Unittest GPU CE #30
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: Unittest GPU CE | |
| on: | |
| schedule: | |
| - cron: "0 1 * * *" # every day at 09:00 Beijing time (UTC+8) | |
| workflow_dispatch: # allow to manually trigger the workflow | |
| inputs: | |
| paddle_whl: | |
| description: "paddle_whl" | |
| required: false | |
| default: "https://paddle-qa.bj.bcebos.com/paddle-pipeline/Develop-GpuAll-LinuxCentos-Gcc11-Cuda126-Cudnn95-Trt105-Py310-Compile/latest/paddlepaddle_gpu-0.0.0-cp310-cp310-linux_x86_64.whl" | |
| type: string | |
| repo_branch: | |
| description: "repo_branch" | |
| required: false | |
| type: string | |
| concurrency: | |
| group: unittest-ce-${{ github.workflow }} | |
| cancel-in-progress: true | |
| env: | |
| COMMIT_ID: ${{ github.event.pull_request.head.sha }} | |
| TASK: PaddleFormers-CE-unittest-gpu | |
| ce_scripts: /workspace/PaddleFormers/scripts/unit_test | |
| BRANCH: ${{ github.event.pull_request.base.ref }} | |
| AGILE_COMPILE_BRANCH: ${{ github.event.pull_request.base.ref }} | |
| CE_name: unittest-gpu-ce | |
| no_proxy: "localhost,bj.bcebos.com,su.bcebos.com,bcebos.com,apiin.im.baidu.com,gitee.com,aliyun.com,.baidu.com,.tuna.tsinghua.edu.cn" | |
| defaults: | |
| run: | |
| shell: bash | |
| jobs: | |
| unittest-gpu-ce: | |
| name: unittest-gpu-ce | |
| runs-on: [self-hosted, ernie-8gpu] | |
| steps: | |
| - name: Determine Image Name | |
| run: | | |
| echo "IMAGE_NAME=ccr-2vdh3abv-pub.cnc.bj.baidubce.com/paddlepaddle/paddle:cuda126-dev-latest" >> "$GITHUB_ENV" | |
| - name: Run Container | |
| env: | |
| work_dir: ${{ github.workspace }} | |
| FLAGS_dynamic_static_unified_comm: "True" | |
| python_version: "3.10" | |
| paddle_whl: ${{ github.event.inputs.paddle_whl || 'https://paddle-qa.bj.bcebos.com/paddle-pipeline/Develop-GpuAll-LinuxCentos-Gcc11-Cuda126-Cudnn95-Trt105-Py310-Compile/latest/paddlepaddle_gpu-0.0.0-cp310-cp310-linux_x86_64.whl' }} | |
| repo_branch: ${{ github.event.inputs.repo_branch || '' }} | |
| run: | | |
| container_name=${TASK}-$(date +%Y%m%d-%H%M%S) | |
| echo "container_name=${container_name}" >> "$GITHUB_ENV" | |
| echo "Workspace path: ${{ github.workspace }}" | |
| DOCKER_VER=$(docker version --format '{{.Server.Version}}' | cut -d. -f1,2) | |
| if (( $(echo "$DOCKER_VER < 19.03" | bc -l) )); then | |
| GPU_OPTION="--runtime=nvidia" | |
| else | |
| GPU_OPTION="--gpus all" | |
| fi | |
| echo "DOCKER_VER=${DOCKER_VER}" | |
| echo "GPU_OPTION=${GPU_OPTION}" | |
| docker run -d -t --name ${container_name} --net=host -v /dev/shm:/dev/shm --shm-size=32G \ | |
| -v $work_dir/../../..:$work_dir/../../.. \ | |
| -v $work_dir:/workspace \ | |
| -v /home/.cache/pip:/home/.cache/pip \ | |
| -v /home/paddle-1/models/:/home/models/ \ | |
| -e PF_HOME=/home/models/ \ | |
| -e BRANCH \ | |
| -e AGILE_COMPILE_BRANCH \ | |
| -e COMMIT_ID \ | |
| -e work_dir \ | |
| -e ce_scripts \ | |
| -e no_proxy \ | |
| -e CE_name \ | |
| -e paddle_whl \ | |
| -e repo_branch \ | |
| -e FLAGS_dynamic_static_unified_comm \ | |
| -e python_version \ | |
| -e HF_PROXY_PATH=$work_dir/../../../proxy_huggingface \ | |
| -e AISTUDIO_PROXY_PATH=$work_dir/../../../proxy_aistudio \ | |
| -e "HF_DATASETS_CACHE=$work_dir/../../../paddlenlp/huggingface/datasets" \ | |
| -e "TRANSFORMERS_CACHE=$work_dir/../../../paddlenlp/huggingface" \ | |
| -w /workspace ${GPU_OPTION} --privileged $IMAGE_NAME | |
| - name: Download Code | |
| run: | | |
| docker exec -t $container_name /bin/bash -c ' | |
| rm -rf * .[^.]* | |
| echo "Downloading PaddleFormers.tar" | |
| wget -q --no-proxy https://paddle-qa.bj.bcebos.com/CodeSync/develop/PaddleFormers.tar --no-check-certificate | |
| echo "Extracting PaddleFormers.tar" | |
| tar xf PaddleFormers.tar && rm -rf PaddleFormers.tar | |
| source $work_dir/../../../proxy | |
| cd PaddleFormers | |
| git config --global user.name "PaddleCE" | |
| git config --global user.email "[email protected]" | |
| if [ -n "$repo_branch" ]; then | |
| echo "Switching to branch: $repo_branch" | |
| git fetch origin $repo_branch | |
| git checkout $repo_branch || git checkout -b $repo_branch origin/$repo_branch | |
| git pull | |
| else | |
| echo "No repo_branch provided, just pulling latest changes" | |
| git pull | |
| fi | |
| git submodule update --init --recursive --force | |
| git log --pretty=oneline -10 | |
| ' | |
| - name: Test | |
| run: | | |
| docker exec -t $container_name /bin/bash -c ' | |
| ldconfig | |
| pip config set global.cache-dir "/home/.cache/pip" | |
| set -e | |
| rm -rf /root/.cache/aistudio/ | |
| cd /workspace/PaddleFormers && git config --global --add safe.directory $PWD | |
| source $work_dir/../../../proxy | |
| source $work_dir/../../../AISTUDIO_ACCESS_TOKEN | |
| echo "work_dir = ${work_dir}" | |
| cp -r ${work_dir}/../../../models ./models | |
| echo "Check whether the local model file exists:" | |
| ls -l ./models | |
| timeout 40m bash scripts/unit_test/ci_unittest.sh ${paddle_whl} true | |
| ' | |
| - name: Upload Allure-reports & Logs | |
| if: always() | |
| env: | |
| home_path: ${{ github.workspace }}/../../.. | |
| bos_file: ${{ github.workspace }}/../../../bos/BosClient.py | |
| allure_file: ${{ github.workspace }}/../../../allure-2.19.0/bin/allure | |
| run: | | |
| docker exec -t $container_name /bin/bash -c ' | |
| if [ ! -f "${{ env.bos_file }}" ]; then | |
| wget -q --no-proxy -O ${{ env.home_path }}/bos_new.tar.gz https://xly-devops.bj.bcebos.com/home/bos_new.tar.gz --no-check-certificate | |
| mkdir ${{ env.home_path }}/bos | |
| tar xf ${{ env.home_path }}/bos_new.tar.gz -C ${{ env.home_path }}/bos | |
| fi | |
| # if [ ! -f "${{ env.allure_file }}" ]; then | |
| # wget -q --no-proxy -O ${{ env.home_path }}/allure-2.19.0.zip https://xly-devops.bj.bcebos.com/tools/allure-2.19.0.zip --no-check-certificate | |
| # unzip -q ${{ env.home_path }}/allure-2.19.0.zip | |
| # fi | |
| bos_prefix="schedule/github-ce-$(date +%Y%m%d)" | |
| # # coverage.xml | |
| # cd /workspace/PaddleFormers | |
| # python ${{ env.bos_file }} coverage.xml paddle-github-action/PR/PaddleFormers/unittest-gpu/${bos_prefix}/logs | |
| # echo "cov-report: https://paddle-github-action.bj.bcebos.com/PR/PaddleFormers/unittest-gpu/${bos_prefix}/logs/coverage.xml" | |
| # logs | |
| cd /workspace/PaddleFormers/unittest_logs | |
| for FILE in /workspace/PaddleFormers/unittest_logs/*; do | |
| file=$(basename "$FILE") | |
| python ${{ env.bos_file }} $file paddle-github-action/PR/PaddleFormers/unittest-gpu/${bos_prefix}/logs | |
| echo "$file: https://paddle-github-action.bj.bcebos.com/PR/PaddleFormers/unittest-gpu/${bos_prefix}/logs/$file" | |
| done | |
| # cd /workspace/PaddleFormers/ | |
| # ${{ env.allure_file }} generate result -o report | |
| # tar -czf products.tar.gz report unittest_logs | |
| # python ${{ env.bos_file }} products.tar.gz paddle-github-action/PR/PaddleFormers/unittest-gpu/${bos_prefix}/logs | |
| # echo "report: https://paddle-github-action.bj.bcebos.com/PR/PaddleFormers/unittest-gpu/${bos_prefix}/logs/products.tar.gz" | |
| ' | |
| - name: Terminate And Delete the Container | |
| if: always() | |
| run: | | |
| docker rm -f $container_name 2>/dev/null || true |