Databricks Machine Learning Associate Question 430
Select 4You are working on a machine learning project in Databricks and need to create a feature store table in Unity Catalog to manage features for reuse across multiple ML models. Which of the following steps are required to successfully create a feature store table in Unity Catalog?
- A
Define the feature table schema using a FeatureStoreClient.
- B
Specify a Unity Catalog metastore when creating the feature store table.
- C
Use Delta format for the feature store table.
- D
Run a Databricks SQL query to directly create the feature store table.
- E
Ensure the feature store table is registered in a catalog and schema in Unity Catalog.
Show answer and explanation
Correct answers: A, B, C, E
Explanation
To create a feature store table in Unity Catalog, you must define the schema using the FeatureStoreClient, specify a Unity Catalog metastore, use Delta format, and register the table within a catalog and schema. These steps ensure that the feature table is managed properly and can be reused across multiple ML projects. Running a SQL query alone cannot substitute the required steps for creating a feature store table.
- A. Correct.
Correct: You need to define the feature table schema using the FeatureStoreClient as it is the standard interface for creating and managing feature store tables in Databricks.
- B. Correct.
Correct: Unity Catalog requires specifying a metastore to manage data governance for feature tables.
- C. Correct.
Correct: Feature store tables in Databricks must use Delta format to ensure compatibility with the feature store and efficient data handling.
- D. Incorrect.
Incorrect: Creating a feature store table cannot be done solely through a SQL query. The FeatureStoreClient API is required.
- E. Correct.
Correct: Unity Catalog requires the feature store table to be registered within a catalog and schema to enable governance and reuse.