Ragaaf - adding new metric 'context relevance' #183
Workflow file for this run
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
| # Copyright (C) 2024 Intel Corporation | |
| # SPDX-License-Identifier: Apache-2.0 | |
| name: Model Test on CPU | |
| on: | |
| pull_request: | |
| branches: [main] | |
| types: [opened, reopened, ready_for_review, synchronize] # added `ready_for_review` since draft is skipped | |
| paths: | |
| - .github/workflows/model_test_cpu.yml | |
| - evals/evaluation/** | |
| - evals/metrics/** | |
| - setup.py | |
| workflow_dispatch: | |
| # If there is a new commit, the previous jobs will be canceled | |
| concurrency: | |
| group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }} | |
| cancel-in-progress: true | |
| permissions: write-all | |
| env: | |
| OUT_SCRIPT_PATH: ${{ github.workspace }}/.github/workflows/scripts/models | |
| SCRIPT_PATH: /GenAIEval/.github/workflows/scripts | |
| DOCKER_NAME: "genaieval" | |
| DOCKER_TAG: "latest" | |
| CONTAINER_NAME: "modelTest" | |
| jobs: | |
| Evaluation-Workflow: | |
| runs-on: aise-cluster-cpu | |
| strategy: | |
| matrix: | |
| include: | |
| - modelName: "opt-125m" | |
| datasets: "lambada_openai" | |
| device: "cpu" | |
| tasks: "text-generation" | |
| fail-fast: true | |
| steps: | |
| - name: Clean Up Working Directory | |
| run: sudo rm -rf ${{github.workspace}}/* | |
| - name: Checkout out Repo | |
| uses: actions/checkout@v4 | |
| with: | |
| submodules: "recursive" | |
| fetch-tags: true | |
| # We need this because GitHub needs to clone the branch to pipeline | |
| - name: Docker Build | |
| run: | | |
| docker build -f ${{ github.workspace }}/.github/workflows/docker/common.dockerfile -t ${{ env.DOCKER_NAME }}:${{ env.DOCKER_TAG }} . | |
| - name: Docker Run | |
| run: | | |
| if [[ $(docker ps -a | grep -i '${{ env.CONTAINER_NAME }}'$) ]]; then | |
| docker stop ${{ env.CONTAINER_NAME }} | |
| docker rm -vf ${{ env.CONTAINER_NAME }} || true | |
| fi | |
| docker run -dit --disable-content-trust --privileged --name=${{ env.CONTAINER_NAME }} -v /dev/shm:/dev/shm \ | |
| -v ${{ github.workspace }}:/GenAIEval ${{ env.DOCKER_NAME }}:${{ env.DOCKER_TAG }} | |
| - name: Binary build | |
| run: | | |
| docker exec ${{ env.CONTAINER_NAME }} \ | |
| bash -c "cd /GenAIEval && pip install -r requirements.txt && python setup.py install" | |
| - name: Evaluation | |
| run: | | |
| docker exec ${{ env.CONTAINER_NAME }} \ | |
| bash -c "cd /GenAIEval/.github/workflows/scripts/models \ | |
| && bash -x model_test.sh --model=${{ matrix.modelName }} --device=${{ matrix.device }} --datasets=${{ matrix.datasets }} --tasks=${{ matrix.tasks }}" | |
| - name: Collect Log | |
| run: | | |
| docker exec ${{ env.CONTAINER_NAME }} \ | |
| bash -c "cd /GenAIEval/.github/workflows/scripts/models \ | |
| && bash -x collect_log.sh --model=${{ matrix.modelName }} \ | |
| --device=${{ matrix.device }} \ | |
| --datasets=${{ matrix.datasets }} \ | |
| --tasks=${{ matrix.tasks }}" | |
| - name: Publish pipeline artifact | |
| uses: actions/upload-artifact@v4 | |
| if: ${{ !cancelled() }} | |
| with: | |
| name: ${{ matrix.device }}-${{ matrix.tasks }}-${{ matrix.modelName }}-${{ matrix.datasets }} | |
| path: | | |
| ${{ github.workspace }}/${{ matrix.device }}-${{ matrix.tasks }}-${{ matrix.modelName }}-${{ matrix.datasets }}.log | |
| ${{ github.workspace }}/summary.log | |
| if-no-files-found: ignore # 'warn' or 'ignore' are also available, defaults to `warn` | |
| retention-days: 60 # 1 <= retention-days <= 90 | |
| Genreate-Report: | |
| runs-on: ubuntu-latest | |
| needs: [Evaluation-Workflow] | |
| steps: | |
| - name: Checkout out Repo | |
| uses: actions/checkout@v4 | |
| - name: Download Summary Log | |
| uses: actions/download-artifact@v4 | |
| with: | |
| path: ${{ env.OUT_SCRIPT_PATH }}/log | |
| - name: Display structure of downloaded files | |
| run: ls -R | |
| - name: Analysis Summary | |
| run: | | |
| cd ${{ env.OUT_SCRIPT_PATH }} | |
| ls -R | |
| - name: Download Reference Artifact | |
| id: download-artifact | |
| uses: dawidd6/[email protected] | |
| with: | |
| workflow: model_test_cpu.yml | |
| name: FinalReport | |
| run_id: ${{ vars.ModelTest_CPU_REF_ID }} | |
| path: ${{ env.OUT_SCRIPT_PATH }} | |
| name_is_regexp: true | |
| repo: ${{ github.repository }} | |
| check_artifacts: false | |
| search_artifacts: false | |
| skip_unpack: false | |
| if_no_artifact_found: warn | |
| - name: Display structure of downloaded files | |
| run: cd ${{ env.OUT_SCRIPT_PATH }}/log && ls -R | |
| - name: Generate report | |
| run: | | |
| echo "------ Generating final report.html ------" | |
| cd ${{ env.OUT_SCRIPT_PATH }} | |
| mkdir -p generated | |
| /usr/bin/bash -x generate_report.sh | |
| env: | |
| RUN_DISPLAY_URL: https://https://github.com/opea-project/GenAIEval/actions/runs/${{ github.run_id }} | |
| BUILD_NUMBER: ${{ github.run_id }} | |
| JOB_STATUS: succeed | |
| - name: Publish Report | |
| uses: actions/upload-artifact@v4 | |
| if: ${{ !cancelled() }} | |
| with: | |
| name: FinalReport | |
| path: ${{ env.OUT_SCRIPT_PATH }}/generated | |
| - name: Specify performance regression | |
| if: ${{ !cancelled() }} | |
| run: | | |
| if [ ${{ env.is_perf_reg }} == 'true' ]; then | |
| echo "[Performance Regression] Some model performance regression occurred, please check artifacts and reports." | |
| exit 1 | |
| fi |