4 hours of instruction
This theoretical course gives a comprehensive overview of the components that make up the emerging trend of MLOps. It highlights the importance of Continuous Integration (CI) and Continuous Delivery (CD) in MLOps space and enables teams to adopt automation in model building, testing, and deployment.
OBJECTIVES
- Explain the role of continuous integration and continuous deployment pipelines in machine learning
- Describe the components of a typical ML pipeline
PREREQUISITES
Model Packaging, Deployment & Monitoring
SYLLABUS & TOPICS COVERED
- DataOps and its components
- ‘Define DataOps and describe its advantages’
- ‘Analyze the relationship between MLOps and DataOps’
- ‘Outline the components of the DataOps pipeline’
- Model development
- ‘Identify the components of model development’
- ‘Describe model training’
- ‘Explain model evaluation
- testing
- and packaging’
- Model deployment
- ‘Describe model artifact deployment’
- ‘Explain model serving’
- ‘Summarize model performance monitoring
- logging
- and debugging’
- Machine learning pipeline automation tools
- ‘Explain automation of an ML pipeline’
- ‘Evaluate the role of humans in the automation of an ML pipeline’
- ‘Explore platforms for MLOps’
SOFTWARE REQUIREMENTS
TBD