Summary

In this chapter, we learned about the concept of a feature store from an ML perspective. We described the functionality of Amazon SageMaker Feature Store and walked through several feature store use cases when developing an ML model using a public automotive dataset. In the example code, we showed you how to create a feature group in SageMaker Feature Store and how to ingest and update features and data to a feature group. We also showed you how to access features from the offline store for model training purposes and how to perform a point-in-time (time travel) feature query, which is useful when you need to access features in the past. Finally, we showed you how to access features from the online store for ML inference purposes.

In the next chapter, we will move into the topic of building and training ML models with the SageMaker Studio IDE. Building and training ML models can be challenging in a typical ML life cycle, as it is time-consuming and is compute resource-intensive...