Quiz 2 Question 17 of 20

Before training an ML model with Spark MLlib, a large training DataFrame has very uneven partition sizes (some 50MB, others 500MB). What operation should be applied to optimize training performance?

Select an answer to reveal the explanation.

Motivation