Databricks Machine Learning Professional exam dumps

Databricks Machine Learning Professional practice question 89 of 280

Databricks Certified Machine Learning Professional. Professional level, Databricks. Free question with the correct answer and a full explanation.

Databricks Machine Learning Professional Question 89

Single answer

You are managing a machine learning model registry in Databricks and want to automate the lifecycle of your models. Specifically, you need to ensure that models that have been in the 'Staging' stage for more than 30 days without being promoted to 'Production' are automatically archived, and models that have been archived for more than 90 days are deleted. Which feature or approach should you use to implement this automation?

  1. A

    Use Databricks Model Registry web UI to manually track and update model stages.

  2. B

    Implement a scheduled Databricks workflow with a Python script to track model stages and perform the required transitions.

  3. C

    Set up an automated transition rule in the Databricks Model Registry to handle stage transitions and deletions.

  4. D

    Use Databricks AutoML to automatically manage the lifecycle of models in the registry.

Show answer and explanation

Correct answer: B

Explanation

To automate the lifecycle of models in Databricks Model Registry, you can schedule a Databricks workflow that uses a Python script with the MLflow API. This approach allows you to programmatically implement logic for transitioning models to 'Archived' after 30 days and deleting models after 90 days. The other options do not provide the necessary automation capabilities.

  • A. Incorrect.

    The Databricks Model Registry web UI does not support automation for model lifecycle stages. This approach would require manual intervention and is not suitable for automating the described scenario.

  • B. Correct.

    Using a scheduled Databricks workflow with a Python script is a valid approach to automate model lifecycle transitions and deletions. You can use the MLflow API within the script to implement the required logic.

  • C. Incorrect.

    The Databricks Model Registry does not currently support automated transition rules for archiving or deleting models. This option is incorrect.

  • D. Incorrect.

    Databricks AutoML is designed to assist with model development and experimentation, not lifecycle management of registered models. This option is not applicable to the scenario.

Timed practice exam

Take a Databricks Machine Learning Professional practice test under exam conditions

60 questions in 120 minutes, drawn from this bank, with a score report and a per-question review when you finish.

Start timed exam