Phase 3: CI/CD: GitHub Actions for ML Pipelines

name: Train and Register Model

on:
  push:
    branches:
      - main
    paths:
      - 'src/**'
      - 'data/**'
      - '.github/workflows/train.yml'
  pull_request:
    branches:
      - main

jobs:
  train:
    runs-on: ubuntu-latest
    timeout-minutes: 30
    environment: ml-prod
    
    steps:
      - name: Checkout code
        uses: actions/checkout@v4
        with:
          fetch-depth: 0
      
      - name: Set up Python
        uses: actions/setup-python@v5
        with:
          python-version: '3.11'
          cache: 'pip'
      
      - name: Install dependencies
        run: |
          pip install --upgrade pip
          pip install -r requirements.txt
          pip install dvc dvc-s3 mlflow
      
      - name: Configure DVC remote
        env:
          AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_ID }}
          AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
        run: |
          dvc remote modify myremote url s3://my-bucket/dvc-storage
          dvc pull
      
      - name: Train model
        env:
          MLFLOW_TRACKING_URI: ${{ secrets.MLFLOW_TRACKING_URI }}
          MLFLOW_TRACKING_USERNAME: ${{ secrets.MLFLOW_USERNAME }}
          MLFLOW_TRACKING_PASSWORD: ${{ secrets.MLFLOW_PASSWORD }}
        run: |
          python src/train.py \
            --data-path data/train.csv \
            --model-output models/model.pkl \
            --experiment-name production
      
      - name: Evaluate model
        run: |
          python src/evaluate.py \
            --model-path models/model.pkl \
            --test-path data/test.csv
      
      - name: Check validation thresholds
        run: |
          ACCURACY=$(python src/get_metric.py --metric accuracy)
          if (( $(echo "$ACCURACY < 0.85" | bc -l) )); then
            echo "Accuracy $ACCURACY below threshold 0.85"
            exit 1
          fi
      
      - name: Register model to MLflow
        if: github.event_name == 'push' && github.ref == 'refs/heads/main'
        env:
          MLFLOW_TRACKING_URI: ${{ secrets.MLFLOW_TRACKING_URI }}
          MLFLOW_TRACKING_USERNAME: ${{ secrets.MLFLOW_USERNAME }}
          MLFLOW_TRACKING_PASSWORD: ${{ secrets.MLFLOW_PASSWORD }}
        run: |
          python src/register_model.py \
            --model-uri runs:/${{ env.RUN_ID }}/model \
            --stage Production
      
      - name: Commit DVC lock updates
        if: github.event_name == 'push' && github.ref == 'refs/heads/main'
        run: |
          git config user.name "ML Bot"
          git config user.email "ml-bot@company.com"
          git add dvc.lock
          git diff --quiet && git diff --staged --quiet || git commit -m "[ci] Update DVC lock after training"
          git push
      
      - name: Upload test results
        if: always()
        uses: actions/upload-artifact@v4
        with:
          name: test-results
          path: |
            metrics.json
            confusion_matrix.png
          retention-days: 30