Teach practical aspects of productionizing ML services — from collecting requirements to model deployment and monitoring.
Data scientists and ML engineers. Also software engineers and data engineers interested in learning about putting ML in production.
- Python
- Docker
- Being comfortable with command line
- Prior exposure to machine learning
- Prior programming experience (at least 1+ year)
Course start: January 8th Course end: January 24th Final project presentation: January 31st
- What is MLOps
- Why do we need MLOps
- Running example: NY Taxi trips dataset
- Course overview
- Environment preparation
- Practice
- Homework
- Introduction to workflows orchestration
- Introduction to Prefect
- From notebooks to workflows
- Continuous Training
- Next steps and resources
- Practice
- Homework
- Testing Data Quality
- Creating Data Pipelines
- Feature Stores
- Data Leakage
- Training / Serving Skew
- Practice
- Homework
- Experiment tracking intro
- What is MLflow
- Experiment tracking with MLflow
- Saving and loading models with MLflow
- Model registry
- Practice
- Homework
- Web service: model deployment with FastAPI
- Docker: containerizing a web service
- MLflow: retrieving a model from the model registry
- Locust: load testing a web service
- End-to-end project with all the things above
- Keener Jeremy
- Rohart Capucine