Databricks Generative AI Engineer Associate Question 213
Select 4You have trained a machine learning model using Databricks and logged it with MLflow. Your organization requires all production-ready models to be registered with Unity Catalog for better governance and sharing. Which of the following steps must be performed to register the model to Unity Catalog using MLflow?
- A
Ensure the Unity Catalog is enabled for your Databricks workspace.
- B
Use the MLflow
log_modelmethod to log the model to the Unity Catalog. - C
Assign a Unity Catalog model registry URI before registering the model.
- D
Use the MLflow
register_modelfunction and specify the Unity Catalog model registry URI. - E
Grant appropriate permissions for the model in the Unity Catalog after registration.
Show answer and explanation
Correct answers: A, C, D, E
Explanation
Registering a model to Unity Catalog using MLflow involves enabling Unity Catalog in your Databricks workspace, specifying the Unity Catalog model registry URI, registering the model using the register_model function, and finally setting up permissions for the model in Unity Catalog. The log_model function is unrelated to the registration process.
- A. Correct.
Correct: Unity Catalog must be enabled in your Databricks workspace to manage models and ensure governance.
- B. Incorrect.
Incorrect: The
log_modelmethod is used for logging a model artifact, not directly for registering it to Unity Catalog. - C. Correct.
Correct: You need to specify the Unity Catalog model registry URI to integrate MLflow with Unity Catalog.
- D. Correct.
Correct: The
register_modelfunction in MLflow is used to register the model, and the Unity Catalog model registry URI must be specified. - E. Correct.
Correct: After registering the model, appropriate permissions must be assigned within Unity Catalog to control access to the model.