Concept icon
Concept

Keep the Test Set Out of Routine Model Decisions

It is reasonable to run the test set at regular intervals, such as every week or every month, to report overall progress. It should not be used to choose model changes, tune parameters, or decide whether to revert to an earlier version. If those decisions are made from test-set results, the model begins to fit the test set itself, and the test set no longer provides an unbiased estimate of real-world performance.

0

1

Concept icon
Updated 2026-08-12

Contributors are:

Who are from:

Tags

Machine Learning

Deep Learning

Machine Learning Strategy

Supervised Learning

Dive into Deep Learning @ D2L

Data Science

Machine Learning Yearning @ DeepLearning.AI