A BatchFeatureTable object is a representation of a table in the feature store that contains feature values from batch serving. The object includes metadata on the Deployment and the BatchRequestTable used to create it.
Creating BatchFeatureTable Objects
A BatchFeatureTable Object is created by getting batch features from a Deployment object and a BatchRequestTable object.
To get batch features, follow those steps:
Get a BatchRequestTable object from the catalog using the
batch_request_table = catalog.get_batch_request_table(<batch_request_table_name>)
Get a Deployment object from the catalog using the [
deployment = catalog.get_deployment(<deployment_name>)
- Use the
compute_batch_feature_table() method to retrieve batch features, passing in the BatchRequestTable object and optionally a desired name for the BatchFeatureTable:
batch_features = deployment.compute_batch_feature_table(
batch_feature_table_name = <batch_feature_table_name>
You can then download or delete the BatchFeatureTable object using the
delete() methods. Deleting the object will delete the table in the feature store.
You can convert the table to a Pandas DataFrame by using the
Getting BatchFeatureTable Lineage
The BatchFeatureTable object contains information on the Deployment and BatchRequestTable used to create it which can be accessed using the
batch_request_table_id properties. This returns the Object IDs of the two objects.
Listing and Retrieving BatchFeatureTable Objects
To list the BatchFeatureTable objects in the catalog, use the
To retrieve a specific BatchFeatureTable by its name from the catalog, use the
batch_features = catalog.get_batch_feature_table(<batch_feature_table_name>)
To retrieve a specific BatchFeatureTable by its Object ID from the catalog, use the
batch_features = catalog.get_batch_feature_table_by_id(<batch_feature_table_id>)