Databricks Generative AI Engineer Associate Question 211
Select 3You have trained a machine learning model using Databricks and want to register it to Unity Catalog using MLflow for centralized governance and sharing across workspaces. Which of the following steps is required to successfully register the model to Unity Catalog?
- A
Ensure that the MLflow Model Registry is configured to use Unity Catalog.
- B
Assign a schema in Unity Catalog where the model will be stored.
- C
Use the MLflow CLI to directly register the model to Unity Catalog.
- D
Ensure your Databricks workspace has Unity Catalog enabled and properly configured.
- E
Manually copy the model's artifacts to a Unity Catalog-enabled storage location.
Show answer and explanation
Correct answers: A, B, D
Explanation
To register a model to Unity Catalog using MLflow, the MLflow Model Registry must be configured to work with Unity Catalog, and the model must be associated with a schema in Unity Catalog. Additionally, the Databricks workspace must have Unity Catalog enabled and set up. These steps ensure that the model can be managed and governed properly within Unity Catalog's structure and security framework. Manual artifact copying and using the MLflow CLI alone are not sufficient for this process.
- A. Correct.
Correct. Configuring the MLflow Model Registry to use Unity Catalog is required to ensure model registration integrates with Unity Catalog's governance features.
- B. Correct.
Correct. Models registered to Unity Catalog must be associated with a specific schema, as schemas define their organizational structure in the catalog.
- C. Incorrect.
Incorrect. While the MLflow CLI can interact with MLflow Model Registry, registering models to Unity Catalog specifically requires interaction via the Databricks environment and not the CLI alone.
- D. Correct.
Correct. Unity Catalog must be enabled and configured in the workspace to allow models to be registered and governed within it.
- E. Incorrect.
Incorrect. Manually copying model artifacts is not a required or recommended step for registering a model to Unity Catalog. MLflow handles the artifact management automatically during registration.