Next steps and resources

This guide walks you through on a few considerations when planning the lifecycle of the machine learning models you want to bring to production. It discusses challenges and best practices in four areas—data, training, deployment, and monitoring—and includes additional relevant resources.

AWS provides the Well-Architected Framework, which helps cloud architects build secure, high-performing, resilient, and efficient infrastructures for a variety of applications, workloads, and technology domains. For additional reading, see the Machine Learning Lens offered by AWS Well-Architected.

Resources

Amazon SageMaker AI documentation

Amazon SageMaker AI Feature Store
Feature Store security and access control
Shapley values
Amazon SageMaker AI Debugger
Amazon SageMaker AI Pipelines
Amazon SageMaker AI default project templates
SageMaker AI real-time inference
Automatically scale Amazon SageMaker AI models
Amazon SageMaker AI asynchronous inference
SageMaker AI Model Monitor

AWS developer tools

AWS CodePipeline

AWS blog posts

Understanding the key capabilities of Amazon SageMaker AI Feature Store
Testing data quality at scale with PyDeequ
Amazon SageMaker AI Experiments
Safely deploying and monitoring Amazon SageMaker endpoints with CodePipeline and AWS CodeDeploy
Deploy shadow ML models in Amazon SageMaker AI
A/B Testing ML models in production using Amazon SageMaker AI

Warning Javascript is disabled or is unavailable in your browser.

To use the Amazon Web Services Documentation, Javascript must be enabled. Please refer to your browser's Help pages for instructions.

Document Conventions

Monitoring

Document history