What is the purpose of deploying a custom model to a model endpoint?

To visualize the model's feature importance

To make the model available for real-time predictions via an API

To store the model in a data lake for future training

To create a backup of the model artifacts

A team has deployed a custom model to a Databricks model endpoint. They notice that during peak hours, requests are slow and some time out. They want to improve performance without over-provisioning during off-peak hours. What should they configure?

Increase the compute size to a larger VM type and keep the same number of instances.

Deploy a second endpoint and use a load balancer to distribute traffic.

Set a longer timeout for the endpoint to allow requests to complete.

Increase the scale-out settings to allow more concurrent requests and enable auto-scaling based on traffic.

A team wants to deploy a model endpoint that can handle a variable workload with unpredictable spikes. They also want to minimize cost during low-traffic periods. What should they configure?

Disable auto-scaling and manually adjust the instance count based on expected load.

Set a fixed number of instances high enough to handle the maximum expected load.

Enable auto-scaling with a minimum of one instance and a maximum of several instances.

Use a single instance and rely on the endpoint's built-in queuing to handle spikes.

A data scientist has deployed a custom model to a Databricks model endpoint. They want to test the endpoint by sending a sample request from a Python notebook. What should they do?

Use the Databricks SQL API to query the endpoint.

Use the Databricks SDK for Python to call the endpoint's `predict` method with the input data.

Use the MLflow client to load the model and call `predict` locally.

Use the `requests` library to send a GET request to the endpoint URL with the input data in the query string.

What is the purpose of packaging a custom model with its dependencies and environment for deployment?

To ensure the model runs consistently across different serving environments

To reduce the size of the model file for faster storage

To convert the model into a format that can be used by any programming language

To encrypt the model so that only authorized users can access it

DatabricksCertified Machine Learning Associate

Domain 4Objective 2

Deploy a Custom Model to a Model Endpoint MACHINE-LEARNING-ASSOCIATE Practice Questions (Page 3)

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 11–15

11foundation · easy
What is the purpose of deploying a custom model to a model endpoint?
Select an answer first
12application · medium
A team has deployed a custom model to a Databricks model endpoint. They notice that during peak hours, requests are slow and some time out. They want to improve performance without over-provisioning during off-peak hours. What should they configure?
Select an answer first
13expert · hard
A team wants to deploy a model endpoint that can handle a variable workload with unpredictable spikes. They also want to minimize cost during low-traffic periods. What should they configure?
Select an answer first
14application · medium
A data scientist has deployed a custom model to a Databricks model endpoint. They want to test the endpoint by sending a sample request from a Python notebook. What should they do?
Select an answer first
15foundation · easy
What is the purpose of packaging a custom model with its dependencies and environment for deployment?
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.