A data science team has trained a scikit-learn model and wants to deploy it to a Databricks model endpoint. They have already logged the model with MLflow, including a conda environment. They now need to create the endpoint using the Databricks UI. What is the minimal set of actions they must take?

Create a cluster, install the model as a library, and then attach the cluster to the endpoint.

Upload the model artifacts to DBFS and then reference the path when creating the endpoint.

Register the model in the Unity Catalog, select the model version, and choose a compute size and scale-out settings.

Export the model as a pickle file and upload it directly to the endpoint configuration.

Which of the following is required when creating a model endpoint in Databricks?

A registered model version in the MLflow Model Registry

A team has a model endpoint serving version 1. They have trained version 2 and want to deploy it, but they want to be able to quickly revert to version 1 if issues arise. They also want to minimize the number of endpoints. What should they do?

Delete version 1 from the registry to force the endpoint to use version 2.

Create a second endpoint for version 2 and keep the original endpoint for version 1.

Update the existing endpoint to serve version 2, and keep version 1 in the registry for rollback.

Use the API to update the endpoint to version 2 and then immediately delete version 1.

How do you typically invoke a model endpoint to get predictions?

Read the model's output from a Delta table

Send an HTTP request to the endpoint's REST API with input data

Run a SQL query against the endpoint

Use the Databricks CLI to run a notebook

A team has deployed a model endpoint and wants to roll back to a previous model version if the new version performs poorly. What should they do?

Delete the new model version from the registry.

Retrain the model and redeploy as a new version.

Create a new endpoint with the previous version and redirect traffic.

Update the endpoint to serve the previous model version using the UI or API.

DatabricksCertified Machine Learning Associate

Domain 4Objective 2

Deploy a Custom Model to a Model Endpoint MACHINE-LEARNING-ASSOCIATE Practice Questions (Page 4)

Part of the Section 4: Model Deployment domain, which makes up ~30% of our current practice bank. Databricks does not publish an official question count, but from its 90-minute exam (~35–60 total, ~11–18 in this domain), expect 2–3 from this objective — we provide 27 practice questions to prepare you well beyond it. (estimate)

27questions here

6free pages

7concepts

Questions 16–20

16application · medium
A data science team has trained a scikit-learn model and wants to deploy it to a Databricks model endpoint. They have already logged the model with MLflow, including a conda environment. They now need to create the endpoint using the Databricks UI. What is the minimal set of actions they must take?
Select an answer first
17foundation · easy
Which of the following is required when creating a model endpoint in Databricks?
Select an answer first
18expert · hard
A team has a model endpoint serving version 1. They have trained version 2 and want to deploy it, but they want to be able to quickly revert to version 1 if issues arise. They also want to minimize the number of endpoints. What should they do?
Select an answer first
19foundation · easy
How do you typically invoke a model endpoint to get predictions?
Select an answer first
20application · medium
A team has deployed a model endpoint and wants to roll back to a previous model version if the new version performs poorly. What should they do?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to MACHINE-LEARNING-ASSOCIATE

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by Databricks. “MACHINE-LEARNING-ASSOCIATE” is a trademark of its owner, used for identification only.