test-slow.yml

name: Slow Tests

# DISABLED: This workflow requires a self-hosted GPU runner to be available.
# Currently disabled because no GPU-enabled runners are available.
# Can be manually triggered via workflow_dispatch if a GPU runner becomes available.

on:
  workflow_dispatch:

jobs:
  test-slow-gpu:
    name: Slow Tests on GPU
    runs-on: [self-hosted, Windows, X64, gpu]
    timeout-minutes: 60
    continue-on-error: true

    steps:
    - name: Checkout code
      uses: actions/checkout@v6
      with:
        lfs: true

    - name: Create venv in RUNNER_TEMP
      # Python 3.11 is required because the base dependency nvidia-physicsnemo
      # requires Python >= 3.11.
      run: |
        & "C:\Program Files\Python311\python.exe" -m venv "$env:RUNNER_TEMP\monai-physio-venv"
        echo "$env:RUNNER_TEMP\monai-physio-venv\Scripts" >> $env:GITHUB_PATH

    - name: Check GPU availability
      run: nvidia-smi

    - name: Cache uv packages
      uses: actions/cache@v4
      with:
        path: ~\AppData\Local\uv\cache
        key: ${{ runner.os }}-uv-${{ hashFiles('pyproject.toml') }}
        restore-keys: |
          ${{ runner.os }}-uv-

    - name: Cache test data
      uses: actions/cache@v4
      with:
        path: |
          tests/data/
          tests/results/
        key: test-data-${{ hashFiles('tests/test_*.py') }}-v2
        restore-keys: |
          test-data-

    - name: Upgrade pip and build tools
      run: |
        python -m pip install --upgrade pip setuptools wheel uv

    - name: Install package with test dependencies
      # uv respects [tool.uv.sources], routing torch to pytorch-cu130 for cuda13.
      # PhysicsNeMo is a base dependency, so it is always installed; it is
      # what --run-physicsnemo (Tutorial 9 and any tests marked
      # requires_physicsnemo) needs.
      # Invoke via python -m uv so uv targets the active venv interpreter.
      run: |
        python -m uv pip install -e ".[dev_cuda13]" --no-build-isolation-package torch-scatter

    - name: Assert CUDA is accessible
      run: |
        python -c "
        import sys, torch
        print(f'PyTorch {torch.__version__}  |  CUDA toolkit {torch.version.cuda}')
        if not torch.cuda.is_available():
            print('ERROR: torch.cuda.is_available() returned False', file=sys.stderr)
            sys.exit(1)
        n = torch.cuda.device_count()
        if n == 0:
            print('ERROR: torch.cuda.device_count() == 0', file=sys.stderr)
            sys.exit(1)
        print(f'OK: {n} GPU(s) visible')
        "

    - name: Run slow tests
      # Self-hosted GPU runner: enable every opt-in bucket via --run-all.
      # Tests whose host requirements (e.g. a licensed Simpleware install)
      # aren't met on the runner will runtime-skip cleanly via their
      # internal availability guards.
      run: |
        pytest tests/ -v --run-all --cov=monai_physio --cov-report=xml --cov-report=term
      env:
        CUDA_VISIBLE_DEVICES: 0

    - name: Upload coverage to Codecov
      uses: codecov/codecov-action@v4
      with:
        token: ${{ secrets.CODECOV_TOKEN }}
        file: ./coverage.xml
        flags: slow-tests-gpu
        name: codecov-slow-gpu
        fail_ci_if_error: false