How can Snowflake integrate with an external model serving platform like Amazon SageMaker?

By using Snowpipe to stream data to SageMaker

By using an external function that calls the SageMaker endpoint

By using a Snowflake connector to SageMaker

By importing the SageMaker model directly into Snowflake

A company uses Azure ML to host a large deep-learning model that cannot be embedded in Snowflake. They want to call this model from Snowflake for real-time inference on a small number of rows. Which approach should they use?

Copy the data to Azure ML and run inference there.

Use a Snowpark stored procedure to call the Azure ML API.

Create an external function that calls the Azure ML endpoint.

Create a Python UDF that imports the model from Azure ML.

A team has a PyTorch model that they want to deploy in Snowflake to score a large table. They plan to use Snowpark Python. Which deployment method will give the best performance for batch inference on millions of rows?

Create a scalar Python UDF that processes one row at a time.

Use an external function to call a remote inference service.

Create a stored procedure that loops through the table and inserts predictions.

Create a vectorized Python UDF that processes batches of rows.

A team has deployed a model and wants to monitor its performance over time. They have a table of predictions, but they need to compare predictions against actual outcomes that arrive later. What is the best way to set this up?

Use an external function to send predictions to a monitoring service.

Create a view that joins predictions with actual outcomes using a common key.

Use a stored procedure to update predictions with actuals.

Store predictions and actuals in separate tables and never join them.

SnowflakeSnowPro Advanced — Data Scientist

Domain 5Objective 1

Deploy Models in Snowflake SNOWPRO-ADVANCED-DATA-SCIENTIST Practice Questions (Page 1)

Part of the Model Deployment and Operationalization domain, which makes up ~24% of our current practice bank.

27questions here

6free pages

8concepts

Questions 1–5

1foundation · easy
How can Snowflake integrate with an external model serving platform like Amazon SageMaker?
Select an answer first
2foundation · easy
In Snowpark, which object can be used to deploy a model for inference on a schedule?
Select an answer first
3application · medium
A company uses Azure ML to host a large deep-learning model that cannot be embedded in Snowflake. They want to call this model from Snowflake for real-time inference on a small number of rows. Which approach should they use?
Select an answer first
4application · medium
A team has a PyTorch model that they want to deploy in Snowflake to score a large table. They plan to use Snowpark Python. Which deployment method will give the best performance for batch inference on millions of rows?
Select an answer first
5expert · hard
A team has deployed a model and wants to monitor its performance over time. They have a table of predictions, but they need to compare predictions against actual outcomes that arrive later. What is the best way to set this up?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to SNOWPRO-ADVANCED-DATA-SCIENTIST

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 Snowflake. “SNOWPRO-ADVANCED-DATA-SCIENTIST” is a trademark of its owner, used for identification only.