Model Packaging, Deployment & Monitoring

4 hours of instruction

This practical, hands-on course dives into the details of implementation of model deployment – the essential part of the ML cycle in production.

OBJECTIVES

1. By the end of this course, participants will be able to package and deploy models using the infrastructure they have set up using the chosen cloud provider services

PREREQUISITES

Model Development & Testing

SYLLABUS & TOPICS COVERED

1. Introducing CodePipeline in AWS
   • Explain CodePipeline and how it works
   • Describe the setup and configuration options for approval actions in CodePipeline
   • Approve or reject an action in CodePipeline
2. Model optimization with compilation jobs
   • Explain compilation jobs and their importance
   • Complete prerequisite for compilation jobs
   • Create and configure compilation jobs
3. Using a compilation job
   • Create a model
   • Deploy a model
4. Evaluating ML model operations
   • ‘Describe Model Monitor and the repository layout’
   • ‘Adjust triggers
   • parameters
   • and baselines’

SOFTWARE REQUIREMENTS

API Gateway, AWS Sagemaker, Access to AWS accounts, CodeBuild, CodePipeline, Lambda, S3