Concept icon
Concept

Why Different Dev and Test Distributions Complicate Diagnosis

When the development set and the test set come from different distributions, a model can look strong on development data and weak on test data for more than one reason. The main possibilities are that the model overfit the development set, the test set is genuinely harder, or the test set reflects a different distribution rather than a more difficult one.

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

Related