CERTIFIED-MACHINE-LEARNING-PROFESSIONAL · Question #43
CERTIFIED-MACHINE-LEARNING-PROFESSIONAL Question #43: Real Exam Question with Answer & Explanation
The correct answer is A: fs.create_table. Note: The marked correct answer (A) appears to be incorrect based on Databricks Feature Store documentation. The actual correct answer is E - fs.read_table. fs.read_table is specifically designed to read a Feature Store table and return a Spark DataFrame of the associated dataset
Question
Which of the following operations in Feature Store Client fs can be used to return a Spark DataFrame of a data set associated with a Feature Store table?
Options
- Afs.create_table
- Bfs.write_table
- Cfs.get_table
- DThere is no way to accomplish this task with fs
- Efs.read_table
Explanation
Note: The marked correct answer (A) appears to be incorrect based on Databricks Feature Store documentation. The actual correct answer is E - fs.read_table.
fs.read_table is specifically designed to read a Feature Store table and return a Spark DataFrame of the associated dataset, making it the right tool when you need to work with feature data programmatically. fs.create_table (A) creates a new Feature Store table and returns a metadata object, not a DataFrame - its purpose is provisioning, not reading. fs.write_table (B) writes/updates data into an existing table and returns nothing useful for reading. fs.get_table (C) returns a FeatureTable metadata object (name, schema, description, etc.) - not the actual data rows. Option D is wrong because the capability clearly exists via fs.read_table.
Memory tip: Think read → DataFrame - just like spark.read returns data, fs.read_table reads a Feature Store table into a Spark DataFrame. The verb "read" always signals data retrieval, while "get" signals metadata retrieval.
Community Discussion
No community discussion yet for this question.