Update NMBottleneck bn momentum & epsilon #166
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
KSGulin
requested review from
natuan,
markurtz and
horheynm
and removed request for
a team
anmarques
approved these changes
Feb 7, 2023
corey-nm
approved these changes
Feb 7, 2023
KSGulin
added a commit
that referenced
this pull request
Feb 7, 2023
KSGulin
added a commit
that referenced
this pull request
Feb 7, 2023
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
When replacing the Bottleneck layers with our own custom, quantizable layers, the momentum and epsilon for batch norm weren't getting copied over properly, resulting in small differences in the output. With this fix, the outputs are identical
Test plan
Manual self-distillation runs without training, verifying the distillation loss is 0