Concept icon
Concept

Training–Dev Distribution Gap

A training–dev distribution gap happens when a model handles fresh examples that come from the same distribution as the training data, but performs much worse on the dev/test data. The usual reason is that the training examples are not a good representation of the dev/test examples.

0

1

Concept icon
Updated 2026-08-12

Contributors are:

Who are from:

Tags

Machine Learning

Deep Learning

Supervised Learning

Dive into Deep Learning @ D2L

Data Science

Machine Learning Strategy

Machine Learning Yearning @ DeepLearning.AI