This repository has been archived by the owner on Nov 3, 2022. It is now read-only.
Add AdaMod optimizer to keras-contrib #532
Open
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
mffigueroa
changed the title
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.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.
- What I did
I copied the Adam optimizer from the main keras repo and modified it according to the Adamod paper.
- How I did it
Mainly I added the exponential averaging of the learning rate after the gradients are divided by the 2nd moment, using the new beta_3 parameter. This exponential average is then used as an upper bound for this adaptive learning rate. The 1st and 2nd moment bias corrections had to be separated out into 2 different statements because the 1st moment term is applied after this upper bounding according to the Adamod paper. See this paper for further details.
- How you can verify it
I added a unit test in the same fashion as the other unit tests in the optimizers directory.
This pull request fixes #531