Essay

Explain how pipeline component error counts dictate development priorities.

Question: A developer reviews 100 misclassified dev-set images from an image processing pipeline. They find that 90 errors are attributable to the cat detector and 10 errors are attributable to the cat breed classifier. Explain how the developer should use these error counts to determine which component to prioritize for improvement.

Sample answer: The developer should analyze the error counts to determine which component is the primary source of failure. Since the cat detector is responsible for 90 of the errors and the breed classifier is only responsible for 10, the cat detector represents the major bottleneck in the pipeline. Therefore, the developer can safely conclude that they should focus more attention on improving the cat detector to achieve the most significant performance gains.

Key points:

  • Attribute the 100 dev-set errors to their respective components.
  • Compare the counts: 90 errors from the cat detector versus 10 from the breed classifier.
  • Focus more attention on improving the cat detector based on it being the dominant source of errors.

Rubric: The response must explain that error counts attribute failures to specific pipeline components, identify the cat detector as causing 90 errors and the breed classifier as causing 10, and conclude that the team should prioritize improving the cat detector.

0

1

Updated 2026-05-27

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