SOL-C01 Question #240: Real Exam Question with Answer & Explanation

Question

A data scientist wants to leverage Snowflake's A1 Data Cloud capabilities to train a machine learning model directly on data stored in Snowflake. The data is stored in a table named 'customer_data' and contains features like 'age', 'income', 'transaction _ history', and a target variable `churn'. The data scientist intends to use Snowpark ML for model training. Which of the following steps are REQUIRED to enable this scenario?

Options

• AGrant the necessary privileges to the data scientist's role to access the `customer_datas table
• BInstall the Snowpark library in the data scientist's local environment and configure it to connect to
• CCreate a Snowflake Anaconda channel integration to manage Python dependencies for Snowpark
• DEnable the external access integration for outbound network access to the internet. This will allow
• ECreate a service account in Snowflake and grant the necessary privileges to this service account.

Explanation

Snowpark ML relies on Anaconda to manage Python dependencies within Snowflake. Therefore, you must set up a Snowflake Anaconda channel integration (Option C) so you can use Python packages. The data scientist also needs read access to the data (Option A). Options D and E involve more complex setups (external access and service accounts), which are not necessary for the baseline scenario of training a model directly on data within Snowflake using Snowpark ML and Anaconda packages. Option B describes setting up Snowpark on local IDEs to connect to Snowflake, it doesn't relate to using Snowflake native capability.

No community discussion yet for this question.