• 01-intro

  • setup environment
  • download data
  • added training notebook
  • done homework

• 02-experiment-tracking

  • setup mlflow
  • do experiment tracking
  • Model tracking
  • Model Registry
  • Launch tracking server:- mlflow server --backend-store-uri sqlite:///mlflow.db --default-artifact-root artifacts

• 03-orchestration

  • setup mage: git clone https://github.com/mage-ai/mlops.git
  • run mage server: cd 03-orchestration/mlops && scripts/start.sh
  • unit_1_data_preparation
    • make pipeline
    • data ingestion
    • data transformation
    • data visualization
    • data exporters
  • unit_2_training
    • sklearn models: hyperparameter tuning
    • sklearn models: training models with best hyperparameters
    • XGBoost model: hyperparameter tuning
    • XGBoost model: training model with best hyperparameters
  • unit_3_observability
    • didn't did all the charts and deployment

• 04-deployment

• 05-monitoring

  • Environment setup(conda activate py11)
  • Prepare reference data and model
  • Evidently metrics calculation
  • Evidently dashboard
  • Grafana dashboard with dummy data
  • Grafana Dashboard with real data
  • Saved dashboard to config json file
  • Evidently ReportPreset and TestSuitePreset
  • Done Homework

• 06-best-practices

  • Unit Testing
  • Integration Testing
  • Local Stack(Kinesis test)
  • pylint, black, isort
  • pre-commit hooks
  • Makefile
  • Homework
  • terraform (IAC Infrastructure as Code)
    • terraform setup
    • terraform basic

• Tool to download github repo sub directory: https://download-directory.github.io/
• Module 3 sorted video playlist: https://www.youtube.com/playlist?list=PLJlrBE4yPIzg9W9LaAp-3DOtZ9JnDd23J
• Docker remove all cache: docker system prune -a
• conda deactivate

https://certificate.datatalks.club/mlops-zoomcamp/2024/286c615358c35738f2f81a0421901bd9d92992ab.pdf