Snowflake SnowPro Advanced Data Scientist - SnowPro Advanced Data Scientist DSA-C03 Exam
Page: 2 / 13
Total 61 questions
Question #6 (Topic: Exam A)
A Data Scientist wants to get the coefficient of determination for two columns in a table.
Which function can be used for that?
Which function can be used for that?
A. REGR_INTERCEPT
B. REGR_SLOPE
C. REGR_R2
D. REGR_SXY
Answer: C
Question #7 (Topic: Exam A)
Which Snowflake functions are used in Linear Regression to find the equation of the best fit line? (Choose two.)
A. REGR_INTERCEPT
B. REGR_SLOPE
C. REGR_SXX
D. REGR_SXY
E. REGR_SYY
Answer: AB
Question #8 (Topic: Exam A)
A Data Scientist needs to track prediction accuracy across model versions in Snowflake Model Registry.
How can the Data Scientist associate an accuracy score with a model version with the LEAST operational overhead?
How can the Data Scientist associate an accuracy score with a model version with the LEAST operational overhead?
A. Set a prediction accuracy as a key-value pair metric score when creating a model version.
B. Create pre-define tags for prediction accuracy and tag model versions with accuracy scores.
C. Create a prediction accuracy time series with Snowpark ML time series package to keep track of the metric.
D. Create a Snowflake table to track prediction score by model version. Then write prediction accuracy scores into the table after registering a model version.
Answer: A
Question #9 (Topic: Exam A)
A Data Scientist wants to incorporate extensive fraud analysis in the transactions that are being fed into a data pipeline. One of the features they want to create is how many transactions a customer is doing over a period of time, going back to the last day, last seven days, and last year.
What is the MOST efficient way to build that feature?
What is the MOST efficient way to build that feature?
A. Write a SQL statement to retrieve data into memory and use Python code to calculate the new feature.
B. Copy the data into a bucket location, preprocess it using an external tool, and ingest it into Snowflake.
C. Use window functions to partition by customer and use sum functions to calculate the number of transactions.
D. Use a DataFrame to pull the data into memory to use Python functions to calculate the feature and write back into a table.
Answer: C
Question #10 (Topic: Exam A)
What are considerations for using Snowflake Notebooks? (Choose two.)
A. Notebooks support only SQL and Python cells.
B. Notebook drafts are automatically saved every three seconds.
C. Notebooks can be scheduled with third-party tools using the EXECUTE NOTEBOOK DB.SCHEMA.NOTEBOOK_NAME(); command.
D. The default state of a Notebook is to be connected to a virtual warehouse so that when opened, results are automatically generated.
E. A Notebook will keep its virtual warehouse active for 10 minutes after the last use of the websocket connection, if not closed.
Answer: AE
