Databricks Machine Learning Associate exam dumps

Databricks Machine Learning Associate practice question 487 of 656

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

Databricks Machine Learning Associate Question 487

Single answer

You are managing machine learning models in a Databricks workspace using MLflow. You have a currently deployed champion model with the alias 'production' and a newly trained challenger model that has shown better performance in testing. How can you promote the challenger model to replace the champion model as the 'production' model using MLflow aliases?

  1. A

    Use the MLflow CLI command mlflow models alias update to update the alias 'production' to point to the challenger model.

  2. B

    Delete the existing alias 'production' and create a new alias 'production' pointing to the challenger model.

  3. C

    Update the 'production' alias directly from the MLflow UI by selecting the challenger model and assigning the alias.

  4. D

    Use the MLflow Python API to create a new alias 'production' for the challenger model without deleting the old alias.

Show answer and explanation

Correct answer: A

Explanation

To promote a challenger model to a champion model in MLflow, you can use the CLI or Python API to update the alias pointing to the champion model (e.g., 'production') to the new challenger model version. This process ensures a smooth transition without deleting or recreating aliases, which could disrupt workflows.

  • A. Correct.

    This is the correct option because the MLflow CLI command mlflow models alias update allows you to reassign an alias like 'production' to a new model version, effectively promoting the challenger model without requiring the deletion of the alias.

  • B. Incorrect.

    This is incorrect because MLflow aliases are designed to be updated without requiring deletion and recreation, which avoids unnecessary complexity.

  • C. Incorrect.

    This is incorrect because the MLflow UI currently does not support directly updating aliases. Alias operations need to be performed through the CLI or API.

  • D. Incorrect.

    This is incorrect because creating a new alias with the same name ('production') without deleting the old alias would result in a conflict. The correct approach is to update the alias.

Timed practice exam

Take a Databricks Machine Learning Associate practice test under exam conditions

48 questions in 90 minutes, drawn from this bank, with a score report and a per-question review when you finish.

Start timed exam

More Databricks Machine Learning Associate practice questions

  1. 488.You are managing machine learning models in Databricks' Model Registry. A new model version has...
  2. 489.You are managing machine learning models in Databricks Model Registry. A new model version (Version...
  3. 490.You are managing a machine learning workflow in Databricks and have a registered model with two...
  4. 491.You are managing machine learning models in a Databricks MLflow model registry. Your team has...
  5. 492.You are tasked with preparing a dataset for a machine learning project on Databricks. The dataset...
Back to the Databricks Machine Learning Associate exam dumps