.github/workflows/train_model.yml

name: Train Final Model

on:
  workflow_dispatch:  # Manual trigger only
    inputs:
      model_type:
        description: 'Model to train'
        required: true
        type: choice
        options:
          - Random Forest
          - Gradient Boosting
          - Logistic Regression
        default: 'Random Forest'
      cv_folds:
        description: 'Number of cross-validation folds'
        required: false
        type: number
        default: 10
      n_iter:
        description: 'Number of RandomizedSearchCV iterations'
        required: false
        type: number
        default: 100
      registry_name:
        description: 'Model name in Hopsworks registry'
        required: false
        type: string
        default: 'phishing_detector'

jobs:
  train-model:
    name: Train and Upload Model
    runs-on: ubuntu-latest
    timeout-minutes: 120  # 2 hours max for extensive hyperparameter search

    steps:
      - name: Checkout code
        uses: actions/checkout@v4
        with:
          lfs: true  # Pull Git LFS files if needed

      - name: Set up Python
        uses: actions/setup-python@v5
        with:
          python-version: '3.11'

      - name: Install uv
        uses: astral-sh/setup-uv@v4
        with:
          version: "latest"

      - name: Install dependencies
        run: uv sync --group dev

      - name: Train model with extensive hyperparameter search
        env:
          HOPSWORKS_API_KEY: ${{ secrets.HOPSWORKS_API_KEY }}
          HOPSWORKS_PROJECT: ${{ secrets.HOPSWORKS_PROJECT_NAME }}
        run: |
          uv run python src/phising_detection/models/train_final_model.py \
            --model "${{ github.event.inputs.model_type }}" \
            --cv-folds ${{ github.event.inputs.cv_folds }} \
            --n-iter ${{ github.event.inputs.n_iter }} \
            --registry-name "${{ github.event.inputs.registry_name }}" \
            --feature-group urlscan_features \
            --feature-group-version 1

      - name: Training Summary
        if: success()
        run: |
          echo "✅ Model training completed successfully!"
          echo "Model: ${{ github.event.inputs.model_type }}"
          echo "CV Folds: ${{ github.event.inputs.cv_folds }}"
          echo "Iterations: ${{ github.event.inputs.n_iter }}"
          echo "Registry Name: ${{ github.event.inputs.registry_name }}"
          echo "Model uploaded to Hopsworks Model Registry"

      - name: Notify on failure
        if: failure()
        run: |
          echo "❌ Model training failed!"
          echo "Check the logs above for details"