Learn Before
Diagnose a plot where the red dev-error curve stays above the blue training-error curve.
Case context: A plotted result uses a blue curve for training error and a red curve for dev error. The red curve lies strictly above the blue curve.
Question: Does this plot show the usual relationship described in the source, and what does it indicate about performance on the two sets?
Sample answer: Yes. The plot shows the usual relationship because the dev-error curve lies above the training-error curve. It indicates that the algorithm performs better on the training set than on the dev set.
Key points:
- The red curve represents dev error.
- The blue curve represents training error.
- Dev error lies above training error.
- The algorithm performs better on the training set.
Rubric: The response should identify the relationship as usual and correctly connect the higher dev-error curve with better performance on the training set.
0
1
Tags
Machine Learning
Deep Learning
Supervised Learning
Dive into Deep Learning @ D2L
Data Science
Machine Learning Strategy
Machine Learning Yearning @ DeepLearning.AI
Related
Which relationship between training and dev error is usual for a learning algorithm?
The dev-error curve usually lies strictly above the training-error curve.
The algorithm usually has lower error on the _____ set.
Match each element of the error plot to its usual interpretation.
Order the reasoning that predicts the usual positions of the two error curves.
Explain why the dev-error curve is usually above the training-error curve.
Diagnose a plot where the red dev-error curve stays above the blue training-error curve.
What does a dev-error curve above the training-error curve indicate?
How should the two curves usually be positioned when blue denotes training error and red denotes dev error?
A plot with training error strictly above dev error displays the usual relationship.