Fix shared parameter gradient scaling #283
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
gasteigerjo
changed the title
levineds
pushed a commit
that referenced
this pull request
Jul 11, 2024
* Fix shared parameter grad scaling * Fix shared_parameters in GemNet * Warn once if a shared parameter has no gradient Co-authored-by: Abhishek Das <das.abhshk@gmail.com>
beomseok-kang
pushed a commit
to beomseok-kang/fairchem
that referenced
this pull request
Jan 27, 2025
* Fix shared parameter grad scaling * Fix shared_parameters in GemNet * Warn once if a shared parameter has no gradient Co-authored-by: Abhishek Das <das.abhshk@gmail.com> Former-commit-id: 60bcc6178ea31c183f8533acc34919866a7511c7
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Outdated suggestions cannot be applied.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.
Gradient scaling for shared parameters didn't work previous since it used
trainer.modelinstead oftrainer.model.module. Also, GemNet-T passed modules instead of parameters. This fixes these two issues.Additionally, we now also show a warning if a shared parameter has no gradient, which usually means that it doesn't point to a PyTorch parameter.