Free MLOps course run by DataTalksClub. The program comprises seven modules followed by a capstone project as mentioned below and spans across several months.
-
Module 1: Introduction
- What is MLOps
- MLOps maturity model
- Running example: NY Taxi trips dataset
- Why do we need MLOps
- Course overview
- Environment preparation
- Homework
Week1: Notes, code and assignment
Source: Neptune.ai
-
Module 2: Experiment tracking and model management
- Experiment tracking intro
- Getting started with MLflow
- Experiment tracking with MLflow
- Saving and loading models with MLflow
- Model registry
- MLflow in practice
- Homework
-
Module 3: Orchestration and ML Pipelines
- Workflow orchestration
- Prefect 2.0
- Turning a notebook into a pipeline
- Deployment of Prefect flow
- Homework
-
Module 4: Model Deployment
- Batch vs online
- For online: web services vs streaming
- Serving models in Batch mode
- Web services
- Streaming (Kinesis/SQS + AWS Lambda)
- Homework
Week4 Part-1 [Web-service]: Notes, code and assignment
-
Module 5: Model Monitoring
- ML monitoring vs software monitoring
- Data quality monitoring
- Data drift / concept drift
- Batch vs real-time monitoring
- Tools: Evidently, Prometheus and Grafana
- Homework
-
Module 6: Best Practices
- Devops
- Virtual environments and Docker
- Python: logging, linting
- Testing: unit, integration, regression
- CI/CD (github actions)
- Infrastructure as code (terraform, cloudformation)
- Cookiecutter
- Makefiles
- Homework
-
Module 7: Processes
- CRISP-DM, CRISP-ML
- ML Canvas
- Data Landscape canvas
- MLOps Stack Canvas
- Documentation practices in ML projects (Model Cards Toolkit)
-
Project
- End-to-end project with all the things above
Our heartfelt gratitude to all the instructors for taking time and teaching us.
Larysa Visengeriyeva
Cristian Martinez
Kevin Kho
Theofilos Papapanagiotou
Alexey Grigorev
Emeli Dral
Sejal Vaidya
More details:
https://github.com/datatalksclub/mlops-zoomcamp