model_jobs.yml

name: model jobs

on:
  workflow_call:
    inputs:
      folder_slices:
        required: true
        type: string
      machine_type:
        required: true
        type: string
      slice_id:
        required: true
        type: number

env:
  HF_HOME: /mnt/cache
  TRANSFORMERS_IS_CI: yes
  OMP_NUM_THREADS: 8
  MKL_NUM_THREADS: 8
  RUN_SLOW: yes
  # For gated repositories, we still need to agree to share information on the Hub repo. page in order to get access.
  # This token is created under the bot `hf-transformers-bot`.
  HF_HUB_READ_TOKEN: ${{ secrets.HF_HUB_READ_TOKEN }}
  SIGOPT_API_TOKEN: ${{ secrets.SIGOPT_API_TOKEN }}
  TF_FORCE_GPU_ALLOW_GROWTH: true
  RUN_PT_TF_CROSS_TESTS: 1
  CUDA_VISIBLE_DEVICES: 0,1

jobs:
  model_job:
    name: " "
    strategy:
      fail-fast: false
      matrix:
        folders: ${{ fromJson(inputs.folder_slices)[inputs.slice_id] }}
    runs-on: ['${{ inputs.machine_type }}', nvidia-gpu, t4, daily-ci]
    container:
      image: huggingface/transformers-all-latest-gpu
      options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
    steps:
      - name: Echo input and matrix info
        shell: bash
        run: |
          echo "${{ inputs.folder_slices }}"
          echo "${{ matrix.folders }}"
          echo "${{ toJson(fromJson(inputs.folder_slices)[inputs.slice_id]) }}"

      - name: Echo folder ${{ matrix.folders }}
        shell: bash
        # For folders like `models/bert`, set an env. var. (`matrix_folders`) to `models_bert`, which will be used to
        # set the artifact folder names (because the character `/` is not allowed).
        run: |
          echo "${{ matrix.folders }}"
          matrix_folders=${{ matrix.folders }}
          matrix_folders=${matrix_folders/'models/'/'models_'}
          echo "$matrix_folders"
          echo "matrix_folders=$matrix_folders" >> $GITHUB_ENV

      - name: Update clone
        working-directory: /transformers
        run: git fetch && git checkout ${{ github.sha }}

      - name: Reinstall transformers in edit mode (remove the one installed during docker image build)
        working-directory: /transformers
        run: python3 -m pip uninstall -y transformers && python3 -m pip install -e .

      - name: NVIDIA-SMI
        run: |
          nvidia-smi

      - name: Environment
        working-directory: /transformers
        run: |
          python3 utils/print_env.py

      - name: Show installed libraries and their versions
        working-directory: /transformers
        run: pip freeze

      - name: Run all tests on GPU
        working-directory: /transformers
        run: python3 -m pytest -v --make-reports=${{ inputs.machine_type }}_tests_gpu_${{ matrix.folders }} tests/${{ matrix.folders }}

      - name: Failure short reports
        if: ${{ failure() }}
        continue-on-error: true
        run: cat /transformers/reports/${{ inputs.machine_type }}_tests_gpu_${{ matrix.folders }}/failures_short.txt

      - name: Run test
        shell: bash
        run: |
          mkdir -p /transformers/reports/${{ inputs.machine_type }}_tests_gpu_${{ matrix.folders }}
          echo "hello" > /transformers/reports/${{ inputs.machine_type }}_tests_gpu_${{ matrix.folders }}/hello.txt
          echo "${{ inputs.machine_type }}_tests_gpu_${{ matrix.folders }}"

      - name: "Test suite reports artifacts: ${{ inputs.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports"
        if: ${{ always() }}
        uses: actions/upload-artifact@v3
        with:
          name: ${{ inputs.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports
          path: /transformers/reports/${{ inputs.machine_type }}_tests_gpu_${{ matrix.folders }}