validate-model.yaml

# .github/workflows/test-code.yaml

# Name of the workflow
name: Validate model

# Trigger when?
on:
  push: # this will trigger the workflow/pipeline only if there is push on `main` branch
    branches:
      - main
      - dev
    paths: # the push should be specifically to the folders `src` or `scripts` to trigger this workflow, otherwise, the workflow will not be triggered
      - 'src/**'
      - 'scripts/**'
      - 'services/airflow/dags/**'
      - 'pipelines/**'
      - 'tests/**'
      - 'configs/**'
      - '.github/**'

# Allows to only read the contents of the repository
# `contents: read` permits an action to list the commits
# `contents: write` allows the action to create a release
permissions:
  contents: read

# Declare environment variables to be used in this workflow file
env:
  message: "Validating model!"

# Tasks
jobs:
  # Task name
  validate-model:
    # OS to run the task
    runs-on: ubuntu-latest # The ubuntu-latest label currently uses the Ubuntu 22.04 runner image

    defaults: # Set working directory of the job
      run:
        shell: bash # Set the default shell
        working-directory: .

    # The steps of the task/job
    steps:
      - name: Checking out our code
        uses: actions/checkout@v4

      - name: setup python
        uses: actions/setup-python@v5
        with:
          python-version: '3.11.0'
          cache: 'pip' # caching pip dependencies
      - name: install python packages
        run: |
          
          pip3.11 install -r requirements.txt 

      - name: create-json
        id: create-json
        uses: jsdaniell/create-json@1.1.2
        with:
          name: "gdrive-credentials.json"
          json: ${{ secrets.GDRIVE_CREDENTIALS_DATA }}

      - name: Configure DVC remote with service account
        run: |
          dvc remote modify storage gdrive_use_service_account true
          dvc remote modify storage --local gdrive_service_account_json_file_path gdrive-credentials.json

      - name: Pull DVC data
        run: | 
          dvc fetch
          dvc pull

      - name: Test git config
        run: |
          git status
          git pull
          git status

      - name: Fetch tags
        run: |
          git fetch --tags --force


      # - name: Git checkout
      #   run: |
      #     git checkout main

      - name: Test tags version
        run: |
          git tag

      - name: Test git checkout
        run: |
          git checkout v1.0 data/samples.dvc
          dvc pull
          dvc checkout data/samples.dvc
          git checkout v2.0 data/samples.dvc
          dvc pull
          dvc checkout data/samples.dvc
          git checkout v3.0 data/samples.dvc
          dvc pull
          dvc checkout data/samples.dvc
          git checkout v4.0 data/samples.dvc
          dvc pull
          dvc checkout data/samples.dvc
          git checkout v5.0 data/samples.dvc
          dvc pull
          dvc checkout data/samples.dvc


      - name: Test dvc checkout
        run: |   
          dvc checkout data/samples.dvc

      - name: Test checkout
        run: |   
          git checkout HEAD data/samples.dvc
          dvc checkout data/samples.dvc

      - name: Export variables
        run: | 
          echo $PWD
          export ZENML_CONFIG_PATH=$PWD/services/zenml
          export PROJECTPATH=$PWD
          export AIRFLOW_HOME=$PWD/services/airflow
          export PYTHONPATH=$PWD/src

      - name: Run ZenML server
        run: zenml down && zenml up

      - name: Generate test artifacts
        run: | 
          export ZENML_CONFIG_PATH=/home/runner/work/Sum2024MLOps/services/zenml
          export PROJECTPATH=/home/runner/work/Sum2024MLOps
          export AIRFLOW_HOME=/home/runner/work/Sum2024MLOps/services/airflow
          export PYTHONPATH=/home/runner/work/Sum2024MLOps
          python3.11 src/utils.py 1
          python3.11 pipelines/data_prepare.py
          python3.11 src/utils.py 2
          python3.11 pipelines/data_prepare.py

      - name: Run MLflow server
        run: mlflow ui &

      - name: Run experiments
        run: | 
          mlflow run . --env-manager=local

      - name: Run validation for champion model
        run: python3.11 src/validate.py

  # Another job
  print_info:
    runs-on: ubuntu-latest
    needs: validate-model
    steps:
      - name: print my password
        run: | 
          echo My password is ${{ secrets.PASSWORD }}
          echo My name is '${{ vars.NAME }}'

      - name: print message
        run: echo $message