benchmark_test #45
This file contains 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: benchmark_test | |
| on: | |
| workflow_dispatch: | |
| inputs: | |
| repo_org: | |
| required: false | |
| description: 'Tested repository organization name. Default is InternLM' | |
| type: string | |
| default: 'InternLM/lmdeploy' | |
| repo_ref: | |
| required: false | |
| description: 'Set branch or tag or commit id. Default is "main"' | |
| type: string | |
| default: 'main' | |
| benchmark_type: | |
| required: true | |
| description: 'Set benchmark type. Default is "["generation", "throughtput", "api_server"]"' | |
| type: string | |
| default: "['apiserver', 'generation', 'throughput']" | |
| offline_mode: | |
| required: true | |
| description: 'Whether start a offline mode, if true, you should prepare code and whl package by yourself' | |
| type: boolean | |
| default: false | |
| env: | |
| HOST_PIP_CACHE_DIR: /nvme/github-actions/pip-cache | |
| HOST_LOCALTIME: /usr/share/zoneinfo/Asia/Shanghai | |
| OUTPUT_FOLDER: cuda11.8_dist_${{ github.run_id }} | |
| REPORT_DIR: /nvme/qa_test_models/benchmark-reports/${{ github.run_id }} | |
| ACTIONS_ALLOW_USE_UNSECURE_NODE_VERSION: true | |
| FAIL_CONFIG: ${{ github.run_attempt != 1 && '--lf --lfnf none' || '--lf'}} | |
| jobs: | |
| linux-build: | |
| if: ${{github.event_name == 'schedule' || (!cancelled() && !inputs.offline_mode)}} | |
| strategy: | |
| matrix: | |
| pyver: [py310] | |
| runs-on: ubuntu-latest | |
| env: | |
| PYTHON_VERSION: ${{ matrix.pyver }} | |
| PLAT_NAME: manylinux2014_x86_64 | |
| DOCKER_TAG: cuda11.8 | |
| steps: | |
| - name: Checkout repository | |
| uses: actions/checkout@v3 | |
| with: | |
| repository: ${{ github.event.inputs.repo_org || 'InternLM/lmdeploy' }} | |
| ref: ${{github.event.inputs.repo_ref || 'main'}} | |
| - name: Build | |
| run: | | |
| echo ${PYTHON_VERSION} | |
| echo ${PLAT_NAME} | |
| echo ${DOCKER_TAG} | |
| echo ${OUTPUT_FOLDER} | |
| echo ${GITHUB_RUN_ID} | |
| # remove -it | |
| sed -i 's/docker run --rm -it/docker run --rm/g' builder/manywheel/build_wheel.sh | |
| bash builder/manywheel/build_wheel.sh ${PYTHON_VERSION} ${PLAT_NAME} ${DOCKER_TAG} ${OUTPUT_FOLDER} | |
| - name: Upload Artifacts | |
| uses: actions/upload-artifact@v4 | |
| with: | |
| if-no-files-found: error | |
| path: builder/manywheel/${{ env.OUTPUT_FOLDER }} | |
| retention-days: 1 | |
| name: my-artifact-${{ github.run_id }}-${{ matrix.pyver }} | |
| benchmark: | |
| needs: linux-build | |
| if: ${{github.event_name == 'schedule' || !cancelled()}} | |
| runs-on: [self-hosted, linux-a100] | |
| strategy: | |
| fail-fast: false | |
| matrix: | |
| benchmark_type: ${{fromJSON(github.event.inputs.benchmark_type)}} | |
| timeout-minutes: 480 | |
| container: | |
| image: openmmlab/lmdeploy:latest-cu11 | |
| options: "--gpus=all --ipc=host --user root -e PIP_CACHE_DIR=/root/.cache/pip -e NVIDIA_DISABLE_REQUIRE=1 --pull never" | |
| volumes: | |
| - /nvme/github-actions/pip-cache:/root/.cache/pip | |
| - /nvme/github-actions/packages:/root/packages | |
| - /nvme/qa_test_models:/nvme/qa_test_models | |
| - /mnt/shared:/mnt/shared | |
| - /usr/share/zoneinfo/Asia/Shanghai:/etc/localtime:ro | |
| steps: | |
| - name: Clone repository | |
| uses: actions/checkout@v3 | |
| if: ${{github.event_name == 'schedule' || !inputs.offline_mode}} | |
| with: | |
| repository: ${{ github.event.inputs.repo_org || 'InternLM/lmdeploy' }} | |
| ref: ${{github.event.inputs.repo_ref || 'main'}} | |
| - name: Copy repository - offline | |
| if: ${{inputs.offline_mode}} | |
| run: cp -r /nvme/qa_test_models/offline_pkg/lmdeploy/. . | |
| - name: Download Artifacts | |
| if: ${{github.event_name == 'schedule' || !inputs.offline_mode}} | |
| uses: actions/download-artifact@v4 | |
| with: | |
| name: my-artifact-${{ github.run_id }}-py310 | |
| - name: Install lmdeploy - dependency | |
| run: | | |
| # manually install flash attn | |
| # the install packeage from. https://github.com/Dao-AILab/flash-attention/releases | |
| python3 -m pip install /root/packages/flash_attn-2.6.3+cu118torch2.3cxx11abiFALSE-cp310-cp310-linux_x86_64.whl | |
| python3 -m pip install -e /root/packages/AutoAWQ_kernels | |
| python3 -m pip install /root/packages/autoawq-0.2.6-cp310-cp310-manylinux2014_x86_64.whl --no-deps | |
| python3 -m pip install /root/packages/xformers-0.0.27+cu118-cp310-cp310-manylinux2014_x86_64.whl --no-deps | |
| python3 -m pip install -r /nvme/qa_test_models/offline_pkg/requirements.txt | |
| - name: Install lmdeploy | |
| if: ${{github.event_name == 'schedule' || !inputs.offline_mode}} | |
| run: | | |
| python3 -m pip install lmdeploy-*.whl --no-deps | |
| python3 -m pip install -r requirements/test.txt | |
| - name: Install lmdeploy - offline | |
| if: ${{inputs.offline_mode}} | |
| run: | | |
| python3 -m pip install /nvme/qa_test_models/offline_pkg/py310/lmdeploy-*.whl --no-deps | |
| python3 -m pip install -r requirements/test.txt | |
| - name: Check env | |
| run: | | |
| pip uninstall -y nvidia-nccl-cu11 | |
| python3 -m pip list | |
| lmdeploy check_env | |
| mkdir ${{env.REPORT_DIR}}/allure-results/.pytest_cache -p | |
| ln -s ${{env.REPORT_DIR}}/allure-results/.pytest_cache autotest | |
| - name: Run other benchmark | |
| run: | | |
| pytest autotest/benchmark/test_${{matrix.benchmark_type}}_performance.py -n 8 --run_id ${{ github.run_id }} -m gpu_num_1 ${{env.FAIL_CONFIG}} --alluredir=${{env.REPORT_DIR}}/allure-results || true | |
| pytest autotest/benchmark/test_${{matrix.benchmark_type}}_performance.py -n 4 --run_id ${{ github.run_id }} -m gpu_num_2 ${{env.FAIL_CONFIG}} --alluredir=${{env.REPORT_DIR}}/allure-results || true | |
| pytest autotest/benchmark/test_${{matrix.benchmark_type}}_performance.py -n 2 --run_id ${{ github.run_id }} -m gpu_num_4 ${{env.FAIL_CONFIG}} --alluredir=${{env.REPORT_DIR}}/allure-results | |
| - name: Clear workfile | |
| if: always() | |
| run: | | |
| chmod -R 777 $REPORT_DIR | |
| export workdir=$(pwd) | |
| cd .. | |
| rm -rf $workdir | |
| mkdir $workdir | |
| chmod -R 777 $workdir | |
| get_result_overview: | |
| if: always() && !cancelled() | |
| needs: [benchmark] | |
| timeout-minutes: 5 | |
| runs-on: [self-hosted, linux-a100] | |
| steps: | |
| - name: Clone repository | |
| uses: actions/checkout@v3 | |
| with: | |
| repository: ${{ github.event.inputs.repo_org || 'InternLM/lmdeploy' }} | |
| ref: ${{github.event.inputs.repo_ref || 'main'}} | |
| - name: Get overview | |
| run: | | |
| pip install pandas fire mmengine | |
| python3 .github/scripts/action_tools.py generate_benchmark_report $REPORT_DIR |