Case Study

Choose between pipeline components based on their training-data availability.

Case context: A team is designing a non-end-to-end pipeline and is comparing two candidate components. It can easily collect data to train Component A, but collecting data to train Component B would be difficult.

Question: Using only the stated component-selection factor, which component should the team favor, and what should it diagnose about Component B?

Sample answer: The team should favor Component A because its training data can be collected easily. It should diagnose Component B as a weaker candidate under this criterion because the data needed to train it is difficult to collect.

Key points:

  • Favor Component A under the stated criterion.
  • Component A has easily collectable training data.
  • Component B is weakened by difficult data collection.
  • The decision concerns a non-end-to-end pipeline.

Rubric: The response should favor Component A, identify easy training-data collection as the reason, and characterize Component B only in terms of its weaker training-data availability.

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