train_deploy.yaml

name: Train and Deploy ML Model

on:
  push:
    branches:
      - main
      - citest

jobs:
  build-deploy:
    runs-on: ubuntu-latest

    steps:
      # Step 1: Checkout repository
      - name: Checkout repository
        uses: actions/checkout@v2

      # Step 2: Set up Google Cloud SDK
      - name: Set up Cloud SDK
        uses: google-github-actions/setup-gcloud@v0.2.0
        with:
          project_id: ${{ secrets.GCP_PROJECT_ID }}
          service_account_key: ${{ secrets.GCP_SERVICE_ACCOUNT_KEY }}
          export_default_credentials: true

      # Step 3: Build Docker Image
      - name: Build Docker Image
        run: |
          IMAGE_NAME="gcr.io/${{ secrets.GCP_PROJECT_ID }}/mpg"
          MODEL_VERSION="v2"
          echo "Building Docker image..."
          # docker build ./ -t ${IMAGE_NAME}:${MODEL_VERSION} -f GCP/gcpdeploy/trainer/Dockerfile
          # echo "Pushing Docker image..."
          # docker push ${IMAGE_NAME}:${MODEL_VERSION}
          echo "Tagging image as 'latest'..."
          # docker tag ${IMAGE_NAME}:${MODEL_VERSION} ${IMAGE_NAME}:latest
          # docker push ${IMAGE_NAME}:latest

      # Step 4: Retrieve Image Digest
      - name: Retrieve Image Digest
        id: digest
        run: |
          IMAGE_NAME="gcr.io/${{ secrets.GCP_PROJECT_ID }}/mpg"
          MODEL_VERSION="v2"
          DIGEST=$(gcloud container images describe ${IMAGE_NAME}:${MODEL_VERSION} --format="value(image_summary.fully_qualified_digest)")
          if [ -z "$DIGEST" ]; then
            echo "Error: Failed to retrieve image digest. Ensure the image was pushed successfully."
            exit 1
          fi
          echo "Digest: $DIGEST"
          echo "::set-output name=digest::$DIGEST"
          
      # Step 5: Upload Model to Vertex AI
      - name: Upload or Retrieve Model
        id: model-upload
        run: |
          IMAGE_NAME="gcr.io/${{ secrets.GCP_PROJECT_ID }}/mpg"
          MODEL_VERSION="v2"
          DIGEST=${{ steps.digest.outputs.digest }}

          # Try uploading the model
          MODEL_ID=$(gcloud ai models upload \
              --region=us-east1 \
              --display-name=model-${MODEL_VERSION} \
              --artifact-uri=gs://stock_price_prediction_dataset/model_checkpoints/ \
              --container-image-uri=${DIGEST} \
              --format="value(name)")

          # If upload fails, try retrieving the existing model
          if [ -z "$MODEL_ID" ]; then
            echo "Upload failed, trying to retrieve existing model..."
            MODEL_ID=$(gcloud ai models list \
                --region=us-east1 \
                --filter="display_name:model-${MODEL_VERSION}" \
                --format="value(name)" | head -n 1)
          fi

          if [ -z "$MODEL_ID" ]; then
            echo "Error: Failed to upload or retrieve the model."
            exit 1
          fi

          echo "Model ID: $MODEL_ID"
          echo "::set-output name=model_id::$MODEL_ID"

      # Step 6: Create or Retrieve Endpoint
      - name: Create or Retrieve Endpoint
        id: endpoint
        run: |
          ENDPOINT_NAME="model-endpoint-v2"
          ENDPOINT_ID=$(gcloud ai endpoints list \
              --region=us-east1 \
              --filter="display_name:${ENDPOINT_NAME}" \
              --format="value(name)")
          if [ -z "$ENDPOINT_ID" ]; then
            echo "Creating new endpoint: ${ENDPOINT_NAME}"
            ENDPOINT_ID=$(gcloud ai endpoints create \
                --region=us-east1 \
                --display-name=${ENDPOINT_NAME} \
                --format="value(name)")
            if [ -z "$ENDPOINT_ID" ]; then
              echo "Error: Failed to create endpoint."
              exit 1
            fi
          else
            echo "Using existing endpoint: ${ENDPOINT_NAME}"
          fi
          echo "Endpoint ID: $ENDPOINT_ID"
          echo "::set-output name=endpoint_id::$ENDPOINT_ID"

      # Step 7: Deploy Model to Endpoint
      - name: Deploy Model
        run: |
          ENDPOINT_ID=${{ steps.endpoint.outputs.endpoint_id }}
          MODEL_ID=${{ steps.model-upload.outputs.model_id }}
          # gcloud ai endpoints deploy-model \
          #     $ENDPOINT_ID \
          #     --region=us-east1 \
          #     --model=$MODEL_ID \
          #     --display-name=model-v2 \
          #     --machine-type=n1-highmem-4 \
          #     --traffic-split="0=100"
          # if [ $? -ne 0 ]; then
          #   echo "Error: Model deployment failed."
          #   exit 1
          # fi
          echo "Model deployed successfully."

      # Step 8: Notify Completion
      - name: Notify Completion
        run: echo "Model deployment process completed successfully."