check_example_ci

.github/workflows/self-scheduled.yml

@@ -7,54 +7,27 @@ name: Self-hosted runner (scheduled)
 # `docker/transformers-pytorch-deepspeed-latest-gpu/Dockerfile`
 
 on:
-  repository_dispatch:
-  schedule:
-    - cron: "0 2 * * *"
+  push:
+    branches:
+      - check_example_ci
 
 env:
-  HF_HOME: /mnt/cache
+  HF_HOME: /transformers/mycache
+#  HF_HOME: /mnt/cache
   TRANSFORMERS_IS_CI: yes
-  OMP_NUM_THREADS: 8
-  MKL_NUM_THREADS: 8
+  #OMP_NUM_THREADS: 8
+  #MKL_NUM_THREADS: 8
   RUN_SLOW: yes
-  SIGOPT_API_TOKEN: ${{ secrets.SIGOPT_API_TOKEN }}
-  TF_FORCE_GPU_ALLOW_GROWTH: true
-  RUN_PT_TF_CROSS_TESTS: 1
+  #SIGOPT_API_TOKEN: ${{ secrets.SIGOPT_API_TOKEN }}
+  #TF_FORCE_GPU_ALLOW_GROWTH: true
+  #RUN_PT_TF_CROSS_TESTS: 1
 
 jobs:
-  check_runner_status:
-    name: Check Runner Status
-    runs-on: ubuntu-latest
-    steps:
-      - name: Checkout transformers
-        uses: actions/checkout@v3
-        with:
-          fetch-depth: 2
-
-      - name: Check Runner Status
-        run: python utils/check_self_hosted_runner.py --target_runners single-gpu-scheduled-ci-runner-docker,multi-gpu-scheduled-ci-runner-docker --token ${{ secrets.ACCESS_REPO_INFO_TOKEN }}
-
-  check_runners:
-    name: Check Runners
-    needs: check_runner_status
-    strategy:
-      matrix:
-        machine_type: [single-gpu, multi-gpu]
-    runs-on: ${{ format('{0}-{1}', matrix.machine_type, 'docker') }}
-    container:
-      image: huggingface/transformers-all-latest-gpu
-      options: --gpus 0 --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
-    steps:
-      - name: NVIDIA-SMI
-        run: |
-          nvidia-smi
-
   setup:
     name: Setup
-    needs: check_runners
     strategy:
       matrix:
-        machine_type: [single-gpu, multi-gpu]
+        machine_type: [single-gpu]
     runs-on: ${{ format('{0}-{1}', matrix.machine_type, 'docker') }}
     container:
       image: huggingface/transformers-all-latest-gpu
@@ -88,120 +61,6 @@ jobs:
         run: |
           nvidia-smi
 
-  run_tests_single_gpu:
-    name: Model tests
-    strategy:
-      fail-fast: false
-      matrix:
-        folders: ${{ fromJson(needs.setup.outputs.matrix) }}
-        machine_type: [single-gpu]
-    runs-on: ${{ format('{0}-{1}', matrix.machine_type, 'docker') }}
-    container:
-      image: huggingface/transformers-all-latest-gpu
-      options: --gpus 0 --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
-    needs: setup
-    steps:
-      - 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 ${{ github.sha }}
-
-      - name: NVIDIA-SMI
-        run: |
-          nvidia-smi
-
-      - name: Environment
-        working-directory: /transformers
-        run: |
-          python3 utils/print_env.py
-
-      - name: Show installed libraries and their versions
-        working-directory: /transformers
-        run: pip freeze
-
-      - name: Run all tests on GPU
-        working-directory: /transformers
-        run: python3 -m pytest -v --make-reports=${{ matrix.machine_type }}_tests_gpu_${{ matrix.folders }} tests/${{ matrix.folders }}
-
-      - name: Failure short reports
-        if: ${{ failure() }}
-        continue-on-error: true
-        run: cat /transformers/reports/${{ matrix.machine_type }}_tests_gpu_${{ matrix.folders }}/failures_short.txt
-
-      - name: Test suite reports artifacts
-        if: ${{ always() }}
-        uses: actions/upload-artifact@v3
-        with:
-          name: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports
-          path: /transformers/reports/${{ matrix.machine_type }}_tests_gpu_${{ matrix.folders }}
-
-  run_tests_multi_gpu:
-    name: Model tests
-    strategy:
-      fail-fast: false
-      matrix:
-        folders: ${{ fromJson(needs.setup.outputs.matrix) }}
-        machine_type: [multi-gpu]
-    runs-on: ${{ format('{0}-{1}', matrix.machine_type, 'docker') }}
-    container:
-      image: huggingface/transformers-all-latest-gpu
-      options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
-    needs: setup
-    steps:
-      - 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 ${{ github.sha }}
-
-      - name: NVIDIA-SMI
-        run: |
-          nvidia-smi
-
-      - name: Environment
-        working-directory: /transformers
-        run: |
-          python3 utils/print_env.py
-
-      - name: Show installed libraries and their versions
-        working-directory: /transformers
-        run: pip freeze
-
-      - name: Run all tests on GPU
-        working-directory: /transformers
-        run: python3 -m pytest -v --make-reports=${{ matrix.machine_type }}_tests_gpu_${{ matrix.folders }} tests/${{ matrix.folders }}
-
-      - name: Failure short reports
-        if: ${{ failure() }}
-        continue-on-error: true
-        run: cat /transformers/reports/${{ matrix.machine_type }}_tests_gpu_${{ matrix.folders }}/failures_short.txt
-
-      - name: Test suite reports artifacts
-        if: ${{ always() }}
-        uses: actions/upload-artifact@v3
-        with:
-          name: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports
-          path: /transformers/reports/${{ matrix.machine_type }}_tests_gpu_${{ matrix.folders }}
-
   run_examples_gpu:
     name: Examples directory
     strategy:
@@ -212,6 +71,7 @@ jobs:
     container:
       image: huggingface/transformers-all-latest-gpu
       options: --gpus 0 --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
+      # options: --gpus 0 --shm-size "16gb" --ipc host
     needs: setup
     steps:
       - name: Update clone
@@ -233,262 +93,10 @@ jobs:
 
       - name: Run examples tests on GPU
         working-directory: /transformers
+#          pip install -r examples/pytorch/_tests_requirements.txt
         run: |
-          pip install -r examples/pytorch/_tests_requirements.txt
-          python3 -m pytest -v --make-reports=${{ matrix.machine_type }}_examples_gpu examples/pytorch
-
-      - name: Failure short reports
-        if: ${{ failure() }}
-        continue-on-error: true
-        run: cat /transformers/reports/${{ matrix.machine_type }}_examples_gpu/failures_short.txt
-
-      - name: Test suite reports artifacts
-        if: ${{ always() }}
-        uses: actions/upload-artifact@v3
-        with:
-          name: ${{ matrix.machine_type }}_run_examples_gpu
-          path: /transformers/reports/${{ matrix.machine_type }}_examples_gpu
-
-  run_pipelines_torch_gpu:
-    name: PyTorch pipelines
-    strategy:
-      fail-fast: false
-      matrix:
-        machine_type: [single-gpu, multi-gpu]
-    runs-on: ${{ format('{0}-{1}', matrix.machine_type, 'docker') }}
-    container:
-      image: huggingface/transformers-pytorch-gpu
-      options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
-    needs: setup
-    steps:
-      - name: Update clone
-        working-directory: /transformers
-        run: git fetch && git checkout ${{ github.sha }}
-
-      - name: NVIDIA-SMI
-        run: |
-          nvidia-smi
-
-      - name: Environment
-        working-directory: /transformers
-        run: |
-          python3 utils/print_env.py
-
-      - name: Show installed libraries and their versions
-        working-directory: /transformers
-        run: pip freeze
-
-      - name: Run all pipeline tests on GPU
-        working-directory: /transformers
-        run: |
-          python3 -m pytest -n 1 -v --dist=loadfile --make-reports=${{ matrix.machine_type }}_tests_torch_pipeline_gpu tests/pipelines
-
-      - name: Failure short reports
-        if: ${{ failure() }}
-        continue-on-error: true
-        run: cat /transformers/reports/${{ matrix.machine_type }}_tests_torch_pipeline_gpu/failures_short.txt
-
-      - name: Test suite reports artifacts
-        if: ${{ always() }}
-        uses: actions/upload-artifact@v3
-        with:
-          name: ${{ matrix.machine_type }}_run_tests_torch_pipeline_gpu
-          path: /transformers/reports/${{ matrix.machine_type }}_tests_torch_pipeline_gpu
-
-  run_pipelines_tf_gpu:
-    name: TensorFlow pipelines
-    strategy:
-      fail-fast: false
-      matrix:
-        machine_type: [single-gpu, multi-gpu]
-    runs-on: ${{ format('{0}-{1}', matrix.machine_type, 'docker') }}
-    container:
-      image: huggingface/transformers-tensorflow-gpu
-      options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
-    needs: setup
-    steps:
-      - name: Update clone
-        working-directory: /transformers
-        run: |
-          git fetch && git checkout ${{ github.sha }}
-
-      - name: NVIDIA-SMI
-        run: |
-          nvidia-smi
-
-      - name: Environment
-        working-directory: /transformers
-        run: |
-          python3 utils/print_env.py
-
-      - name: Show installed libraries and their versions
-        working-directory: /transformers
-        run: pip freeze
-
-      - name: Run all pipeline tests on GPU
-        working-directory: /transformers
-        run: |
-          python3 -m pytest -n 1 -v --dist=loadfile --make-reports=${{ matrix.machine_type }}_tests_tf_pipeline_gpu tests/pipelines
-
-      - name: Failure short reports
-        if: ${{ always() }}
-        run: |
-          cat /transformers/reports/${{ matrix.machine_type }}_tests_tf_pipeline_gpu/failures_short.txt
-
-      - name: Test suite reports artifacts
-        if: ${{ always() }}
-        uses: actions/upload-artifact@v3
-        with:
-          name: ${{ matrix.machine_type }}_run_tests_tf_pipeline_gpu
-          path: /transformers/reports/${{ matrix.machine_type }}_tests_tf_pipeline_gpu
-
-  run_all_tests_torch_cuda_extensions_gpu:
-    name: Torch CUDA extension tests
-    strategy:
-      fail-fast: false
-      matrix:
-        machine_type: [single-gpu, multi-gpu]
-    runs-on: ${{ format('{0}-{1}', matrix.machine_type, 'docker') }}
-    needs: setup
-    container:
-      image: huggingface/transformers-pytorch-deepspeed-latest-gpu
-      options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
-    steps:
-      - name: Update clone
-        working-directory: /workspace/transformers
-        run: git fetch && git checkout ${{ github.sha }}
-
-      - name: Remove cached torch extensions
-        run: rm -rf /github/home/.cache/torch_extensions/
-
-      # To avoid unknown test failures
-      - name: Pre build DeepSpeed *again*
-        working-directory: /workspace
-        run: |
-          python3 -m pip uninstall -y deepspeed
-          DS_BUILD_CPU_ADAM=1 DS_BUILD_FUSED_ADAM=1 DS_BUILD_AIO=1 DS_BUILD_UTILS=1 python3 -m pip install deepspeed --global-option="build_ext" --global-option="-j8" --no-cache -v --disable-pip-version-check
-
-      - name: NVIDIA-SMI
-        run: |
-          nvidia-smi
-
-      - name: Environment
-        working-directory: /workspace/transformers
-        run: |
-          python utils/print_env.py
-
-      - name: Show installed libraries and their versions
-        working-directory: /workspace/transformers
-        run: pip freeze
-
-      - name: Run all tests on GPU
-        working-directory: /workspace/transformers
-        run: |
-          python -m pytest -v --make-reports=${{ matrix.machine_type }}_tests_torch_cuda_extensions_gpu tests/deepspeed tests/extended
-
-      - name: Failure short reports
-        if: ${{ failure() }}
-        continue-on-error: true
-        run: cat /workspace/transformers/reports/${{ matrix.machine_type }}_tests_torch_cuda_extensions_gpu/failures_short.txt
-
-      - name: Test suite reports artifacts
-        if: ${{ always() }}
-        uses: actions/upload-artifact@v3
-        with:
-          name: ${{ matrix.machine_type }}_run_tests_torch_cuda_extensions_gpu_test_reports
-          path: /workspace/transformers/reports/${{ matrix.machine_type }}_tests_torch_cuda_extensions_gpu
-
-  run_extract_warnings:
-    name: Extract warnings in CI artifacts
-    runs-on: ubuntu-latest
-    if: always()
-    needs: [
-      check_runner_status,
-      check_runners,
-      setup,
-      run_tests_single_gpu,
-      run_tests_multi_gpu,
-      run_examples_gpu,
-      run_pipelines_tf_gpu,
-      run_pipelines_torch_gpu,
-      run_all_tests_torch_cuda_extensions_gpu
-    ]
-    steps:
-      - name: Checkout transformers
-        uses: actions/checkout@v3
-        with:
-          fetch-depth: 2
-
-      - name: Install transformers
-        run: pip install transformers
-
-      - name: Show installed libraries and their versions
-        run: pip freeze
-
-      - name: Create output directory
-        run: mkdir warnings_in_ci
-
-      - uses: actions/download-artifact@v3
-        with:
-          path: warnings_in_ci
-
-      - name: Show artifacts
-        run: echo "$(python3 -c 'import os; d = os.listdir(); print(d)')"
-        working-directory: warnings_in_ci
-
-      - name: Extract warnings in CI artifacts
-        run: |
-          python3 utils/extract_warnings.py --workflow_run_id ${{ github.run_id }} --output_dir warnings_in_ci --token ${{ secrets.ACCESS_REPO_INFO_TOKEN }} --from_gh
-          echo "$(python3 -c 'import os; import json; fp = open("warnings_in_ci/selected_warnings.json"); d = json.load(fp); d = "\n".join(d) ;print(d)')"
-
-      - name: Upload artifact
-        if: ${{ always() }}
-        uses: actions/upload-artifact@v3
-        with:
-          name: warnings_in_ci
-          path: warnings_in_ci/selected_warnings.json
-
-  send_results:
-    name: Send results to webhook
-    runs-on: ubuntu-latest
-    if: always()
-    needs: [
-      check_runner_status,
-      check_runners,
-      setup,
-      run_tests_single_gpu,
-      run_tests_multi_gpu,
-      run_examples_gpu,
-      run_pipelines_tf_gpu,
-      run_pipelines_torch_gpu,
-      run_all_tests_torch_cuda_extensions_gpu,
-      run_extract_warnings
-    ]
-    steps:
-      - name: Preliminary job status
-        shell: bash
-        # For the meaning of these environment variables, see the job `Setup`
-        run: |
-          echo "Runner availability: ${{ needs.check_runner_status.result }}"
-          echo "Runner status: ${{ needs.check_runners.result }}"
-          echo "Setup status: ${{ needs.setup.result }}"
-
-      - uses: actions/checkout@v3
-      - uses: actions/download-artifact@v3
-      - name: Send message to Slack
-        env:
-          CI_SLACK_BOT_TOKEN: ${{ secrets.CI_SLACK_BOT_TOKEN }}
-          CI_SLACK_CHANNEL_ID: ${{ secrets.CI_SLACK_CHANNEL_ID }}
-          CI_SLACK_CHANNEL_ID_DAILY: ${{ secrets.CI_SLACK_CHANNEL_ID_DAILY }}
-          CI_SLACK_CHANNEL_DUMMY_TESTS: ${{ secrets.CI_SLACK_CHANNEL_DUMMY_TESTS }}
-          CI_SLACK_REPORT_CHANNEL_ID: ${{ secrets.CI_SLACK_CHANNEL_ID_DAILY }}
-          CI_EVENT: scheduled
-          RUNNER_STATUS: ${{ needs.check_runner_status.result }}
-          RUNNER_ENV_STATUS: ${{ needs.check_runners.result }}
-          SETUP_STATUS: ${{ needs.setup.result }}
-        # We pass `needs.setup.outputs.matrix` as the argument. A processing in `notification_service.py` to change
-        # `models/bert` to `models_bert` is required, as the artifact names use `_` instead of `/`.
-        run: |
-          pip install slack_sdk
-          pip show slack_sdk
-          python utils/notification_service.py "${{ needs.setup.outputs.matrix }}"
+          python3 -m pytest -v examples/pytorch/test_accelerate_examples.py::ExamplesTestsNoTrainer::test_run_image_classification_no_trainer
+          ls -l mycache/datasets
+          ls -l mycache/datasets/downloads
+          ls -l mycache/datasets/hf-internal-testing___cats_vs_dogs_sample
+          ls -l mycache/datasets/downloads/extracted