Bugfix for the gating mechanism + update imports#1
Open
tijsmaas wants to merge 2 commits intolaiguokun:masterfrom
Open
Bugfix for the gating mechanism + update imports#1tijsmaas wants to merge 2 commits intolaiguokun:masterfrom
tijsmaas wants to merge 2 commits intolaiguokun:masterfrom
Conversation
The gate was initialized as x*tanh(x) due to a typo. This should have been sigm(x)*tanh(x). My commit also includes missing imports.
This file contains hidden or 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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
I came across a typo in the implementation of wavenet and stochastic wavenet gating mechanism. Originally wavenet uses a GTU:
tanh(x)*sigm(x), however in the code the gating isx*tanh(x). Haven't tested it but likely this explains some of the instability encountered during training.A bugfix and fixes of missing imports (+updated pytorch definitions) are in this pull request. Either way, cool work.