Short Answer

What limitation motivates adding a training-error plot?

Question: Answer in one to three sentences: What problem with the dev-error curve motivates plotting training error?

Sample answer: The dev-error curve by itself can be hard to extrapolate, making it difficult to predict exactly where it will go with more data. Training error is added to help estimate the impact of adding data.

Key points:

  • Dev error alone can be hard to extrapolate.
  • Training error helps estimate the impact of more data.

Rubric: The answer should mention difficult extrapolation from dev error alone and identify training error as the additional plot used to support the estimate.

0

1

Updated 2026-07-20

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