What is the purpose of versioning data in a machine learning pipeline?

To speed up model training by reducing data size

To automatically clean the data

To ensure that experiments can be reproduced with the same data

A company is building a recommendation system that requires real-time features from user interactions and batch features from historical data. They want to use Snowflake for data storage and feature computation, but also need to serve features to an online model with low latency. Which architecture should they use?

Use an external stream processing engine for real-time features and store them in Snowflake, then serve from Snowflake.

Use Snowflake to serve features directly to the online model via a Snowflake connector.

Use Snowflake for batch feature computation and store the features in a low-latency feature store (e.g., Redis) for online serving.

Use Snowflake for both batch and real-time feature computation, and use a materialized view to serve features.

In Snowflake, what is the purpose of a stream in the context of pipeline automation?

To track changes to a table for incremental processing

SnowflakeSnowPro Advanced — Data Scientist

Domain 2Objective 1

Build Data Pipelines for Data Science Workloads SNOWPRO-ADVANCED-DATA-SCIENTIST Practice Questions (Page 4)

Part of the Data Pipelining domain, which makes up ~15% of our current practice bank.

26questions here

6free pages

10concepts

Questions 16–20

16foundation · easy
What is the purpose of versioning data in a machine learning pipeline?
Select an answer first
17foundation · easy
Which SQL function is commonly used to replace NULL values with a specified default value during data cleaning?
Select an answer first
18foundation · easy
In an end-to-end data pipeline for a machine learning workload, which stage is specifically responsible for converting raw data into a format suitable for model training?
Select an answer first
19expert · hard
A company is building a recommendation system that requires real-time features from user interactions and batch features from historical data. They want to use Snowflake for data storage and feature computation, but also need to serve features to an online model with low latency. Which architecture should they use?
Select an answer first
20foundation · easy
In Snowflake, what is the purpose of a stream in the context of pipeline automation?
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.

Build Data Pipelines for Data Science Workloads SNOWPRO-ADVANCED-DATA-SCIENTIST Practice Questions (Page 4)

Questions 16–20

What is the purpose of versioning data in a machine learning pipeline?

Which SQL function is commonly used to replace NULL values with a specified default value during data cleaning?

In an end-to-end data pipeline for a machine learning workload, which stage is specifically responsible for converting raw data into a format suitable for model training?

In Snowflake, what is the purpose of a stream in the context of pipeline automation?

Keep studying — Data Pipelining