nightly.yml

name: Self-hosted runner with slow tests (scheduled)

on:
  workflow_dispatch:
  schedule:
    - cron: "0 2 * * *"

env:
  RUN_SLOW: "yes"
  IS_GITHUB_CI: "1"
  SLACK_API_TOKEN: ${{ secrets.SLACK_API_TOKEN }}


jobs:
  run_all_tests_single_gpu:
    runs-on: [self-hosted, docker-gpu, multi-gpu]
    env:
      CUDA_VISIBLE_DEVICES: "0"
      TEST_TYPE: "single_gpu"
    container:
      image: huggingface/accelerate-gpu:latest
      options: --gpus all --shm-size "16gb"
    defaults:
      run:
        working-directory: accelerate/
        shell: bash
    steps:
      - name: Update clone & pip install
        run: |
          source activate accelerate
          git config --global --add safe.directory '*'
          git fetch && git checkout ${{ github.sha }} 
          pip install -e . --no-deps
          pip install pytest-reportlog tabulate

      - name: Run test on GPUs
        run: |
          source activate accelerate
          make test
          
      - name: Run examples on GPUs
        if: always()
        run: |
          source activate accelerate
          pip uninstall comet_ml -y
          make test_examples
          
      - name: Generate Report
        if: always()
        run: |
          pip install slack_sdk tabulate
          python utils/log_reports.py >> $GITHUB_STEP_SUMMARY

  run_all_tests_multi_gpu:
    runs-on: [self-hosted, docker-gpu, multi-gpu]
    env:
      CUDA_VISIBLE_DEVICES: "0,1"
      TEST_TYPE: "multi_gpu"
    container:
      image: huggingface/accelerate-gpu:latest
      options: --gpus all --shm-size "16gb"
    defaults:
      run:
        working-directory: accelerate/
        shell: bash
    steps:
      - name: Update clone
        run: |
          source activate accelerate
          git config --global --add safe.directory '*'
          git fetch && git checkout ${{ github.sha }}
          pip install -e . --no-deps
          pip install pytest-reportlog tabulate

      - name: Run core and big modeling tests on GPUs
        run: |
          source activate accelerate
          make test_core
          make test_big_modeling
          make test_cli

      - name: Run Integration tests on GPUs
        if: always()
        run: |
          source activate accelerate
          make test_integrations

      - name: Run examples on GPUs
        if: always()
        run: |
          source activate accelerate
          pip uninstall comet_ml -y
          make test_examples

      - name: Generate Report
        if: always()
        run: |
          pip install slack_sdk tabulate
          python utils/log_reports.py >> $GITHUB_STEP_SUMMARY

  
  run-integration-tests:
    needs: [run_all_tests_single_gpu, run_all_tests_multi_gpu]
    if: always()
    uses: ./.github/workflows/self_hosted_integration_tests.yml