[GH-ISSUE #314] Reader nits : Chapter 7 suggestions #3978

New Issue

GiteaMirror · 2026-04-19T12:01:48-05:00

GiteaMirror commented

2026-04-19 12:01:48 -05:00

Originally created by @emmanuel2406 on GitHub (Jul 12, 2024).
Original GitHub issue: https://github.com/harvard-edge/cs249r_book/issues/314

Originally assigned to: @profvjreddi, @jasonjabbour on GitHub.

A few nits I picked up while reading this chapter. Please notify me whether this is the right way to propose these changes.

Video 7.1 and 7.2 links do not work/ not added yet
Section 7.4.2 : The terms "stratify" and "stratification" should be defined earlier before it's used for unfamiliar readers.
Redundancies of paragraphs in section 7.4.2: “No unseen data for final evaluation” is merely describing the need for a final test set and should either be put more forward or removed since it is covered by the paragraph of “reusing the test set multiple times”
Section 7.7.2 Dropout : inconsistency with what p refers to. Should be the probability of dropping a node as opposed to keeping it. The Bernoulli trial definition could confuse some readers.
Section 7.9 & 7.8: Activation section introduces some of the core activation functions with their formulae. 7.8 refers to this section quite a bit, and so switching them would make a better order.

Originally created by @emmanuel2406 on GitHub (Jul 12, 2024). Original GitHub issue: https://github.com/harvard-edge/cs249r_book/issues/314 Originally assigned to: @profvjreddi, @jasonjabbour on GitHub. A few nits I picked up while reading this chapter. Please notify me whether this is the right way to propose these changes. 1. Video 7.1 and 7.2 links do not work/ not added yet 2. Section 7.4.2 : The terms "stratify" and "stratification" should be defined earlier before it's used for unfamiliar readers. 3. Redundancies of paragraphs in section 7.4.2: “No unseen data for final evaluation” is merely describing the need for a final test set and should either be put more forward or removed since it is covered by the paragraph of “reusing the test set multiple times” 4. Section 7.7.2 Dropout : inconsistency with what p refers to. Should be the probability of dropping a node as opposed to keeping it. The Bernoulli trial definition could confuse some readers. 5. Section 7.9 & 7.8: Activation section introduces some of the core activation functions with their formulae. 7.8 refers to this section quite a bit, and so switching them would make a better order.

GiteaMirror added the area: book type: improvement labels 2026-04-19 12:01:48 -05:00

GiteaMirror closed this issue

2026-04-19 12:01:49 -05:00

GiteaMirror commented

2026-04-19 12:01:50 -05:00

@profvjreddi commented on GitHub (Jul 30, 2024):

Thanks @emmanuel2406 we are working through quite a few fixes, so appreciate the feedback. We will push some updates soon.

@profvjreddi commented on GitHub (Jul 30, 2024): Thanks @emmanuel2406 we are working through quite a few fixes, so appreciate the feedback. We will push some updates soon.

GiteaMirror commented

2026-04-19 12:01:51 -05:00

@jasonjabbour commented on GitHub (Sep 1, 2024):

Thanks Emmanuel! This has been addressed in #349. Closing this issue out.

@jasonjabbour commented on GitHub (Sep 1, 2024): Thanks Emmanuel! This has been addressed in #349. Closing this issue out.

GiteaMirror commented

2026-04-19 12:01:51 -05:00

@profvjreddi commented on GitHub (Sep 2, 2024):

Thanks for the feedback, @emmanuel2406. I love it when community members find issues and help us all improve the material. So please keep it coming!

@profvjreddi commented on GitHub (Sep 2, 2024): Thanks for the feedback, @emmanuel2406. I love it when community members find issues and help us all improve the material. So please keep it coming!

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: github-starred/cs249r_book#3978