Rework the Kaiming initialization. #573
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
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Outdated suggestions cannot be applied.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.
Add support for the following in Kaiming initialization:
FanInorFanOutrather than always defaulting toFanIn.This is a breaking change so will only be included in the next major release. It also changes the default behavior when using Kaiming uniform (which is the default for most linear/conv layers): the gain is multiplied by sqrt(2) because of the non-linearity part and there is also a bugfix included here that multiplies the bounds by sqrt(3). Overall the initial values will be scaled up by a factor of sqrt(6).