Databricks Machine Learning Associate Question 126
Single answerYou are managing a machine learning model in Databricks and have registered it in the Model Registry. The model has completed testing in the 'Staging' stage and is ready to be moved to 'Production.' Using the Model Registry UI, what steps should you take to transition the model's stage to 'Production'?
- A
Click the 'Move to Production' button and confirm the transition in the dialog box.
- B
Delete the current version of the model and re-register it with the 'Production' stage.
- C
Select the model version, click 'Transition Stage,' choose 'Production,' and confirm the action.
- D
Export the model, re-import it, and set its stage to 'Production' during re-import.
Show answer and explanation
Correct answer: C
Explanation
The Databricks Model Registry provides a built-in UI to manage the lifecycle of machine learning models. To transition a model's stage, you select the model version, use the 'Transition Stage' option, and confirm the desired stage change. This streamlined process eliminates the need for manual re-registration or external exporting/re-importing.
- A. Incorrect.
Incorrect: There is no single 'Move to Production' button in the Model Registry UI. Stage transitions are handled through the 'Transition Stage' option.
- B. Incorrect.
Incorrect: Deleting and re-registering the model is unnecessary and inefficient. Stage transitions can be done directly in the Model Registry.
- C. Correct.
Correct: The correct way to transition a model's stage in the Model Registry UI is to select the model version, click 'Transition Stage,' choose the desired stage (in this case, 'Production'), and confirm the action.
- D. Incorrect.
Incorrect: Exporting and re-importing the model is not required to transition its stage. The Model Registry provides built-in functionality for stage transitions.