Train/test split

Basic train/test evaluation.

0/9

completed

mediumTest Set Contamination Through Repeated Use
4/9

A data scientist evaluates a model on the same test set repeatedly while tuning hyperparameters. What problem does this cause?