A data science team has trained several versions of a churn prediction model in BigQuery ML. They want to keep track of all versions and their metadata in a central place. What should they use?

A separate BigQuery table to manually log model versions

The Model Registry in Vertex AI

Cloud Storage to store model artifacts

The BigQuery INFORMATION_SCHEMA only

A marketing team wants to build a model to predict which customers are likely to respond to a campaign. They have a labeled dataset in BigQuery and want to use AutoML to train the model without writing SQL. Which service should they use?

Cloud Dataflow to train a custom model

Vertex AI LLM via BigQuery remote connection

A data engineer needs to create a BigQuery ML model that will be retrained monthly. They want to ensure that the model is versioned and that the training job is reproducible. They also want to avoid manual steps. What should they do?

Manually run CREATE MODEL each month and overwrite the existing model.

Create a new dataset for each month's model.

Use a scheduled query that runs CREATE MODEL with a fixed model name, and register the model in the Model Registry after each run.

Use a Cloud Function to retrain the model and store the model in Cloud Storage.

A team has trained a new version of a model and wants to register it in the Model Registry with a specific version name. What is the correct way to do this?

Create a new BigQuery dataset for each model version.

Store the model in Cloud Storage with a versioned filename.

Use the Vertex AI Model Registry to import the model and assign a version.

Overwrite the existing model in BigQuery without versioning.

Google CloudAssociate Data Practitioner

Domain 2Objective 3

2.3 Define, Train, Evaluate, and Use ML Models ASSOCIATE-DATA-PRACTITIONER Practice Questions (Page 5)

Part of the Data Analysis and Presentation domain, which accounts for ~27% of the ASSOCIATE-DATA-PRACTITIONER exam.

25questions here

5free pages

8concepts

~27%of the exam

Questions 21–25

21application · medium
A data science team has trained several versions of a churn prediction model in BigQuery ML. They want to keep track of all versions and their metadata in a central place. What should they use?
Select an answer first
22application · medium
A marketing team wants to build a model to predict which customers are likely to respond to a campaign. They have a labeled dataset in BigQuery and want to use AutoML to train the model without writing SQL. Which service should they use?
Select an answer first
23foundation · easy
A data analyst wants to invoke a pretrained Google large language model (LLM) directly from BigQuery SQL. What must be created first to enable this access?
Select an answer first
24expert · hard
A data engineer needs to create a BigQuery ML model that will be retrained monthly. They want to ensure that the model is versioned and that the training job is reproducible. They also want to avoid manual steps. What should they do?
Select an answer first
25application · medium
A team has trained a new version of a model and wants to register it in the Model Registry with a specific version name. What is the correct way to do this?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

No more pagesBack to ASSOCIATE-DATA-PRACTITIONER

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 Google Cloud. “ASSOCIATE-DATA-PRACTITIONER” is a trademark of its owner, used for identification only.