-size_mnk70-False] SKIPPED 2025-09-09T15:23:22.4078991Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype71-64-size_mnk71-True] SKIPPED 2025-09-09T15:23:22.4080020Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype72-1-size_mnk72-False] SKIPPED 2025-09-09T15:23:22.4081051Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype73-1-size_mnk73-True] SKIPPED 2025-09-09T15:23:22.4082134Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype74-1-size_mnk74-False] SKIPPED 2025-09-09T15:23:22.4083161Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype75-1-size_mnk75-True] SKIPPED 2025-09-09T15:23:22.4084242Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype76-1-size_mnk76-False] SKIPPED 2025-09-09T15:23:22.4085265Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype77-1-size_mnk77-True] SKIPPED 2025-09-09T15:23:22.4086292Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype78-1-size_mnk78-False] SKIPPED 2025-09-09T15:23:22.4087305Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype79-1-size_mnk79-True] SKIPPED 2025-09-09T15:23:22.4088333Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype80-1-size_mnk80-False] SKIPPED 2025-09-09T15:23:22.4089354Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype81-1-size_mnk81-True] SKIPPED 2025-09-09T15:23:22.4090368Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype82-1-size_mnk82-False] SKIPPED 2025-09-09T15:23:22.4091393Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype83-1-size_mnk83-True] SKIPPED 2025-09-09T15:23:22.4092412Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype84-4-size_mnk84-False] SKIPPED 2025-09-09T15:23:22.4093435Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype85-4-size_mnk85-True] SKIPPED 2025-09-09T15:23:22.4094514Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype86-4-size_mnk86-False] SKIPPED 2025-09-09T15:23:22.4095532Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype87-4-size_mnk87-True] SKIPPED 2025-09-09T15:23:22.4096663Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype88-4-size_mnk88-False] SKIPPED 2025-09-09T15:23:22.4097836Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype89-4-size_mnk89-True] SKIPPED 2025-09-09T15:23:22.4098861Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype90-4-size_mnk90-False] SKIPPED 2025-09-09T15:23:22.4099888Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype91-4-size_mnk91-True] SKIPPED 2025-09-09T15:23:22.4100908Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype92-4-size_mnk92-False] SKIPPED 2025-09-09T15:23:22.4101936Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype93-4-size_mnk93-True] SKIPPED 2025-09-09T15:23:22.4103064Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype94-4-size_mnk94-False] SKIPPED 2025-09-09T15:23:22.4104077Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype95-4-size_mnk95-True] SKIPPED 2025-09-09T15:23:22.4105186Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype96-8-size_mnk96-False] SKIPPED 2025-09-09T15:23:22.4106201Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype97-8-size_mnk97-True] SKIPPED 2025-09-09T15:23:22.4107208Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype98-8-size_mnk98-False] SKIPPED 2025-09-09T15:23:22.4108225Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype99-8-size_mnk99-True] SKIPPED 2025-09-09T15:23:22.4109312Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype100-8-size_mnk100-False] SKIPPED 2025-09-09T15:23:22.4110426Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype101-8-size_mnk101-True] SKIPPED 2025-09-09T15:23:22.4111464Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype102-8-size_mnk102-False] SKIPPED 2025-09-09T15:23:22.4112485Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype103-8-size_mnk103-True] SKIPPED 2025-09-09T15:23:22.4113520Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype104-8-size_mnk104-False] SKIPPED 2025-09-09T15:23:22.4114548Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype105-8-size_mnk105-True] SKIPPED 2025-09-09T15:23:22.4115587Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype106-8-size_mnk106-False] SKIPPED 2025-09-09T15:23:22.4116620Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype107-8-size_mnk107-True] SKIPPED 2025-09-09T15:23:22.4117661Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype108-16-size_mnk108-False] SKIPPED 2025-09-09T15:23:22.4118707Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype109-16-size_mnk109-True] SKIPPED 2025-09-09T15:23:22.4119753Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype110-16-size_mnk110-False] SKIPPED 2025-09-09T15:23:22.4120879Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype111-16-size_mnk111-True] SKIPPED 2025-09-09T15:23:22.4121932Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype112-16-size_mnk112-False] SKIPPED 2025-09-09T15:23:22.4122968Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype113-16-size_mnk113-True] SKIPPED 2025-09-09T15:23:22.4124022Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype114-16-size_mnk114-False] SKIPPED 2025-09-09T15:23:22.4125082Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype115-16-size_mnk115-True] SKIPPED 2025-09-09T15:23:22.4126168Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype116-16-size_mnk116-False] SKIPPED 2025-09-09T15:23:22.4127222Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype117-16-size_mnk117-True] SKIPPED 2025-09-09T15:23:22.4347257Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype118-16-size_mnk118-False] SKIPPED 2025-09-09T15:23:22.4348441Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype119-16-size_mnk119-True] SKIPPED 2025-09-09T15:23:22.4349494Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype120-32-size_mnk120-False] SKIPPED 2025-09-09T15:23:22.4350654Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype121-32-size_mnk121-True] SKIPPED 2025-09-09T15:23:22.4351682Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype122-32-size_mnk122-False] SKIPPED 2025-09-09T15:23:22.4352723Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype123-32-size_mnk123-True] SKIPPED 2025-09-09T15:23:22.4353758Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype124-32-size_mnk124-False] SKIPPED 2025-09-09T15:23:22.4354865Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype125-32-size_mnk125-True] SKIPPED 2025-09-09T15:23:22.4355969Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype126-32-size_mnk126-False] SKIPPED 2025-09-09T15:23:22.4357006Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype127-32-size_mnk127-True] SKIPPED 2025-09-09T15:23:22.4358046Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype128-32-size_mnk128-False] SKIPPED 2025-09-09T15:23:22.4359090Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype129-32-size_mnk129-True] SKIPPED 2025-09-09T15:23:22.4360132Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype130-32-size_mnk130-False] SKIPPED 2025-09-09T15:23:22.4361173Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype131-32-size_mnk131-True] SKIPPED 2025-09-09T15:23:22.4362389Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype132-64-size_mnk132-False] SKIPPED 2025-09-09T15:23:22.4363443Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype133-64-size_mnk133-True] SKIPPED 2025-09-09T15:23:22.4364489Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype134-64-size_mnk134-False] SKIPPED 2025-09-09T15:23:22.4365522Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype135-64-size_mnk135-True] SKIPPED 2025-09-09T15:23:22.4366635Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype136-64-size_mnk136-False] SKIPPED 2025-09-09T15:23:22.4367682Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype137-64-size_mnk137-True] SKIPPED 2025-09-09T15:23:22.4368714Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype138-64-size_mnk138-False] SKIPPED 2025-09-09T15:23:22.4369759Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype139-64-size_mnk139-True] SKIPPED 2025-09-09T15:23:22.4370795Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype140-64-size_mnk140-False] SKIPPED 2025-09-09T15:23:22.4371841Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype141-64-size_mnk141-True] SKIPPED 2025-09-09T15:23:22.4372890Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype142-64-size_mnk142-False] SKIPPED 2025-09-09T15:23:22.4373925Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype143-64-size_mnk143-True] SKIPPED 2025-09-09T15:23:22.4374949Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype0-1-size_mnk0-False] SKIPPED 2025-09-09T15:23:22.4376053Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype1-1-size_mnk1-True] SKIPPED 2025-09-09T15:23:22.4377098Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype2-1-size_mnk2-False] SKIPPED 2025-09-09T15:23:22.4378099Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype3-1-size_mnk3-True] SKIPPED 2025-09-09T15:23:22.4379092Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype4-1-size_mnk4-False] SKIPPED 2025-09-09T15:23:22.4380092Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype5-1-size_mnk5-True] SKIPPED 2025-09-09T15:23:22.4381138Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype6-1-size_mnk6-False] SKIPPED 2025-09-09T15:23:22.4382175Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype7-1-size_mnk7-True] SKIPPED 2025-09-09T15:23:22.4383178Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype8-1-size_mnk8-False] SKIPPED 2025-09-09T15:23:22.4384171Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype9-1-size_mnk9-True] SKIPPED 2025-09-09T15:23:22.4385185Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype10-1-size_mnk10-False] SKIPPED 2025-09-09T15:23:22.4386252Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype11-1-size_mnk11-True] SKIPPED 2025-09-09T15:23:22.4387267Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype12-4-size_mnk12-False] SKIPPED 2025-09-09T15:23:22.4388284Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype13-4-size_mnk13-True] SKIPPED 2025-09-09T15:23:22.4389308Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype14-4-size_mnk14-False] SKIPPED 2025-09-09T15:23:22.4390327Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype15-4-size_mnk15-True] SKIPPED 2025-09-09T15:23:22.4391346Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype16-4-size_mnk16-False] SKIPPED 2025-09-09T15:23:22.4392409Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype17-4-size_mnk17-True] SKIPPED 2025-09-09T15:23:22.4393431Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype18-4-size_mnk18-False] SKIPPED 2025-09-09T15:23:22.4394449Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype19-4-size_mnk19-True] SKIPPED 2025-09-09T15:23:22.4395464Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype20-4-size_mnk20-False] SKIPPED 2025-09-09T15:23:22.4396535Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype21-4-size_mnk21-True] SKIPPED 2025-09-09T15:23:22.4397704Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype22-4-size_mnk22-False] SKIPPED 2025-09-09T15:23:22.4398726Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype23-4-size_mnk23-True] SKIPPED 2025-09-09T15:23:22.4399748Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype24-8-size_mnk24-False] SKIPPED 2025-09-09T15:23:22.4400767Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype25-8-size_mnk25-True] SKIPPED 2025-09-09T15:23:22.4401785Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype26-8-size_mnk26-False] SKIPPED 2025-09-09T15:23:22.4402803Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype27-8-size_mnk27-True] SKIPPED 2025-09-09T15:23:22.4403895Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype28-8-size_mnk28-False] SKIPPED 2025-09-09T15:23:22.4404913Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype29-8-size_mnk29-True] SKIPPED 2025-09-09T15:23:22.4405933Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype30-8-size_mnk30-False] SKIPPED 2025-09-09T15:23:22.4407178Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype31-8-size_mnk31-True] SKIPPED 2025-09-09T15:23:22.4408406Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype32-8-size_mnk32-False] SKIPPED 2025-09-09T15:23:22.4409717Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype33-8-size_mnk33-True] SKIPPED 2025-09-09T15:23:22.4410831Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype34-8-size_mnk34-False] SKIPPED 2025-09-09T15:23:22.4634410Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype35-8-size_mnk35-True] SKIPPED 2025-09-09T15:23:22.4635456Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype36-16-size_mnk36-False] SKIPPED 2025-09-09T15:23:22.4636565Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype37-16-size_mnk37-True] SKIPPED 2025-09-09T15:23:22.4637593Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype38-16-size_mnk38-False] SKIPPED 2025-09-09T15:23:22.4638619Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype39-16-size_mnk39-True] SKIPPED 2025-09-09T15:23:22.4639641Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype40-16-size_mnk40-False] SKIPPED 2025-09-09T15:23:22.4640656Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype41-16-size_mnk41-True] SKIPPED 2025-09-09T15:23:22.4641679Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype42-16-size_mnk42-False] SKIPPED 2025-09-09T15:23:22.4642809Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype43-16-size_mnk43-True] SKIPPED 2025-09-09T15:23:22.4643830Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype44-16-size_mnk44-False] SKIPPED 2025-09-09T15:23:22.4644853Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype45-16-size_mnk45-True] SKIPPED 2025-09-09T15:23:22.4645881Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype46-16-size_mnk46-False] SKIPPED 2025-09-09T15:23:22.4646896Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype47-16-size_mnk47-True] SKIPPED 2025-09-09T15:23:22.4647922Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype48-32-size_mnk48-False] SKIPPED 2025-09-09T15:23:22.4648936Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype49-32-size_mnk49-True] SKIPPED 2025-09-09T15:23:22.4649962Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype50-32-size_mnk50-False] SKIPPED 2025-09-09T15:23:22.4650987Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype51-32-size_mnk51-True] SKIPPED 2025-09-09T15:23:22.4652005Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype52-32-size_mnk52-False] SKIPPED 2025-09-09T15:23:22.4653102Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype53-32-size_mnk53-True] SKIPPED 2025-09-09T15:23:22.4654120Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype54-32-size_mnk54-False] SKIPPED 2025-09-09T15:23:22.4655137Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype55-32-size_mnk55-True] SKIPPED 2025-09-09T15:23:22.4656310Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype56-32-size_mnk56-False] SKIPPED 2025-09-09T15:23:22.4657329Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype57-32-size_mnk57-True] SKIPPED 2025-09-09T15:23:22.4658401Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype58-32-size_mnk58-False] SKIPPED 2025-09-09T15:23:22.4659427Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype59-32-size_mnk59-True] SKIPPED 2025-09-09T15:23:22.4660445Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype60-64-size_mnk60-False] SKIPPED 2025-09-09T15:23:22.4661608Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype61-64-size_mnk61-True] SKIPPED 2025-09-09T15:23:22.4662728Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype62-64-size_mnk62-False] SKIPPED 2025-09-09T15:23:22.4663891Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype63-64-size_mnk63-True] SKIPPED 2025-09-09T15:23:22.4665012Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype64-64-size_mnk64-False] SKIPPED 2025-09-09T15:23:22.4666131Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype65-64-size_mnk65-True] SKIPPED 2025-09-09T15:23:22.4667265Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype66-64-size_mnk66-False] SKIPPED 2025-09-09T15:23:22.4668459Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype67-64-size_mnk67-True] SKIPPED 2025-09-09T15:23:22.4669654Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype68-64-size_mnk68-False] SKIPPED 2025-09-09T15:23:22.4670800Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype69-64-size_mnk69-True] SKIPPED 2025-09-09T15:23:22.4671993Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype70-64-size_mnk70-False] SKIPPED 2025-09-09T15:23:22.4673118Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype71-64-size_mnk71-True] SKIPPED 2025-09-09T15:23:22.4674233Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype72-1-size_mnk72-False] SKIPPED 2025-09-09T15:23:22.4675375Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype73-1-size_mnk73-True] SKIPPED 2025-09-09T15:23:22.4676536Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype74-1-size_mnk74-False] SKIPPED 2025-09-09T15:23:22.4677637Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype75-1-size_mnk75-True] SKIPPED 2025-09-09T15:23:22.4678768Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype76-1-size_mnk76-False] SKIPPED 2025-09-09T15:23:22.4679938Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype77-1-size_mnk77-True] SKIPPED 2025-09-09T15:23:22.4681223Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype78-1-size_mnk78-False] SKIPPED 2025-09-09T15:23:22.4682361Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype79-1-size_mnk79-True] SKIPPED 2025-09-09T15:23:22.4683511Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype80-1-size_mnk80-False] SKIPPED 2025-09-09T15:23:22.4684609Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype81-1-size_mnk81-True] SKIPPED 2025-09-09T15:23:22.4685845Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype82-1-size_mnk82-False] SKIPPED 2025-09-09T15:23:22.4687039Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype83-1-size_mnk83-True] SKIPPED 2025-09-09T15:23:22.4688143Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype84-4-size_mnk84-False] SKIPPED 2025-09-09T15:23:22.4689243Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype85-4-size_mnk85-True] SKIPPED 2025-09-09T15:23:22.4690418Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype86-4-size_mnk86-False] SKIPPED 2025-09-09T15:23:22.4691520Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype87-4-size_mnk87-True] SKIPPED 2025-09-09T15:23:22.4692620Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype88-4-size_mnk88-False] SKIPPED 2025-09-09T15:23:22.4693747Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype89-4-size_mnk89-True] SKIPPED 2025-09-09T15:23:22.4694893Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype90-4-size_mnk90-False] SKIPPED 2025-09-09T15:23:22.4696106Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype91-4-size_mnk91-True] SKIPPED 2025-09-09T15:23:22.4697235Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype92-4-size_mnk92-False] SKIPPED 2025-09-09T15:23:22.4698644Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype93-4-size_mnk93-True] SKIPPED 2025-09-09T15:23:22.4699766Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype94-4-size_mnk94-False] SKIPPED 2025-09-09T15:23:22.4700914Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype95-4-size_mnk95-True] SKIPPED 2025-09-09T15:23:22.4907605Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype96-8-size_mnk96-False] SKIPPED 2025-09-09T15:23:22.4908638Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype97-8-size_mnk97-True] SKIPPED 2025-09-09T15:23:22.4909642Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype98-8-size_mnk98-False] SKIPPED 2025-09-09T15:23:22.4910643Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype99-8-size_mnk99-True] SKIPPED 2025-09-09T15:23:22.4911653Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype100-8-size_mnk100-False] SKIPPED 2025-09-09T15:23:22.4912675Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype101-8-size_mnk101-True] SKIPPED 2025-09-09T15:23:22.4913716Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype102-8-size_mnk102-False] SKIPPED 2025-09-09T15:23:22.4914865Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype103-8-size_mnk103-True] SKIPPED 2025-09-09T15:23:22.4915942Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype104-8-size_mnk104-False] SKIPPED 2025-09-09T15:23:22.4916956Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype105-8-size_mnk105-True] SKIPPED 2025-09-09T15:23:22.4918197Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype106-8-size_mnk106-False] SKIPPED 2025-09-09T15:23:22.4919338Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype107-8-size_mnk107-True] SKIPPED 2025-09-09T15:23:22.4920383Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype108-16-size_mnk108-False] SKIPPED 2025-09-09T15:23:22.4921743Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype109-16-size_mnk109-True] SKIPPED 2025-09-09T15:23:22.4922820Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype110-16-size_mnk110-False] SKIPPED 2025-09-09T15:23:22.4923994Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype111-16-size_mnk111-True] SKIPPED 2025-09-09T15:23:22.4925127Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype112-16-size_mnk112-False] SKIPPED 2025-09-09T15:23:22.4926337Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype113-16-size_mnk113-True] SKIPPED 2025-09-09T15:23:22.4927509Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype114-16-size_mnk114-False] SKIPPED 2025-09-09T15:23:22.4928630Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype115-16-size_mnk115-True] SKIPPED 2025-09-09T15:23:22.4929773Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype116-16-size_mnk116-False] SKIPPED 2025-09-09T15:23:22.4930964Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype117-16-size_mnk117-True] SKIPPED 2025-09-09T15:23:22.4932191Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype118-16-size_mnk118-False] SKIPPED 2025-09-09T15:23:22.4933398Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype119-16-size_mnk119-True] SKIPPED 2025-09-09T15:23:22.4934553Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype120-32-size_mnk120-False] SKIPPED 2025-09-09T15:23:22.4935711Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype121-32-size_mnk121-True] SKIPPED 2025-09-09T15:23:22.4936979Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype122-32-size_mnk122-False] SKIPPED 2025-09-09T15:23:22.4938112Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype123-32-size_mnk123-True] SKIPPED 2025-09-09T15:23:22.4939226Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype124-32-size_mnk124-False] SKIPPED 2025-09-09T15:23:22.4940392Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype125-32-size_mnk125-True] SKIPPED 2025-09-09T15:23:22.4941562Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype126-32-size_mnk126-False] SKIPPED 2025-09-09T15:23:22.4942682Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype127-32-size_mnk127-True] SKIPPED 2025-09-09T15:23:22.4943932Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype128-32-size_mnk128-False] SKIPPED 2025-09-09T15:23:22.4945108Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype129-32-size_mnk129-True] SKIPPED 2025-09-09T15:23:22.4946229Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype130-32-size_mnk130-False] SKIPPED 2025-09-09T15:23:22.4957801Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype131-32-size_mnk131-True] SKIPPED 2025-09-09T15:23:22.4959060Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype132-64-size_mnk132-False] SKIPPED 2025-09-09T15:23:22.4960257Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype133-64-size_mnk133-True] SKIPPED 2025-09-09T15:23:22.4961437Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype134-64-size_mnk134-False] SKIPPED 2025-09-09T15:23:22.4962572Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype135-64-size_mnk135-True] SKIPPED 2025-09-09T15:23:22.4963767Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype136-64-size_mnk136-False] SKIPPED 2025-09-09T15:23:22.4964885Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype137-64-size_mnk137-True] SKIPPED 2025-09-09T15:23:22.4966069Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype138-64-size_mnk138-False] SKIPPED 2025-09-09T15:23:22.4967322Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype139-64-size_mnk139-True] SKIPPED 2025-09-09T15:23:22.4968447Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype140-64-size_mnk140-False] SKIPPED 2025-09-09T15:23:22.4969474Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype141-64-size_mnk141-True] SKIPPED 2025-09-09T15:23:22.4970589Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype142-64-size_mnk142-False] SKIPPED 2025-09-09T15:23:22.4971685Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype143-64-size_mnk143-True] SKIPPED 2025-09-09T15:23:22.4972948Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype0-Xq_Wq_dtypes0-1-size_mnk0-False] SKIPPED 2025-09-09T15:23:22.4974159Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype1-Xq_Wq_dtypes1-1-size_mnk1-True] SKIPPED 2025-09-09T15:23:22.4975383Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype2-Xq_Wq_dtypes2-1-size_mnk2-False] SKIPPED 2025-09-09T15:23:22.4976748Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype3-Xq_Wq_dtypes3-1-size_mnk3-True] SKIPPED 2025-09-09T15:23:22.4977941Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype4-Xq_Wq_dtypes4-1-size_mnk4-False] SKIPPED 2025-09-09T15:23:22.4979139Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype5-Xq_Wq_dtypes5-1-size_mnk5-True] SKIPPED 2025-09-09T15:23:22.4980329Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype6-Xq_Wq_dtypes6-1-size_mnk6-False] SKIPPED 2025-09-09T15:23:22.4981521Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype7-Xq_Wq_dtypes7-1-size_mnk7-True] SKIPPED 2025-09-09T15:23:22.4982788Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype8-Xq_Wq_dtypes8-1-size_mnk8-False] SKIPPED 2025-09-09T15:23:22.4984075Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype9-Xq_Wq_dtypes9-1-size_mnk9-True] SKIPPED 2025-09-09T15:23:22.5145389Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype10-Xq_Wq_dtypes10-1-size_mnk10-False] SKIPPED 2025-09-09T15:23:22.5146752Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype11-Xq_Wq_dtypes11-1-size_mnk11-True] SKIPPED 2025-09-09T15:23:22.5147959Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype12-Xq_Wq_dtypes12-4-size_mnk12-False] SKIPPED 2025-09-09T15:23:22.5149227Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype13-Xq_Wq_dtypes13-4-size_mnk13-True] SKIPPED 2025-09-09T15:23:22.5150439Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype14-Xq_Wq_dtypes14-4-size_mnk14-False] SKIPPED 2025-09-09T15:23:22.5151637Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype15-Xq_Wq_dtypes15-4-size_mnk15-True] SKIPPED 2025-09-09T15:23:22.5152848Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype16-Xq_Wq_dtypes16-4-size_mnk16-False] SKIPPED 2025-09-09T15:23:22.5154052Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype17-Xq_Wq_dtypes17-4-size_mnk17-True] SKIPPED 2025-09-09T15:23:22.5155250Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype18-Xq_Wq_dtypes18-4-size_mnk18-False] SKIPPED 2025-09-09T15:23:22.5156458Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype19-Xq_Wq_dtypes19-4-size_mnk19-True] SKIPPED 2025-09-09T15:23:22.5157661Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype20-Xq_Wq_dtypes20-4-size_mnk20-False] SKIPPED 2025-09-09T15:23:22.5158930Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype21-Xq_Wq_dtypes21-4-size_mnk21-True] SKIPPED 2025-09-09T15:23:22.5160149Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype22-Xq_Wq_dtypes22-4-size_mnk22-False] SKIPPED 2025-09-09T15:23:22.5161355Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype23-Xq_Wq_dtypes23-4-size_mnk23-True] SKIPPED 2025-09-09T15:23:22.5162555Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype24-Xq_Wq_dtypes24-1-size_mnk24-False] SKIPPED 2025-09-09T15:23:22.5163756Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype25-Xq_Wq_dtypes25-1-size_mnk25-True] SKIPPED 2025-09-09T15:23:22.5164959Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype26-Xq_Wq_dtypes26-1-size_mnk26-False] SKIPPED 2025-09-09T15:23:22.5166214Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype27-Xq_Wq_dtypes27-1-size_mnk27-True] SKIPPED 2025-09-09T15:23:22.5167417Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype28-Xq_Wq_dtypes28-1-size_mnk28-False] SKIPPED 2025-09-09T15:23:22.5168626Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype29-Xq_Wq_dtypes29-1-size_mnk29-True] SKIPPED 2025-09-09T15:23:22.5169904Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype30-Xq_Wq_dtypes30-1-size_mnk30-False] SKIPPED 2025-09-09T15:23:22.5171105Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype31-Xq_Wq_dtypes31-1-size_mnk31-True] SKIPPED 2025-09-09T15:23:22.5172303Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype32-Xq_Wq_dtypes32-1-size_mnk32-False] SKIPPED 2025-09-09T15:23:22.5173551Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype33-Xq_Wq_dtypes33-1-size_mnk33-True] SKIPPED 2025-09-09T15:23:22.5174797Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype34-Xq_Wq_dtypes34-1-size_mnk34-False] SKIPPED 2025-09-09T15:23:22.5176222Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype35-Xq_Wq_dtypes35-1-size_mnk35-True] SKIPPED 2025-09-09T15:23:22.5177668Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype36-Xq_Wq_dtypes36-4-size_mnk36-False] SKIPPED 2025-09-09T15:23:22.5178989Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype37-Xq_Wq_dtypes37-4-size_mnk37-True] SKIPPED 2025-09-09T15:23:22.5180205Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype38-Xq_Wq_dtypes38-4-size_mnk38-False] SKIPPED 2025-09-09T15:23:22.5181513Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype39-Xq_Wq_dtypes39-4-size_mnk39-True] SKIPPED 2025-09-09T15:23:22.5182857Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype40-Xq_Wq_dtypes40-4-size_mnk40-False] SKIPPED 2025-09-09T15:23:22.5184228Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype41-Xq_Wq_dtypes41-4-size_mnk41-True] SKIPPED 2025-09-09T15:23:22.5185695Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype42-Xq_Wq_dtypes42-4-size_mnk42-False] SKIPPED 2025-09-09T15:23:22.5187019Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype43-Xq_Wq_dtypes43-4-size_mnk43-True] SKIPPED 2025-09-09T15:23:22.5188493Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype44-Xq_Wq_dtypes44-4-size_mnk44-False] SKIPPED 2025-09-09T15:23:22.5189805Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype45-Xq_Wq_dtypes45-4-size_mnk45-True] SKIPPED 2025-09-09T15:23:22.5191204Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype46-Xq_Wq_dtypes46-4-size_mnk46-False] SKIPPED 2025-09-09T15:23:22.5192636Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype47-Xq_Wq_dtypes47-4-size_mnk47-True] SKIPPED 2025-09-09T15:23:22.5193999Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype48-Xq_Wq_dtypes48-1-size_mnk48-False] SKIPPED 2025-09-09T15:23:22.5195380Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype49-Xq_Wq_dtypes49-1-size_mnk49-True] SKIPPED 2025-09-09T15:23:22.5196754Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype50-Xq_Wq_dtypes50-1-size_mnk50-False] SKIPPED 2025-09-09T15:23:22.5198385Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype51-Xq_Wq_dtypes51-1-size_mnk51-True] SKIPPED 2025-09-09T15:23:22.5199747Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype52-Xq_Wq_dtypes52-1-size_mnk52-False] SKIPPED 2025-09-09T15:23:22.5201100Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype53-Xq_Wq_dtypes53-1-size_mnk53-True] SKIPPED 2025-09-09T15:23:22.5202595Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype54-Xq_Wq_dtypes54-1-size_mnk54-False] SKIPPED 2025-09-09T15:23:22.5203923Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype55-Xq_Wq_dtypes55-1-size_mnk55-True] SKIPPED 2025-09-09T15:23:22.5205415Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype56-Xq_Wq_dtypes56-1-size_mnk56-False] SKIPPED 2025-09-09T15:23:22.5206824Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype57-Xq_Wq_dtypes57-1-size_mnk57-True] SKIPPED 2025-09-09T15:23:22.5208174Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype58-Xq_Wq_dtypes58-1-size_mnk58-False] SKIPPED 2025-09-09T15:23:22.5209646Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype59-Xq_Wq_dtypes59-1-size_mnk59-True] SKIPPED 2025-09-09T15:23:22.5380177Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype60-Xq_Wq_dtypes60-4-size_mnk60-False] SKIPPED 2025-09-09T15:23:22.5381437Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype61-Xq_Wq_dtypes61-4-size_mnk61-True] SKIPPED 2025-09-09T15:23:22.5382667Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype62-Xq_Wq_dtypes62-4-size_mnk62-False] SKIPPED 2025-09-09T15:23:22.5383866Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype63-Xq_Wq_dtypes63-4-size_mnk63-True] SKIPPED 2025-09-09T15:23:22.5385207Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype64-Xq_Wq_dtypes64-4-size_mnk64-False] SKIPPED 2025-09-09T15:23:22.5386482Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype65-Xq_Wq_dtypes65-4-size_mnk65-True] SKIPPED 2025-09-09T15:23:22.5387690Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype66-Xq_Wq_dtypes66-4-size_mnk66-False] SKIPPED 2025-09-09T15:23:22.5388893Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype67-Xq_Wq_dtypes67-4-size_mnk67-True] SKIPPED 2025-09-09T15:23:22.5390099Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype68-Xq_Wq_dtypes68-4-size_mnk68-False] SKIPPED 2025-09-09T15:23:22.5391309Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype69-Xq_Wq_dtypes69-4-size_mnk69-True] SKIPPED 2025-09-09T15:23:22.5392519Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype70-Xq_Wq_dtypes70-4-size_mnk70-False] SKIPPED 2025-09-09T15:23:22.5393720Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype71-Xq_Wq_dtypes71-4-size_mnk71-True] SKIPPED 2025-09-09T15:23:22.5394931Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype72-Xq_Wq_dtypes72-1-size_mnk72-False] SKIPPED 2025-09-09T15:23:22.5396213Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype73-Xq_Wq_dtypes73-1-size_mnk73-True] SKIPPED 2025-09-09T15:23:22.5397556Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype74-Xq_Wq_dtypes74-1-size_mnk74-False] SKIPPED 2025-09-09T15:23:22.5398766Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype75-Xq_Wq_dtypes75-1-size_mnk75-True] SKIPPED 2025-09-09T15:23:22.5400056Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype76-Xq_Wq_dtypes76-1-size_mnk76-False] SKIPPED 2025-09-09T15:23:22.5401318Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype77-Xq_Wq_dtypes77-1-size_mnk77-True] SKIPPED 2025-09-09T15:23:22.5402527Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype78-Xq_Wq_dtypes78-1-size_mnk78-False] SKIPPED 2025-09-09T15:23:22.5403738Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype79-Xq_Wq_dtypes79-1-size_mnk79-True] SKIPPED 2025-09-09T15:23:22.5404944Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype80-Xq_Wq_dtypes80-1-size_mnk80-False] SKIPPED 2025-09-09T15:23:22.5406212Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype81-Xq_Wq_dtypes81-1-size_mnk81-True] SKIPPED 2025-09-09T15:23:22.5407417Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype82-Xq_Wq_dtypes82-1-size_mnk82-False] SKIPPED 2025-09-09T15:23:22.5408629Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype83-Xq_Wq_dtypes83-1-size_mnk83-True] SKIPPED 2025-09-09T15:23:22.5409835Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype84-Xq_Wq_dtypes84-4-size_mnk84-False] SKIPPED 2025-09-09T15:23:22.5411037Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype85-Xq_Wq_dtypes85-4-size_mnk85-True] SKIPPED 2025-09-09T15:23:22.5412319Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype86-Xq_Wq_dtypes86-4-size_mnk86-False] SKIPPED 2025-09-09T15:23:22.5413537Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype87-Xq_Wq_dtypes87-4-size_mnk87-True] SKIPPED 2025-09-09T15:23:22.5414744Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype88-Xq_Wq_dtypes88-4-size_mnk88-False] SKIPPED 2025-09-09T15:23:22.5416064Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype89-Xq_Wq_dtypes89-4-size_mnk89-True] SKIPPED 2025-09-09T15:23:22.5417270Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype90-Xq_Wq_dtypes90-4-size_mnk90-False] SKIPPED 2025-09-09T15:23:22.5418480Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype91-Xq_Wq_dtypes91-4-size_mnk91-True] SKIPPED 2025-09-09T15:23:22.5419694Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype92-Xq_Wq_dtypes92-4-size_mnk92-False] SKIPPED 2025-09-09T15:23:22.5420907Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype93-Xq_Wq_dtypes93-4-size_mnk93-True] SKIPPED 2025-09-09T15:23:22.5422181Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype94-Xq_Wq_dtypes94-4-size_mnk94-False] SKIPPED 2025-09-09T15:23:22.5423392Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype95-Xq_Wq_dtypes95-4-size_mnk95-True] SKIPPED 2025-09-09T15:23:22.5424601Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype96-Xq_Wq_dtypes96-1-size_mnk96-False] SKIPPED 2025-09-09T15:23:22.5425850Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype97-Xq_Wq_dtypes97-1-size_mnk97-True] SKIPPED 2025-09-09T15:23:22.5427058Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype98-Xq_Wq_dtypes98-1-size_mnk98-False] SKIPPED 2025-09-09T15:23:22.5428305Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype99-Xq_Wq_dtypes99-1-size_mnk99-True] SKIPPED 2025-09-09T15:23:22.5429535Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype100-Xq_Wq_dtypes100-1-size_mnk100-False] SKIPPED 2025-09-09T15:23:22.5430781Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype101-Xq_Wq_dtypes101-1-size_mnk101-True] SKIPPED 2025-09-09T15:23:22.5432037Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype102-Xq_Wq_dtypes102-1-size_mnk102-False] SKIPPED 2025-09-09T15:23:22.5433448Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype103-Xq_Wq_dtypes103-1-size_mnk103-True] SKIPPED 2025-09-09T15:23:22.5435009Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype104-Xq_Wq_dtypes104-1-size_mnk104-False] SKIPPED 2025-09-09T15:23:22.5436401Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype105-Xq_Wq_dtypes105-1-size_mnk105-True] SKIPPED 2025-09-09T15:23:22.5437740Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype106-Xq_Wq_dtypes106-1-size_mnk106-False] SKIPPED 2025-09-09T15:23:22.5439226Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype107-Xq_Wq_dtypes107-1-size_mnk107-True] SKIPPED 2025-09-09T15:23:22.5440674Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype108-Xq_Wq_dtypes108-4-size_mnk108-False] SKIPPED 2025-09-09T15:23:22.5442013Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype109-Xq_Wq_dtypes109-4-size_mnk109-True] SKIPPED 2025-09-09T15:23:22.5611641Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype110-Xq_Wq_dtypes110-4-size_mnk110-False] SKIPPED 2025-09-09T15:23:22.5612918Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype111-Xq_Wq_dtypes111-4-size_mnk111-True] SKIPPED 2025-09-09T15:23:22.5614184Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype112-Xq_Wq_dtypes112-4-size_mnk112-False] SKIPPED 2025-09-09T15:23:22.5615434Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype113-Xq_Wq_dtypes113-4-size_mnk113-True] SKIPPED 2025-09-09T15:23:22.5616771Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype114-Xq_Wq_dtypes114-4-size_mnk114-False] SKIPPED 2025-09-09T15:23:22.5618124Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype115-Xq_Wq_dtypes115-4-size_mnk115-True] SKIPPED 2025-09-09T15:23:22.5619365Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype116-Xq_Wq_dtypes116-4-size_mnk116-False] SKIPPED 2025-09-09T15:23:22.5620610Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype117-Xq_Wq_dtypes117-4-size_mnk117-True] SKIPPED 2025-09-09T15:23:22.5621943Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype118-Xq_Wq_dtypes118-4-size_mnk118-False] SKIPPED 2025-09-09T15:23:22.5623181Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype119-Xq_Wq_dtypes119-4-size_mnk119-True] SKIPPED 2025-09-09T15:23:22.5624484Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype120-Xq_Wq_dtypes120-1-size_mnk120-False] SKIPPED 2025-09-09T15:23:22.5625767Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype121-Xq_Wq_dtypes121-1-size_mnk121-True] SKIPPED 2025-09-09T15:23:22.5627012Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype122-Xq_Wq_dtypes122-1-size_mnk122-False] SKIPPED 2025-09-09T15:23:22.5628254Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype123-Xq_Wq_dtypes123-1-size_mnk123-True] SKIPPED 2025-09-09T15:23:22.5629498Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype124-Xq_Wq_dtypes124-1-size_mnk124-False] SKIPPED 2025-09-09T15:23:22.5630739Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype125-Xq_Wq_dtypes125-1-size_mnk125-True] SKIPPED 2025-09-09T15:23:22.5631990Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype126-Xq_Wq_dtypes126-1-size_mnk126-False] SKIPPED 2025-09-09T15:23:22.5633237Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype127-Xq_Wq_dtypes127-1-size_mnk127-True] SKIPPED 2025-09-09T15:23:22.5634543Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype128-Xq_Wq_dtypes128-1-size_mnk128-False] SKIPPED 2025-09-09T15:23:22.5635794Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype129-Xq_Wq_dtypes129-1-size_mnk129-True] SKIPPED 2025-09-09T15:23:22.5637033Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype130-Xq_Wq_dtypes130-1-size_mnk130-False] SKIPPED 2025-09-09T15:23:22.5638272Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype131-Xq_Wq_dtypes131-1-size_mnk131-True] SKIPPED 2025-09-09T15:23:22.5639513Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype132-Xq_Wq_dtypes132-4-size_mnk132-False] SKIPPED 2025-09-09T15:23:22.5640750Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype133-Xq_Wq_dtypes133-4-size_mnk133-True] SKIPPED 2025-09-09T15:23:22.5641993Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype134-Xq_Wq_dtypes134-4-size_mnk134-False] SKIPPED 2025-09-09T15:23:22.5643239Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype135-Xq_Wq_dtypes135-4-size_mnk135-True] SKIPPED 2025-09-09T15:23:22.5644478Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype136-Xq_Wq_dtypes136-4-size_mnk136-False] SKIPPED 2025-09-09T15:23:22.5645819Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype137-Xq_Wq_dtypes137-4-size_mnk137-True] SKIPPED 2025-09-09T15:23:22.5647059Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype138-Xq_Wq_dtypes138-4-size_mnk138-False] SKIPPED 2025-09-09T15:23:22.5648343Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype139-Xq_Wq_dtypes139-4-size_mnk139-True] SKIPPED 2025-09-09T15:23:22.5649585Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype140-Xq_Wq_dtypes140-4-size_mnk140-False] SKIPPED 2025-09-09T15:23:22.5650894Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype141-Xq_Wq_dtypes141-4-size_mnk141-True] SKIPPED 2025-09-09T15:23:22.5652130Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype142-Xq_Wq_dtypes142-4-size_mnk142-False] SKIPPED 2025-09-09T15:23:22.5653369Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype143-Xq_Wq_dtypes143-4-size_mnk143-True] SKIPPED 2025-09-09T15:23:22.5654606Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype144-Xq_Wq_dtypes144-1-size_mnk144-False] SKIPPED 2025-09-09T15:23:22.5655957Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype145-Xq_Wq_dtypes145-1-size_mnk145-True] SKIPPED 2025-09-09T15:23:22.5657197Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype146-Xq_Wq_dtypes146-1-size_mnk146-False] SKIPPED 2025-09-09T15:23:22.5658435Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype147-Xq_Wq_dtypes147-1-size_mnk147-True] SKIPPED 2025-09-09T15:23:22.5659679Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype148-Xq_Wq_dtypes148-1-size_mnk148-False] SKIPPED 2025-09-09T15:23:22.5660986Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype149-Xq_Wq_dtypes149-1-size_mnk149-True] SKIPPED 2025-09-09T15:23:22.5662239Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype150-Xq_Wq_dtypes150-1-size_mnk150-False] SKIPPED 2025-09-09T15:23:22.5663483Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype151-Xq_Wq_dtypes151-1-size_mnk151-True] SKIPPED 2025-09-09T15:23:22.5664737Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype152-Xq_Wq_dtypes152-1-size_mnk152-False] SKIPPED 2025-09-09T15:23:22.5665970Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype153-Xq_Wq_dtypes153-1-size_mnk153-True] SKIPPED 2025-09-09T15:23:22.5667208Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype154-Xq_Wq_dtypes154-1-size_mnk154-False] SKIPPED 2025-09-09T15:23:22.5668449Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype155-Xq_Wq_dtypes155-1-size_mnk155-True] SKIPPED 2025-09-09T15:23:22.5669683Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype156-Xq_Wq_dtypes156-4-size_mnk156-False] SKIPPED 2025-09-09T15:23:22.5670929Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype157-Xq_Wq_dtypes157-4-size_mnk157-True] SKIPPED 2025-09-09T15:23:22.5672210Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype158-Xq_Wq_dtypes158-4-size_mnk158-False] SKIPPED 2025-09-09T15:23:22.6006461Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype159-Xq_Wq_dtypes159-4-size_mnk159-True] SKIPPED 2025-09-09T15:23:22.6007890Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype160-Xq_Wq_dtypes160-4-size_mnk160-False] SKIPPED 2025-09-09T15:23:22.6009435Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype161-Xq_Wq_dtypes161-4-size_mnk161-True] SKIPPED 2025-09-09T15:23:22.6010904Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype162-Xq_Wq_dtypes162-4-size_mnk162-False] SKIPPED 2025-09-09T15:23:22.6012330Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype163-Xq_Wq_dtypes163-4-size_mnk163-True] SKIPPED 2025-09-09T15:23:22.6013769Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype164-Xq_Wq_dtypes164-4-size_mnk164-False] SKIPPED 2025-09-09T15:23:22.6015115Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype165-Xq_Wq_dtypes165-4-size_mnk165-True] SKIPPED 2025-09-09T15:23:22.6016698Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype166-Xq_Wq_dtypes166-4-size_mnk166-False] SKIPPED 2025-09-09T15:23:22.6018129Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype167-Xq_Wq_dtypes167-4-size_mnk167-True] SKIPPED 2025-09-09T15:23:22.6019471Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype168-Xq_Wq_dtypes168-1-size_mnk168-False] SKIPPED 2025-09-09T15:23:22.6020901Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype169-Xq_Wq_dtypes169-1-size_mnk169-True] SKIPPED 2025-09-09T15:23:22.6022367Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype170-Xq_Wq_dtypes170-1-size_mnk170-False] SKIPPED 2025-09-09T15:23:22.6023760Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype171-Xq_Wq_dtypes171-1-size_mnk171-True] SKIPPED 2025-09-09T15:23:22.6025191Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype172-Xq_Wq_dtypes172-1-size_mnk172-False] SKIPPED 2025-09-09T15:23:22.6026532Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype173-Xq_Wq_dtypes173-1-size_mnk173-True] SKIPPED 2025-09-09T15:23:22.6027950Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype174-Xq_Wq_dtypes174-1-size_mnk174-False] SKIPPED 2025-09-09T15:23:22.6029300Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype175-Xq_Wq_dtypes175-1-size_mnk175-True] SKIPPED 2025-09-09T15:23:22.6030707Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype176-Xq_Wq_dtypes176-1-size_mnk176-False] SKIPPED 2025-09-09T15:23:22.6032119Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype177-Xq_Wq_dtypes177-1-size_mnk177-True] SKIPPED 2025-09-09T15:23:22.6033456Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype178-Xq_Wq_dtypes178-1-size_mnk178-False] SKIPPED 2025-09-09T15:23:22.6034995Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype179-Xq_Wq_dtypes179-1-size_mnk179-True] SKIPPED 2025-09-09T15:23:22.6036443Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype180-Xq_Wq_dtypes180-4-size_mnk180-False] SKIPPED 2025-09-09T15:23:22.6037804Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype181-Xq_Wq_dtypes181-4-size_mnk181-True] SKIPPED 2025-09-09T15:23:22.6039290Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype182-Xq_Wq_dtypes182-4-size_mnk182-False] SKIPPED 2025-09-09T15:23:22.6040709Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype183-Xq_Wq_dtypes183-4-size_mnk183-True] SKIPPED 2025-09-09T15:23:22.6042105Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype184-Xq_Wq_dtypes184-4-size_mnk184-False] SKIPPED 2025-09-09T15:23:22.6043426Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype185-Xq_Wq_dtypes185-4-size_mnk185-True] SKIPPED 2025-09-09T15:23:22.6044691Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype186-Xq_Wq_dtypes186-4-size_mnk186-False] SKIPPED 2025-09-09T15:23:22.6046081Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype187-Xq_Wq_dtypes187-4-size_mnk187-True] SKIPPED 2025-09-09T15:23:22.6047331Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype188-Xq_Wq_dtypes188-4-size_mnk188-False] SKIPPED 2025-09-09T15:23:22.6048580Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype189-Xq_Wq_dtypes189-4-size_mnk189-True] SKIPPED 2025-09-09T15:23:22.6049924Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype190-Xq_Wq_dtypes190-4-size_mnk190-False] SKIPPED 2025-09-09T15:23:22.6051174Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype191-Xq_Wq_dtypes191-4-size_mnk191-True] SKIPPED 2025-09-09T15:23:22.6052202Z test/test_utils.py::TestTorchVersion::test_torch_version_at_least PASSED 2025-09-09T15:23:22.6052882Z test/test_utils.py::TestTorchVersion::test_torch_version_deprecation PASSED 2025-09-09T15:23:22.6053552Z test/test_utils.py::TestTorchAOBaseTensor::test_default_impls PASSED 2025-09-09T15:23:22.6054277Z test/test_utils.py::TestTorchAOBaseTensor::test_default_impls_with_optional_attr PASSED 2025-09-09T15:23:22.6055063Z test/test_utils.py::TestTorchAOBaseTensor::test_default_impls_with_optional_data PASSED 2025-09-09T15:23:22.6055870Z test/test_utils.py::TestTorchAOBaseTensor::test_print_arg_types PASSED 2025-09-09T15:23:22.6056210Z 2025-09-09T15:23:22.6056442Z =============================== warnings summary =============================== 2025-09-09T15:23:22.6057027Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config8] 2025-09-09T15:23:22.6058184Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/core/config.py:250: UserWarning: Stored version is not the same as current default version of the config: stored_version=2, current_default_version=1, please check the deprecation warning 2025-09-09T15:23:22.6059250Z warnings.warn( 2025-09-09T15:23:22.6059391Z 2025-09-09T15:23:22.6059575Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_copy_bfloat16 2025-09-09T15:23:22.6060028Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_copy_float16 2025-09-09T15:23:22.6060530Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_copy_float32 2025-09-09T15:23:22.6061801Z /pytorch/ao/test/dtypes/test_nf4.py:223: FutureWarning: `torch.testing.assert_allclose()` is deprecated since 1.12 and will be removed in a future release. Please use `torch.testing.assert_close()` instead. You can find detailed upgrade instructions in https://github.com/pytorch/pytorch/issues/61844. 2025-09-09T15:23:22.6063112Z torch.testing.assert_allclose(input_tensor, nf4_to_dtype, atol=0.13, rtol=0.13) 2025-09-09T15:23:22.6063460Z 2025-09-09T15:23:22.6063647Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_copy_bfloat16 2025-09-09T15:23:22.6064152Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_copy_float16 2025-09-09T15:23:22.6064591Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_copy_float32 2025-09-09T15:23:22.6065838Z /pytorch/ao/test/dtypes/test_nf4.py:229: FutureWarning: `torch.testing.assert_allclose()` is deprecated since 1.12 and will be removed in a future release. Please use `torch.testing.assert_close()` instead. You can find detailed upgrade instructions in https://github.com/pytorch/pytorch/issues/61844. 2025-09-09T15:23:22.6067057Z torch.testing.assert_allclose( 2025-09-09T15:23:22.6067254Z 2025-09-09T15:23:22.6067362Z test/float8/test_base.py: 36 warnings 2025-09-09T15:23:22.6068511Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/float8/float8_linear.py:261: DeprecationWarning: torch.get_autocast_gpu_dtype() is deprecated. Please use torch.get_autocast_dtype('cuda') instead. (Triggered internally at /pytorch/torch/csrc/autograd/init.cpp:826.) 2025-09-09T15:23:22.6069683Z autocast_dtype = torch.get_autocast_gpu_dtype() 2025-09-09T15:23:22.6069917Z 2025-09-09T15:23:22.6070197Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype3] 2025-09-09T15:23:22.6070770Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype4] 2025-09-09T15:23:22.6071289Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype5] 2025-09-09T15:23:22.6071814Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype6] 2025-09-09T15:23:22.6072366Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype7] 2025-09-09T15:23:22.6073386Z /pytorch/ao/test/float8/test_float8_utils.py:67: DeprecationWarning: an integer is required (got type float). Implicit conversion to integers using __int__ is deprecated, and may be removed in a future version of Python. 2025-09-09T15:23:22.6074363Z non_float32_tensor = torch.tensor([3.0], dtype=invalid_dtype) 2025-09-09T15:23:22.6074690Z 2025-09-09T15:23:22.6074893Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_mm_0_cuda 2025-09-09T15:23:22.6075378Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_mm_1_cuda 2025-09-09T15:23:22.6076673Z /pytorch/ao/test/kernel/test_autotuner.py:50: FutureWarning: `torch.testing.assert_allclose()` is deprecated since 1.12 and will be removed in a future release. Please use `torch.testing.assert_close()` instead. You can find detailed upgrade instructions in https://github.com/pytorch/pytorch/issues/61844. 2025-09-09T15:23:22.6077906Z torch.testing.assert_allclose(out32_1, out32_2) 2025-09-09T15:23:22.6078139Z 2025-09-09T15:23:22.6078354Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_scaled_mm_0_cuda 2025-09-09T15:23:22.6078882Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_scaled_mm_1_cpu 2025-09-09T15:23:22.6079397Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_scaled_mm_2_cuda 2025-09-09T15:23:22.6079917Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_scaled_mm_3_cpu 2025-09-09T15:23:22.6081249Z /pytorch/ao/test/kernel/test_autotuner.py:96: FutureWarning: `torch.testing.assert_allclose()` is deprecated since 1.12 and will be removed in a future release. Please use `torch.testing.assert_close()` instead. You can find detailed upgrade instructions in https://github.com/pytorch/pytorch/issues/61844. 2025-09-09T15:23:22.6082468Z torch.testing.assert_allclose(out32_1, out32_2) 2025-09-09T15:23:22.6082751Z 2025-09-09T15:23:22.6083057Z test/prototype/test_codebook_quant.py::TestCodebookQuantization::test_choose_qparams_codebook 2025-09-09T15:23:22.6084337Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/testing/_internal/common_utils.py:903: UserWarning: index_reduce() is in beta and the API may change at any time. (Triggered internally at /pytorch/aten/src/ATen/native/TensorAdvancedIndexing.cpp:1531.) 2025-09-09T15:23:22.6085437Z return callable(*args, **kwargs) 2025-09-09T15:23:22.6085646Z 2025-09-09T15:23:22.6085890Z test/prototype/test_parametrization.py::TestFakeSparsity::test_jit_trace 2025-09-09T15:23:22.6087525Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/sparsity/sparsifier/utils.py:134: TracerWarning: Converting a tensor to a Python boolean might cause the trace to be incorrect. We can't record the data flow of Python values, so this value will be treated as a constant in the future. This means that the trace might not generalize to other inputs! 2025-09-09T15:23:22.6088999Z assert self.mask.shape == x.shape 2025-09-09T15:23:22.6089195Z 2025-09-09T15:23:22.6089421Z test/prototype/test_scheduler.py::TestScheduler::test_lambda_scheduler 2025-09-09T15:23:22.6089929Z test/prototype/test_scheduler.py::TestCubicScheduler::test_step 2025-09-09T15:23:22.6091204Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/sparsity/scheduler/base_scheduler.py:133: UserWarning: Detected call of `scheduler.step()` before `sparsifier.step()`. You have to make sure you run the sparsifier.step() BEFORE any calls to the scheduler.step(). 2025-09-09T15:23:22.6092357Z warnings.warn( 2025-09-09T15:23:22.6092493Z 2025-09-09T15:23:22.6092758Z test/quantization/pt2e/test_graph_utils.py::TestGraphUtils::test_conv_bn_conv_relu 2025-09-09T15:23:22.6094038Z /pytorch/ao/test/quantization/pt2e/test_graph_utils.py:42: FutureWarning: export(f, *args, **kwargs) is deprecated, use export(f)(*args, **kwargs) instead. If you don't migrate, we may break your export call in the future if your user defined kwargs conflict with future kwargs added to export(f). 2025-09-09T15:23:22.6095244Z m, guards = torchdynamo.export( # noqa: F841© 2025-09-09T15:23:22.6095477Z 2025-09-09T15:23:22.6095732Z test/quantization/pt2e/test_graph_utils.py::TestGraphUtils::test_conv_bn_relu 2025-09-09T15:23:22.6098314Z /pytorch/ao/test/quantization/pt2e/test_graph_utils.py:86: FutureWarning: export(f, *args, **kwargs) is deprecated, use export(f)(*args, **kwargs) instead. If you don't migrate, we may break your export call in the future if your user defined kwargs conflict with future kwargs added to export(f). 2025-09-09T15:23:22.6099479Z m, guards = torchdynamo.export( # noqa: F841 2025-09-09T15:23:22.6099710Z 2025-09-09T15:23:22.6100030Z test/quantization/pt2e/test_graph_utils.py::TestGraphUtils::test_customized_equivalet_types_dict 2025-09-09T15:23:22.6101369Z /pytorch/ao/test/quantization/pt2e/test_graph_utils.py:118: FutureWarning: export(f, *args, **kwargs) is deprecated, use export(f)(*args, **kwargs) instead. If you don't migrate, we may break your export call in the future if your user defined kwargs conflict with future kwargs added to export(f). 2025-09-09T15:23:22.6102515Z m, guards = torchdynamo.export( # noqa: F841 2025-09-09T15:23:22.6102738Z 2025-09-09T15:23:22.6102908Z test/quantization/pt2e/test_quantize_pt2e.py: 18 warnings 2025-09-09T15:23:22.6103349Z test/quantization/pt2e/test_quantize_pt2e_qat.py: 91 warnings 2025-09-09T15:23:22.6103804Z test/quantization/pt2e/test_representation.py: 8 warnings 2025-09-09T15:23:22.6104590Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/testing/pt2e/_xnnpack_quantizer.py:289: UserWarning: XNNPACKQuantizer is deprecated! 2025-09-09T15:23:22.6105354Z warnings.warn(f"{self.__class__.__name__} is deprecated!") 2025-09-09T15:23:22.6105621Z 2025-09-09T15:23:22.6105967Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_fold_all_ops_before_quantize 2025-09-09T15:23:22.6106753Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_flatten_recipe2 2025-09-09T15:23:22.6108926Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/export/_unlift.py:81: UserWarning: Attempted to insert a get_attr Node with no underlying reference in the owning GraphModule! Call GraphModule.add_submodule to add the necessary submodule, GraphModule.add_parameter to add the necessary Parameter, or nn.Module.register_buffer to add the necessary buffer 2025-09-09T15:23:22.6110703Z getattr_node = gm.graph.get_attr(lifted_node) 2025-09-09T15:23:22.6110964Z 2025-09-09T15:23:22.6111395Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_fold_all_ops_before_quantize 2025-09-09T15:23:22.6112822Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/fx/graph.py:1772: UserWarning: Node weight target weight weight of does not reference an nn.Module, nn.Parameter, or buffer, which is what 'get_attr' Nodes typically target 2025-09-09T15:23:22.6114018Z warnings.warn( 2025-09-09T15:23:22.6114166Z 2025-09-09T15:23:22.6114486Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_model_is_exported 2025-09-09T15:23:22.6116061Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/fx/_symbolic_trace.py:913: UserWarning: Was not able to add assertion to guarantee correct input x to specialized function. It is up to the user to make sure that your inputs match the inputs you specialized the function with. 2025-09-09T15:23:22.6117387Z warnings.warn( 2025-09-09T15:23:22.6117529Z 2025-09-09T15:23:22.6117820Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_reentrant 2025-09-09T15:23:22.6118651Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_fold_bn_erases_bn_node 2025-09-09T15:23:22.6119601Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_fold_bn_erases_bn_node 2025-09-09T15:23:22.6120933Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/quantization/pt2e/utils.py:145: UserWarning: must run observer before calling calculate_qparams. Returning default values. 2025-09-09T15:23:22.6121929Z warnings.warn( 2025-09-09T15:23:22.6122070Z 2025-09-09T15:23:22.6122357Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_reentrant 2025-09-09T15:23:22.6123698Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/quantization/pt2e/observer.py:1350: UserWarning: must run observer before calling calculate_qparams. Returning default scale and zero point 2025-09-09T15:23:22.6124888Z warnings.warn( 2025-09-09T15:23:22.6125036Z 2025-09-09T15:23:22.6125502Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_bn_bias_derived_qspec 2025-09-09T15:23:22.6126621Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_bn_per_channel_weight_bias 2025-09-09T15:23:22.6127704Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_per_channel_weight_custom_dtype 2025-09-09T15:23:22.6128763Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_bn_bias_derived_qspec 2025-09-09T15:23:22.6129822Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_bn_per_channel_weight_bias 2025-09-09T15:23:22.6130901Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_per_channel_weight_custom_dtype 2025-09-09T15:23:22.6132550Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/quantization/pt2e/observer.py:253: UserWarning: Please use quant_min and quant_max to specify the range for observers. reduce_range will be deprecated in a future release of PyTorch. 2025-09-09T15:23:22.6133787Z warnings.warn( 2025-09-09T15:23:22.6133935Z 2025-09-09T15:23:22.6134374Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_filter_conv2d_recipe 2025-09-09T15:23:22.6139125Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/quantization/pt2e/quantizer/x86_inductor_quantizer.py:1325: UserWarning: The input of maxpool2d is not quantized, skip annotate maxpool2d with config QuantizationConfig(input_activation=QuantizationSpec(dtype=torch.uint8, observer_or_fake_quant_ctr=functools.partial(, eps=0.000244140625){}, quant_min=0, quant_max=255, qscheme=torch.per_tensor_affine, ch_axis=None, is_dynamic=False), output_activation=QuantizationSpec(dtype=torch.uint8, observer_or_fake_quant_ctr=functools.partial(, eps=0.000244140625){}, quant_min=0, quant_max=255, qscheme=torch.per_tensor_affine, ch_axis=None, is_dynamic=False), weight=QuantizationSpec(dtype=torch.int8, observer_or_fake_quant_ctr=functools.partial(, eps=0.000244140625){}, quant_min=-128, quant_max=127, qscheme=torch.per_channel_symmetric, ch_axis=0, is_dynamic=False), bias=None, is_qat=False). 2025-09-09T15:23:22.6143364Z warnings.warn( 2025-09-09T15:23:22.6143499Z 2025-09-09T15:23:22.6143858Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_flatten_recipe2 2025-09-09T15:23:22.6145089Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/fx/graph.py:1772: UserWarning: Node cls_token target cls_token cls_token of does not reference an nn.Module, nn.Parameter, or buffer, which is what 'get_attr' Nodes typically target 2025-09-09T15:23:22.6146077Z warnings.warn( 2025-09-09T15:23:22.6146211Z 2025-09-09T15:23:22.6146696Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_set_module_name_and_module_type_with_mixed_configs 2025-09-09T15:23:22.6147975Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/quantization/pt2e/quantizer/x86_inductor_quantizer.py:484: UserWarning: Mixed dynamic and static quantization config is not supported. 2025-09-09T15:23:22.6149008Z warnings.warn( 2025-09-09T15:23:22.6149142Z 2025-09-09T15:23:22.6149726Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_set_module_name_and_module_type_with_mixed_configs 2025-09-09T15:23:22.6151101Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/quantization/pt2e/quantizer/x86_inductor_quantizer.py:383: UserWarning: Skip the quantization config for . 2025-09-09T15:23:22.6152067Z warnings.warn( 2025-09-09T15:23:22.6152200Z 2025-09-09T15:23:22.6152661Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int8dq_base_0_multiple_tokens 2025-09-09T15:23:22.6153398Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/_inductor/lowering.py:7007: UserWarning: 2025-09-09T15:23:22.6154005Z Online softmax is disabled on the fly since Inductor decides to 2025-09-09T15:23:22.6154456Z split the reduction. Cut an issue to PyTorch if this is an 2025-09-09T15:23:22.6154883Z important use case and you want to speed it up with online 2025-09-09T15:23:22.6155226Z softmax. 2025-09-09T15:23:22.6155416Z 2025-09-09T15:23:22.6155600Z warnings.warn( 2025-09-09T15:23:22.6155751Z 2025-09-09T15:23:22.6156003Z test/quantization/test_qat.py::TestQAT::test_legacy_quantize_api_e2e 2025-09-09T15:23:22.6157176Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/quantization/qat/utils.py:84: UserWarning: 'IntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead: 2025-09-09T15:23:22.6158220Z 2025-09-09T15:23:22.6158543Z base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) 2025-09-09T15:23:22.6159026Z quantize_(model, QATConfig(base_config, step="prepare")) 2025-09-09T15:23:22.6159376Z # train (not shown) 2025-09-09T15:23:22.6159687Z quantize_(model, QATConfig(base_config, step="convert")) 2025-09-09T15:23:22.6160020Z 2025-09-09T15:23:22.6160320Z Alternatively, if you prefer to pass in fake quantization configs: 2025-09-09T15:23:22.6160736Z 2025-09-09T15:23:22.6161113Z activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) 2025-09-09T15:23:22.6161698Z weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) 2025-09-09T15:23:22.6162090Z qat_config = QATConfig( 2025-09-09T15:23:22.6162369Z activation_config=activation_config, 2025-09-09T15:23:22.6162680Z weight_config=weight_config, 2025-09-09T15:23:22.6162959Z step="prepare", 2025-09-09T15:23:22.6163192Z ) 2025-09-09T15:23:22.6163392Z quantize_(model, qat_config) 2025-09-09T15:23:22.6163644Z 2025-09-09T15:23:22.6163997Z Please see https://github.com/pytorch/ao/issues/2630 for more details. 2025-09-09T15:23:22.6164384Z 2025-09-09T15:23:22.6164617Z warnings.warn( 2025-09-09T15:23:22.6164755Z 2025-09-09T15:23:22.6164970Z test/quantization/test_qat.py::TestQAT::test_legacy_quantize_api_e2e 2025-09-09T15:23:22.6166232Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/quantization/qat/utils.py:84: UserWarning: 'FromIntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead: 2025-09-09T15:23:22.6167291Z 2025-09-09T15:23:22.6167603Z base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) 2025-09-09T15:23:22.6168076Z quantize_(model, QATConfig(base_config, step="prepare")) 2025-09-09T15:23:22.6168425Z # train (not shown) 2025-09-09T15:23:22.6168735Z quantize_(model, QATConfig(base_config, step="convert")) 2025-09-09T15:23:22.6169064Z 2025-09-09T15:23:22.6169373Z Alternatively, if you prefer to pass in fake quantization configs: 2025-09-09T15:23:22.6169736Z 2025-09-09T15:23:22.6170126Z activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) 2025-09-09T15:23:22.6170717Z weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) 2025-09-09T15:23:22.6171115Z qat_config = QATConfig( 2025-09-09T15:23:22.6171394Z activation_config=activation_config, 2025-09-09T15:23:22.6171701Z weight_config=weight_config, 2025-09-09T15:23:22.6171980Z step="prepare", 2025-09-09T15:23:22.6172200Z ) 2025-09-09T15:23:22.6172406Z quantize_(model, qat_config) 2025-09-09T15:23:22.6172650Z 2025-09-09T15:23:22.6172959Z Please see https://github.com/pytorch/ao/issues/2630 for more details. 2025-09-09T15:23:22.6173335Z 2025-09-09T15:23:22.6173585Z warnings.warn( 2025-09-09T15:23:22.6173720Z 2025-09-09T15:23:22.6173925Z test/quantization/test_qat.py::TestQAT::test_qat_fp8a4w_quantizer 2025-09-09T15:23:22.6177228Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/autograd/graph.py:824: UserWarning: torchao::dequantize_affine_float8: an autograd kernel was not registered to the Autograd key(s) but we are trying to backprop through it. This may lead to silently incorrect behavior. This behavior is deprecated and will be removed in a future version of PyTorch. If your operator is differentiable, please ensure you have registered an autograd kernel to the correct Autograd key (e.g. DispatchKey::Autograd, DispatchKey::CompositeImplicitAutograd). If your operator is not differentiable, or to squash this warning and use the previous behavior, please register torch::CppFunction::makeFallthrough() to DispatchKey::Autograd. (Triggered internally at /pytorch/torch/csrc/autograd/autograd_not_implemented_fallback.cpp:62.) 2025-09-09T15:23:22.6180514Z return Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass 2025-09-09T15:23:22.6180928Z 2025-09-09T15:23:22.6181196Z test/sparsity/test_marlin.py::SparseMarlin24::test_quant_sparse_marlin_layout_compile 2025-09-09T15:23:22.6181833Z test/sparsity/test_sparse_api.py::TestQuantSemiSparse::test_sparse_marlin_compile_True 2025-09-09T15:23:22.6185109Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/autograd/graph.py:824: UserWarning: torchao::marlin_24_gemm: an autograd kernel was not registered to the Autograd key(s) but we are trying to backprop through it. This may lead to silently incorrect behavior. This behavior is deprecated and will be removed in a future version of PyTorch. If your operator is differentiable, please ensure you have registered an autograd kernel to the correct Autograd key (e.g. DispatchKey::Autograd, DispatchKey::CompositeImplicitAutograd). If your operator is not differentiable, or to squash this warning and use the previous behavior, please register torch::CppFunction::makeFallthrough() to DispatchKey::Autograd. (Triggered internally at /pytorch/torch/csrc/autograd/autograd_not_implemented_fallback.cpp:62.) 2025-09-09T15:23:22.6188436Z return Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass 2025-09-09T15:23:22.6188889Z 2025-09-09T15:23:22.6189206Z test/sparsity/test_sparse_api.py::TestBlockSparseWeight::test_sparse_compile_False_input_shape_1 2025-09-09T15:23:22.6190927Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/sparsity/blocksparse.py:198: UserWarning: Sparse BSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /pytorch/aten/src/ATen/SparseCsrTensorImpl.cpp:53.) 2025-09-09T15:23:22.6192464Z bsr_tensor = dense_tensor.to_sparse_bsr(blocksize) 2025-09-09T15:23:22.6192699Z 2025-09-09T15:23:22.6193018Z test/sparsity/test_sparse_api.py::TestBlockSparseWeight::test_sparse_compile_False_input_shape_1 2025-09-09T15:23:22.6194498Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/kernel/bsr_triton_ops.py:240: UserWarning: bsr_dense_addmm uses non-optimal triton kernel parameters for M=2048 K=1024 N=1 Ms=64, Ks=64 beta=0 alpha=1 dtype=torch.float16 out_dtype=torch.float16. To find optimal triton kernel parameters, run with BSR_AUTOTUNE=1 2025-09-09T15:23:22.6195753Z warn_once( 2025-09-09T15:23:22.6195902Z 2025-09-09T15:23:22.6196218Z test/sparsity/test_sparse_api.py::TestBlockSparseWeight::test_sparse_compile_False_input_shape_1 2025-09-09T15:23:22.6197827Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/kernel/bsr_triton_ops.py:240: UserWarning: bsr_dense_addmm uses non-optimal triton kernel parameters for M=1024 K=2048 N=1 Ms=64, Ks=64 beta=0 alpha=1 dtype=torch.float16 out_dtype=torch.float16. To find optimal triton kernel parameters, run with BSR_AUTOTUNE=1 2025-09-09T15:24:09.6119689Z warn_once( 2025-09-09T15:24:09.6120184Z 2025-09-09T15:24:09.6120580Z test/sparsity/test_sparse_api.py::TestBlockSparseWeight::test_sparse_compile_False_input_shape_1024 2025-09-09T15:24:09.6122130Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/kernel/bsr_triton_ops.py:240: UserWarning: bsr_dense_addmm uses non-optimal triton kernel parameters for M=2048 K=1024 N=1024 Ms=64, Ks=64 beta=0 alpha=1 dtype=torch.float16 out_dtype=torch.float16. To find optimal triton kernel parameters, run with BSR_AUTOTUNE=1 2025-09-09T15:24:09.6123387Z warn_once( 2025-09-09T15:24:09.6123521Z 2025-09-09T15:24:09.6123847Z test/sparsity/test_sparse_api.py::TestBlockSparseWeight::test_sparse_compile_False_input_shape_1024 2025-09-09T15:24:09.6125303Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/kernel/bsr_triton_ops.py:240: UserWarning: bsr_dense_addmm uses non-optimal triton kernel parameters for M=1024 K=2048 N=1024 Ms=64, Ks=64 beta=0 alpha=1 dtype=torch.float16 out_dtype=torch.float16. To find optimal triton kernel parameters, run with BSR_AUTOTUNE=1 2025-09-09T15:24:09.6126515Z warn_once( 2025-09-09T15:24:09.6126640Z 2025-09-09T15:24:09.6126929Z test/sparsity/test_sparse_api.py::TestQuantBlockSparseWeight::test_sparse_compile_False 2025-09-09T15:24:09.6128319Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/kernel/bsr_triton_ops.py:240: UserWarning: bsr_dense_addmm uses non-optimal triton kernel parameters for M=256 K=128 N=256 Ms=64, Ks=64 beta=0 alpha=1 dtype=torch.int8 out_dtype=torch.bfloat16. To find optimal triton kernel parameters, run with BSR_AUTOTUNE=1 2025-09-09T15:24:09.6129601Z warn_once( 2025-09-09T15:24:09.6129725Z 2025-09-09T15:24:09.6130011Z test/sparsity/test_sparse_api.py::TestQuantBlockSparseWeight::test_sparse_compile_False 2025-09-09T15:24:09.6131398Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/kernel/bsr_triton_ops.py:240: UserWarning: bsr_dense_addmm uses non-optimal triton kernel parameters for M=128 K=256 N=256 Ms=64, Ks=64 beta=0 alpha=1 dtype=torch.int8 out_dtype=torch.bfloat16. To find optimal triton kernel parameters, run with BSR_AUTOTUNE=1 2025-09-09T15:24:09.6132588Z warn_once( 2025-09-09T15:24:09.6132802Z 2025-09-09T15:24:09.6133036Z test/sparsity/test_wanda.py::TestWandaSparsifier::test_one_layer_mlp_2x4 2025-09-09T15:24:09.6133993Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/sparsity/wanda.py:46: UserWarning: WandaSparsifier got semi_structured_bock_size=4, sparsity_level fixed to 50% (2:4) sparsity 2025-09-09T15:24:09.6134901Z warnings.warn( 2025-09-09T15:24:09.6135038Z 2025-09-09T15:24:09.6135267Z -- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html 2025-09-09T15:24:09.6136316Z ======== 2101 passed, 3397 skipped, 210 warnings in 3852.10s (1:04:12) ========= 2025-09-09T15:24:09.6222776Z ##[group]Run pmeier/pytest-results-action@a2c1430e2bddadbad9f49a6f9b879f062c6b19b1 2025-09-09T15:24:09.6223307Z with: 2025-09-09T15:24:09.6223611Z path: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T15:24:09.6224019Z fail-on-empty: false 2025-09-09T15:24:09.6224254Z env: 2025-09-09T15:24:09.6224491Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T15:24:09.6224818Z REPOSITORY: pytorch/ao 2025-09-09T15:24:09.6225056Z PR_NUMBER: 2963 2025-09-09T15:24:09.6226850Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T15:24:09.6228815Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T15:24:09.6229346Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T15:24:09.6229928Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T15:24:09.6230342Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T15:24:09.6230668Z ##[endgroup] 2025-09-09T15:24:09.6850893Z Prepare all required actions 2025-09-09T15:24:09.6890321Z ##[group]Run ./test-infra/.github/actions/chown-directory 2025-09-09T15:24:09.6890645Z with: 2025-09-09T15:24:09.6890914Z directory: /home/ec2-user/actions-runner/_work/ao/ao/ 2025-09-09T15:24:09.6891372Z ALPINE_IMAGE: 308535385114.dkr.ecr.us-east-1.amazonaws.com/tool/alpine 2025-09-09T15:24:09.6891757Z env: 2025-09-09T15:24:09.6892000Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T15:24:09.6892340Z REPOSITORY: pytorch/ao 2025-09-09T15:24:09.6892597Z PR_NUMBER: 2963 2025-09-09T15:24:09.6894393Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T15:24:09.6896414Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T15:24:09.6897070Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T15:24:09.6897869Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T15:24:09.6898289Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T15:24:09.6898607Z ##[endgroup] 2025-09-09T15:24:09.6924070Z ##[group]Run docker run --rm -v "${DIRECTORY}":/v -w /v "${ALPINE_IMAGE}" chown -R "$(id -u):$(id -g)" . 2025-09-09T15:24:09.6924713Z docker run --rm -v "${DIRECTORY}":/v -w /v "${ALPINE_IMAGE}" chown -R "$(id -u):$(id -g)" . 2025-09-09T15:24:09.6939821Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T15:24:09.6940224Z env: 2025-09-09T15:24:09.6940537Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T15:24:09.6940971Z REPOSITORY: pytorch/ao 2025-09-09T15:24:09.6941401Z PR_NUMBER: 2963 2025-09-09T15:24:09.6943293Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T15:24:09.6945205Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T15:24:09.6945747Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T15:24:09.6946244Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T15:24:09.6946673Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T15:24:09.6947123Z ALPINE_IMAGE: 308535385114.dkr.ecr.us-east-1.amazonaws.com/tool/alpine 2025-09-09T15:24:09.6947578Z DIRECTORY: /home/ec2-user/actions-runner/_work/ao/ao/ 2025-09-09T15:24:09.6947904Z ##[endgroup] 2025-09-09T15:24:09.7206511Z Unable to find image '308535385114.dkr.ecr.us-east-1.amazonaws.com/tool/alpine:latest' locally 2025-09-09T15:24:09.9202141Z latest: Pulling from tool/alpine 2025-09-09T15:24:09.9202480Z 540db60ca938: Pulling fs layer 2025-09-09T15:24:10.0278196Z 540db60ca938: Verifying Checksum 2025-09-09T15:24:10.1442228Z 540db60ca938: Pull complete 2025-09-09T15:24:10.1554447Z Digest: sha256:def822f9851ca422481ec6fee59a9966f12b351c62ccb9aca841526ffaa9f748 2025-09-09T15:24:10.1596906Z Status: Downloaded newer image for 308535385114.dkr.ecr.us-east-1.amazonaws.com/tool/alpine:latest 2025-09-09T15:24:11.3039206Z Prepare all required actions 2025-09-09T15:24:11.3065512Z ##[group]Run ./test-infra/.github/actions/chown-directory 2025-09-09T15:24:11.3065841Z with: 2025-09-09T15:24:11.3066105Z directory: /home/ec2-user/actions-runner/_work/_temp 2025-09-09T15:24:11.3066564Z ALPINE_IMAGE: 308535385114.dkr.ecr.us-east-1.amazonaws.com/tool/alpine 2025-09-09T15:24:11.3066949Z env: 2025-09-09T15:24:11.3067206Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T15:24:11.3067533Z REPOSITORY: pytorch/ao 2025-09-09T15:24:11.3067781Z PR_NUMBER: 2963 2025-09-09T15:24:11.3069605Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T15:24:11.3071547Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T15:24:11.3072077Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T15:24:11.3072584Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T15:24:11.3073112Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T15:24:11.3073447Z ##[endgroup] 2025-09-09T15:24:11.3100404Z ##[group]Run docker run --rm -v "${DIRECTORY}":/v -w /v "${ALPINE_IMAGE}" chown -R "$(id -u):$(id -g)" . 2025-09-09T15:24:11.3101059Z docker run --rm -v "${DIRECTORY}":/v -w /v "${ALPINE_IMAGE}" chown -R "$(id -u):$(id -g)" . 2025-09-09T15:24:11.3115743Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T15:24:11.3116088Z env: 2025-09-09T15:24:11.3116334Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T15:24:11.3116676Z REPOSITORY: pytorch/ao 2025-09-09T15:24:11.3116928Z PR_NUMBER: 2963 2025-09-09T15:24:11.3118677Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T15:24:11.3120712Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T15:24:11.3121311Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T15:24:11.3121810Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T15:24:11.3122233Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T15:24:11.3122695Z ALPINE_IMAGE: 308535385114.dkr.ecr.us-east-1.amazonaws.com/tool/alpine 2025-09-09T15:24:11.3123141Z DIRECTORY: /home/ec2-user/actions-runner/_work/_temp 2025-09-09T15:24:11.3123462Z ##[endgroup] 2025-09-09T15:24:12.3498622Z ##[group]Run # Only do these steps if we actually want to upload an artifact 2025-09-09T15:24:12.3499194Z # Only do these steps if we actually want to upload an artifact 2025-09-09T15:24:12.3499620Z if [[ -n "${UPLOAD_ARTIFACT_NAME}" ]]; then 2025-09-09T15:24:12.3500124Z  # If the default execution path is followed then we should get a wheel in the dist/ folder 2025-09-09T15:24:12.3500674Z  # attempt to just grab whatever is in there and scoop it all up 2025-09-09T15:24:12.3501166Z  if find "dist/" -name "*.whl" >/dev/null 2>/dev/null; then 2025-09-09T15:24:12.3501556Z  mv -v dist/*.whl "${RUNNER_ARTIFACT_DIR}/" 2025-09-09T15:24:12.3501899Z  fi 2025-09-09T15:24:12.3502172Z  if [[ -d "artifacts-to-be-uploaded" ]]; then 2025-09-09T15:24:12.3502585Z  mv -v artifacts-to-be-uploaded/* "${RUNNER_ARTIFACT_DIR}/" 2025-09-09T15:24:12.3502951Z  fi 2025-09-09T15:24:12.3503172Z fi 2025-09-09T15:24:12.3503362Z  2025-09-09T15:24:12.3503568Z upload_docs=0 2025-09-09T15:24:12.3503932Z # Check if there are files in the documentation folder to upload, note that 2025-09-09T15:24:12.3504391Z # empty folders do not count 2025-09-09T15:24:12.3504805Z if find "${RUNNER_DOCS_DIR}" -mindepth 1 -maxdepth 1 -type f | read -r; then 2025-09-09T15:24:12.3505342Z  # TODO: Add a check here to test if on ec2 because if we're not on ec2 then this 2025-09-09T15:24:12.3505795Z  # upload will probably not work correctly 2025-09-09T15:24:12.3506105Z  upload_docs=1 2025-09-09T15:24:12.3506342Z fi 2025-09-09T15:24:12.3506630Z echo "upload-docs=${upload_docs}" >> "${GITHUB_OUTPUT}" 2025-09-09T15:24:12.3516481Z shell: /usr/bin/bash -e {0} 2025-09-09T15:24:12.3516734Z env: 2025-09-09T15:24:12.3516980Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T15:24:12.3517324Z REPOSITORY: pytorch/ao 2025-09-09T15:24:12.3517564Z PR_NUMBER: 2963 2025-09-09T15:24:12.3519325Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T15:24:12.3521409Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T15:24:12.3521989Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T15:24:12.3522484Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T15:24:12.3522906Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T15:24:12.3523242Z UPLOAD_ARTIFACT_NAME: 2025-09-09T15:24:12.3523577Z ##[endgroup] 2025-09-09T15:24:12.3674150Z Prepare all required actions 2025-09-09T15:24:12.3708781Z ##[group]Run ./test-infra/.github/actions/teardown-linux 2025-09-09T15:24:12.3709119Z with: 2025-09-09T15:24:12.3709309Z env: 2025-09-09T15:24:12.3709548Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T15:24:12.3709882Z REPOSITORY: pytorch/ao 2025-09-09T15:24:12.3710117Z PR_NUMBER: 2963 2025-09-09T15:24:12.3712055Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T15:24:12.3713993Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T15:24:12.3714525Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T15:24:12.3715020Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T15:24:12.3715445Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T15:24:12.3715763Z ##[endgroup] 2025-09-09T15:24:12.3737351Z ##[group]Run set -eou pipefail 2025-09-09T15:24:12.3737663Z set -eou pipefail 2025-09-09T15:24:12.3737911Z  2025-09-09T15:24:12.3738242Z echo "Holding runner for 2 hours until all ssh sessions have logged out" 2025-09-09T15:24:12.3738664Z for _ in $(seq 1440); do 2025-09-09T15:24:12.3738968Z  # Break if no ssh session exists anymore 2025-09-09T15:24:12.3739291Z  if [ "$(who)" = "" ]; then 2025-09-09T15:24:12.3739559Z  break 2025-09-09T15:24:12.3739782Z  fi 2025-09-09T15:24:12.3739999Z  echo "." 2025-09-09T15:24:12.3740233Z  sleep 5 2025-09-09T15:24:12.3740455Z done 2025-09-09T15:24:12.3749394Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T15:24:12.3749743Z env: 2025-09-09T15:24:12.3749990Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T15:24:12.3750327Z REPOSITORY: pytorch/ao 2025-09-09T15:24:12.3750570Z PR_NUMBER: 2963 2025-09-09T15:24:12.3752330Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T15:24:12.3754250Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T15:24:12.3754796Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T15:24:12.3755308Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T15:24:12.3755855Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T15:24:12.3756188Z ##[endgroup] 2025-09-09T15:24:12.3785625Z Holding runner for 2 hours until all ssh sessions have logged out 2025-09-09T15:24:12.3892487Z ##[group]Run # ignore expansion of "docker ps -q" since it could be empty 2025-09-09T15:24:12.3893048Z # ignore expansion of "docker ps -q" since it could be empty 2025-09-09T15:24:12.3893444Z # shellcheck disable=SC2046 2025-09-09T15:24:12.3893751Z docker stop $(docker ps -q) || true 2025-09-09T15:24:12.3894074Z # Prune all of the docker images 2025-09-09T15:24:12.3894377Z docker system prune -af 2025-09-09T15:24:12.3903845Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T15:24:12.3904194Z env: 2025-09-09T15:24:12.3904743Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T15:24:12.3905080Z REPOSITORY: pytorch/ao 2025-09-09T15:24:12.3905334Z PR_NUMBER: 2963 2025-09-09T15:24:12.3907098Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T15:24:12.3909017Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T15:24:12.3909548Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T15:24:12.3910049Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T15:24:12.3910537Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T15:24:12.3910867Z ##[endgroup] 2025-09-09T15:24:13.7367360Z c429987d1971 2025-09-09T15:24:21.5279504Z Deleted Containers: 2025-09-09T15:24:21.5280025Z c429987d197112cb1c1dca45efc7b01c8ed17990b5489cf4c359012ede81fdb6 2025-09-09T15:24:21.5280420Z 2025-09-09T15:24:29.2011687Z Deleted Images: 2025-09-09T15:24:29.2012051Z untagged: public.ecr.aws/docker/library/python:3.13 2025-09-09T15:24:29.2012848Z untagged: public.ecr.aws/docker/library/python@sha256:74503e0bff6cf811f029590a05e0218cc9ba3e099a4b7df0ab84a67df081e1bc 2025-09-09T15:24:29.2013681Z deleted: sha256:77f2b24be2b3987f6d59918787d226acb4e6612644bacb3dd37adc494e477d9e 2025-09-09T15:24:29.2014314Z deleted: sha256:1b9aa91044866f8707424c8fe367f924a48557eac69f7485fd6d2a3a116c74d5 2025-09-09T15:24:29.2014882Z deleted: sha256:b86402d18e73d4825a3bd2a09244a93487ba4687ca7c9dcba0f73e160840845c 2025-09-09T15:24:29.2015472Z deleted: sha256:5755f8963eb047a0086073c3a7dd0731296d6751a7445f3693a52b30020a5b65 2025-09-09T15:24:29.2016107Z deleted: sha256:7f33dbfa9475d25622f49ed51f4164c97de1303331c77dfdc738e084d100f50c 2025-09-09T15:24:29.2016681Z deleted: sha256:19daa38049795ba2c166dd898c81b17e31f4b5f98c1337846c6515fff97d8782 2025-09-09T15:24:29.2017288Z deleted: sha256:483bd23b5d7e66fc0f8a92dbfacc3d72fad97ef47dd4767889979a803bc1f5b8 2025-09-09T15:24:29.2017906Z deleted: sha256:185e04da9d947141fd703dbf36361bdc2ff77cc27cbf500fb9f4881cb5ddbe95 2025-09-09T15:24:29.2018483Z untagged: 308535385114.dkr.ecr.us-east-1.amazonaws.com/tool/alpine:latest 2025-09-09T15:24:29.2019255Z untagged: 308535385114.dkr.ecr.us-east-1.amazonaws.com/tool/alpine@sha256:def822f9851ca422481ec6fee59a9966f12b351c62ccb9aca841526ffaa9f748 2025-09-09T15:24:29.2020062Z deleted: sha256:6dbb9cc54074106d46d4ccb330f2a40a682d49dda5f4844962b7dce9fe44aaec 2025-09-09T15:24:29.2020666Z deleted: sha256:b2d5eeeaba3a22b9b8aa97261957974a6bd65274ebd43e1d81d0a7b8b752b116 2025-09-09T15:24:29.2021150Z untagged: pytorch/almalinux-builder:cuda12.6 2025-09-09T15:24:29.2021733Z untagged: pytorch/almalinux-builder@sha256:be7f2a4c6f467933b154ac0b3ded894ad1bf06ce95f8f8d908dba108e68806f3 2025-09-09T15:24:29.2022652Z deleted: sha256:f2a1aa8f0f7d42816c0fe19d8838a50734adb5eb5de36e7f04cffdbf8de9d63c 2025-09-09T15:24:29.2023241Z deleted: sha256:c9e150f3add3ad75da73e417a090d5b5fc1b81cc89f87daa049075292636557c 2025-09-09T15:24:29.2023829Z deleted: sha256:a34ecd86569aa00449c46f68a40f7e90b90cc6a7e7d286262816aee915322a7b 2025-09-09T15:24:29.2024401Z deleted: sha256:892ada8183495e70a17e9cfa280133ba3bc836b5e37564e0bbe9d7f8234a3848 2025-09-09T15:24:29.2024990Z deleted: sha256:eadbce3bbb3f1a27ce48c9c43e762dc84723582ebdf7fcaf004049bd5efebcd4 2025-09-09T15:24:29.2025578Z deleted: sha256:614eb17bcea9533c6fafc9f12884904851ce23b0a4609d3e6eb24ebccee3a49d 2025-09-09T15:24:29.2026168Z deleted: sha256:5d7a767e6e9b52e3bd6ae5896a9c32ca564f0880f3bec9eb38678eec35ebf104 2025-09-09T15:24:29.2026747Z deleted: sha256:c375116670abbf5fd150e1ee6ee8f0ea2d14351ec3e6a55e2a897f23ada5cfd6 2025-09-09T15:24:29.2027706Z deleted: sha256:9bd5a8ab14bca745f798e5e88495f278c13d3fc531c4b668c5c8f7c91d334f9b 2025-09-09T15:24:29.2028292Z deleted: sha256:4dc4b770bc6212696f6b304b6c20f9c2c34f899bc1a36629fd3595ea776b846f 2025-09-09T15:24:29.2028870Z deleted: sha256:baf10146ebddd70194ce3977095cc8ed0e6d7df2f88979af5b49262c60a3f3e1 2025-09-09T15:24:29.2029450Z deleted: sha256:4ab11f18df8904dad268820aa8d575115da52680d411d7559440c0b5bc04b0f1 2025-09-09T15:24:29.2030021Z deleted: sha256:3597cd684ef75bc762bbad866d470bde44719232d96ccf3cfcda286583915b14 2025-09-09T15:24:29.2030610Z deleted: sha256:1beac1c8161bfd0d6ac4d67b32f665ccfb1b312f7af82b05d30ac79551275838 2025-09-09T15:24:29.2031208Z deleted: sha256:3391cdf48fdf7b9acbb1ac360ba1210ca8a8d727313aa25f38a977087f547d99 2025-09-09T15:24:29.2031794Z deleted: sha256:45cdedd84abf879e7a255cb152d814c309dbdade3f6a917f6bd54d76277d71d6 2025-09-09T15:24:29.2032379Z deleted: sha256:3e51a87c61a2c818401711cbc3a136cfaa58de4c8453602a012cbdda61e20cd8 2025-09-09T15:24:29.2032960Z deleted: sha256:c494b964f63b3ad12c0b2478a96315d32ddb3d60869a8ff249c5f70b78918f47 2025-09-09T15:24:29.2033627Z deleted: sha256:7dd5e41bb10ca14c9e763dd5cbb9e15c7a4727c5cbc49b40cfd262872ea7b3dd 2025-09-09T15:24:29.2034336Z deleted: sha256:0ebf188609b1c3acf20d30c01cd9d08f623b552df32b6bbd9c9ea72c713cbda7 2025-09-09T15:24:29.2034917Z deleted: sha256:46a0108e49ebe8b5c8a25a046961a83a96de1d1d2b3c83fdb1cce9c24033d3f3 2025-09-09T15:24:29.2035495Z deleted: sha256:add74a43e7c81f71d6040c8c766aa51293da349336c272faccf238aac3218502 2025-09-09T15:24:29.2036066Z deleted: sha256:ebe35679e50c08c29cc5c163b411dcd19c66cbb41714d123f4eaebcd8c836766 2025-09-09T15:24:29.2036649Z deleted: sha256:ea669f8859ecce20d165c51dd89be8761f0ad6f7e899116e386a4f9dd403b565 2025-09-09T15:24:29.2037216Z deleted: sha256:ff4f19608a1944c0c2807cd533515673285a9632dc74bf020e83e18630d1ae35 2025-09-09T15:24:29.2037555Z 2025-09-09T15:24:29.2037665Z Total reclaimed space: 28.29GB 2025-09-09T15:24:29.2125297Z ##[group]Run set +e 2025-09-09T15:24:29.2125566Z set +e 2025-09-09T15:24:29.2125809Z if [[ "${NO_SUDO}" == "false" ]]; then 2025-09-09T15:24:29.2126178Z  sudo rm -rf "${GITHUB_WORKSPACE:?}/${REPOSITORY:?}" 2025-09-09T15:24:29.2126519Z else 2025-09-09T15:24:29.2126781Z  rm -rf "${GITHUB_WORKSPACE:?}/${REPOSITORY:?}" 2025-09-09T15:24:29.2127088Z fi 2025-09-09T15:24:29.2127287Z set -e 2025-09-09T15:24:29.2136262Z shell: /usr/bin/bash -e {0} 2025-09-09T15:24:29.2136512Z env: 2025-09-09T15:24:29.2136759Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T15:24:29.2137093Z REPOSITORY: pytorch/ao 2025-09-09T15:24:29.2137360Z PR_NUMBER: 2963 2025-09-09T15:24:29.2139170Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv echo "::group::Install newer objcopy that supports --set-section-alignment" dnf install -y gcc-toolset-10-binutils export PATH=/opt/rh/gcc-toolset-10/root/usr/bin/:$PATH python -m pip install --upgrade pip pip install torch==2.7.0 sed -i '' dev-requirements.txt pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T15:24:29.2141186Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T15:24:29.2141715Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T15:24:29.2142215Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T15:24:29.2142634Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T15:24:29.2142966Z NO_SUDO: false 2025-09-09T15:24:29.2143178Z ##[endgroup] 2025-09-09T15:24:29.7347454Z Post job cleanup. 2025-09-09T15:24:29.8435564Z Post job cleanup. 2025-09-09T15:24:29.9449542Z [command]/usr/bin/git version 2025-09-09T15:24:29.9499558Z git version 2.47.1 2025-09-09T15:24:29.9544547Z Temporarily overriding HOME='/home/ec2-user/actions-runner/_work/_temp/dec619b3-d3ac-4c1c-870f-1da1dd6e6ca8' before making global git config changes 2025-09-09T15:24:29.9545546Z Adding repository directory to the temporary git global config as a safe directory 2025-09-09T15:24:29.9549993Z [command]/usr/bin/git config --global --add safe.directory /home/ec2-user/actions-runner/_work/ao/ao/test-infra 2025-09-09T15:24:29.9592955Z [command]/usr/bin/git config --local --name-only --get-regexp core\.sshCommand 2025-09-09T15:24:29.9633204Z [command]/usr/bin/git submodule foreach --recursive sh -c "git config --local --name-only --get-regexp 'core\.sshCommand' && git config --local --unset-all 'core.sshCommand' || :" 2025-09-09T15:24:30.0064814Z [command]/usr/bin/git config --local --name-only --get-regexp http\.https\:\/\/github\.com\/\.extraheader 2025-09-09T15:24:30.0094161Z http.https://github.com/.extraheader 2025-09-09T15:24:30.0108049Z [command]/usr/bin/git config --local --unset-all http.https://github.com/.extraheader 2025-09-09T15:24:30.0145343Z [command]/usr/bin/git submodule foreach --recursive sh -c "git config --local --name-only --get-regexp 'http\.https\:\/\/github\.com\/\.extraheader' && git config --local --unset-all 'http.https://github.com/.extraheader' || :" 2025-09-09T15:24:30.0644555Z A job completed hook has been configured by the self-hosted runner administrator 2025-09-09T15:24:30.0676022Z ##[group]Run '/home/ec2-user/runner-scripts/after_job.sh' 2025-09-09T15:24:30.0684165Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T15:24:30.0684536Z ##[endgroup] 2025-09-09T15:24:30.0804856Z [!ALERT!] Swap in detected! [!ALERT!] 2025-09-09T15:24:41.4874603Z [!ALERT!] Swap out detected [!ALERT!] 2025-09-09T15:24:59.8022785Z Cleaning up orphan processes