From the course: Machine Learning with SageMaker by Pearson

Unlock this course with a free trial

Join today to access over 25,200 courses taught by industry experts.

Managing SageMaker Feature Store

Managing SageMaker Feature Store - Amazon SageMaker Tutorial

From the course: Machine Learning with SageMaker by Pearson

Start my 1-month free trial Buy for my team

Managing SageMaker Feature Store

“

You can think of SageMaker Feature Store as a database. And in fact, it is actually backed by a database, whether it's online in DynamoDB or offline in S3. So it's purpose-built repository for managing features. You can store feature data inside Feature Store. You use a thing called a Feature Group. You associate some data with a Feature Group. You use a thing called a feature group. You associate some information with that feature group, and then you can store data into that feature group. This enables you to reuse features across training and inference jobs, and it does integrate with SageMaker pipeline workflows. So it is a central repository for features. As I mentioned, you can think of it like a database. A feature is simply a single piece of information of information that is used to feed the training of a machine learning model. Feature groups are a logical grouping of features. And then a record is a single piece of data stored within that particular feature group. Here's the…

Contents