Snowflake DSA-C03 試験概要:

Snowflake DSA-C03 試験シラバストピック:

Snowflake SnowPro Advanced: Data Scientist Certification 認定 DSA-C03 試験問題：

1. A retail company is using Snowflake to store transaction data'. They want to create a derived feature called 'customer _ recency' to represent the number of days since a customer's last purchase. The transactions table 'TRANSACTIONS has columns 'customer_id' (INT) and 'transaction_date' (DATE). Which of the following SQL queries is the MOST efficient and scalable way to derive this feature as a materialized view in Snowflake?

A) Option B
B) Option D
C) Option A
D) Option E
E) Option C

2. You have deployed a vectorized Python UDF in Snowflake to perform sentiment analysis on customer reviews. The UDF uses a pre-trained transformer model loaded from a Stage. The model consumes a significant amount of memory (e.g., 5GB). Users are reporting intermittent 'Out of Memory' errors when calling the UDF, especially during peak usage. Which of the following strategies, used IN COMBINATION, would MOST effectively mitigate these errors and optimize resource utilization?

A) Increase the value of 'MAX BATCH_ROWS' for the UDF to process larger batches of data at once.
B) Reduce the value of 'MAX for the UDF to process smaller batches of data.
C) Increase the warehouse size to provide more memory per node.
D) Partition the input data into smaller chunks using SQL queries and call the UDF on each partition separately.
E) Implement lazy loading of the model within the UDF, ensuring it's only loaded once per warehouse node and reused across multiple invocations within that node.

3. You are tasked with building a fraud detection model using Snowflake and Snowpark Python. The model needs to identify fraudulent transactions in real-time with high precision, even if it means missing some actual fraud cases. Which combination of optimization metric and model tuning strategy would be most appropriate for this scenario, considering the importance of minimizing false positives (incorrectly flagging legitimate transactions as fraudulent)?

A) Precision, optimized with a threshold adjustment to minimize false positives.
B) AUC-ROC, optimized with a randomized search focusing on hyperparameters related to model complexity.
C) F 1-Score, optimized to balance precision and recall equally.
D) Log Loss, optimized with a grid search focusing on hyperparameters that improve overall accuracy.
E) Recall, optimized with a threshold adjustment to minimize false negatives.

4. You are developing a Snowflake Native App that leverages Snowflake Cortex for text summarization. The app needs to process user-provided text input in real-time and return a summarized version. You want to expose this functionality as a secure and scalable REST API endpoint within the Snowflake environment. Which of the following strategies are MOST suitable for achieving this, considering best practices for security and performance?

A) Utilize a Snowflake Stored Procedure written in SQL that invokes the 'SNOWFLAKE.CORTEX.SUMMARIZE' function, and then create a Snowflake API Integration to expose the stored procedure as a REST endpoint.
B) Develop a Snowflake Native App containing a Python UDF that calls 'SNOWFLAKCORTEX.SUMMARIZE function, and expose it as a REST API endpoint using Snowflake's API Integration feature within the app package.
C) Write a Snowflake Stored Procedure using Javascript to invoke the 'SNOWFLAKE.CORTEX.SUMMARIZE function, deploy the procedure to a Snowflake stage, and then trigger it via an AWS Lambda function integrated with Snowflake.
D) Create a Snowflake External Function using Python that directly calls the 'SNOWFLAKE.CORTEX.SUMMARIZE' function and expose this function via a REST API gateway outside of Snowflake.
E) Develop a Snowflake Native App that includes a Java UDF that calls 'SNOWFLAKE.CORTEX.SUMMARIZE and expose a REST API using Snowflake's built-in REST API capabilities within the Native App framework.

5. You are developing a churn prediction model and want to track its performance across different model versions using the Snowflake Model Registry. After registering a new model version, you need to log evaluation metrics (e.g., AUC, F 1-score) and custom tags associated with the training run. Assuming you have a registered model named 'churn_model' with version 'v2', which of the following code snippets demonstrates the correct way to log these metrics and tags using the Snowflake Python Connector and the 'ModelRegistry' API?

A)

B)

C)

D)

E)

質問と回答：