Case Study

Structuring evaluation workflow for a new cat detector model

Case context: Your machine learning team is reviewing misclassified dev set images to improve a cat detector. During this error analysis phase, the team proposes several different ideas for enhancing the system.

Question: What workflow decision should you make regarding the evaluation of these multiple proposed ideas?

Sample answer: You should decide to efficiently evaluate all of these improvement ideas in parallel.

Key points:

  • Acknowledge the scenario of having several improvement ideas.
  • Decide to evaluate the ideas in parallel.
  • Recognize that this parallel approach maximizes efficiency.

Rubric: The answer should explicitly recommend the parallel evaluation of the several improvement ideas to achieve efficiency.

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