Home / ML Associate practice test / Databricks Machine Learning / Question

Databricks Machine Learning Associate · free practice question

A team computes updated customer aggregates every night and wants to refresh only the corresponding rows in an existing Unity Catalog feature table without…

Exam-style question on Databricks Machine Learning from the free ML Associate practice test, with the answer and a full explanation below.

Updated November 10, 2026

Databricks Machine Learning

A team computes updated customer aggregates every night and wants to refresh only the corresponding rows in an existing Unity Catalog feature table without overwriting untouched entities. Which Databricks Feature Engineering pattern is appropriate?

  1. ARead the table into pandas and rewrite the full dataset with saveAsTable each night, because partial updates are not supported for feature tables once they are governed in Unity Catalog
  2. BExport the feature table to DBFS, update it locally with Python dictionaries, and re-import it, because FeatureEngineeringClient cannot merge changed feature values into existing feature rows
  3. CUse FeatureEngineeringClient.write_table with mode set to merge so rows are updated based on the feature table primary key while unaffected rows remain unchanged
  4. DUse create_training_set to update the feature values, because training-set materialization is the standard Databricks mechanism for merging new feature values into stored feature tables
  5. EUse mlflow.log_metric inside a scheduled job, because feature table refreshes in Databricks are recorded as model metrics rather than data writes to Delta-backed feature tables
Answer & explanation

Correct answer: C

WHY C: The documented pattern for partial updates is FeatureEngineeringClient.write_table with mode='merge', which updates matching primary keys and leaves other rows unchanged. WHY NOT A: Full rewrites are not required for partial updates. WHY NOT B: Local export-import is unnecessary and brittle. WHY NOT D: create_training_set is for model training data assembly, not feature table writes. WHY NOT E: MLflow logging does not write feature table data.

Take the full ML Associate practice test free →

Would you pass today?

This question is one of 45 in the free, timed Databricks Machine Learning Associate practice test. Take it and get a section-by-section score.

Start the free test →