Summary

In this chapter, we discussed the benefits of using multiple accounts to manage and operate machine learning workloads that use Amazon SageMaker across the ML Lifecycle. We also looked at common patterns for account isolation across the ML Lifecycle. Finally, we focused specifically on the SageMaker features that are most often used across accounts, and the considerations you should be aware of when architecting and building end-to-end machine learning solutions.

This chapter wraps up the book where we covered best practices for SageMaker across features spanning the machine learning lifecycle of data preparation, model training, and operations. In this book, we discussed best practices, as well as considerations, that you can draw on when creating your own projects. We used an example use case, using open weather data to demonstrate the concepts throughout the chapters of the book. This was done so you can get hands-on with the concepts and practices discussed. We hope...