dvc-pipeline.yml

name: HparamSearch, Train Test, Infer and Report
on:
  workflow_call:
  workflow_dispatch:
  pull_request:
    branches: [ main ]

jobs:
  hparamSearch-train-test-infer-and-report:
    runs-on: ubuntu-latest
    permissions:
      contents: write
      pull-requests: write
      repository-projects: write
      id-token: write
    steps:
      - uses: actions/checkout@v4
      - uses: iterative/setup-cml@v2
      
      - name: Install uv
        uses: astral-sh/setup-uv@v2

      - name: Set up Python 3.11.7
        run: uv python install 3.11.7

      - name: Install dependencies
        env:
          UV_EXTRA_INDEX_URL: https://download.pytorch.org/whl/cpu
        run: |
          uv sync

      - name: Create credentials.json
        env:
          GDRIVE_CREDENTIALS_DATA: ${{ secrets.GDRIVE_CREDENTIALS_DATA }}
        run: |
          echo $GDRIVE_CREDENTIALS_DATA > credentials_1.json
          cat credentials_1.json  # (Optional) Print the content to ensure it's correct

      - name: Modify DVC Remote
        run: |
          uv run dvc remote modify myremote --local gdrive_service_account_json_file_path credentials_1.json

      - name: DVC Pull Data
        run: |
          uv run dvc pull -v

      - name: Train model
        run: |
          rm report.md
          uv run dvc repro -f

      - name: Create CML report
        env:
          REPO_TOKEN: ${{ secrets.GITHUB_TOKEN }}
        run: |
          # Comment hparams, acc, loss plots and table plots
          cml comment create report.md