Force prediction normalization fix #234
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
This corrects the rescaling of the gradient of normalized energy, as the mean was erroneously added to the force.
laserkelvin
added
bug
melo-gonzo
approved these changes
Jun 4, 2024
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Outdated suggestions cannot be applied.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.
This PR applies two changes:
ForceRegressionTask.predict,denormfromenergywas erroneously applied to forces, which was the wrong behavior. This PR fixes this by replacing this with just a rescaling based on the energy standard deviation.ForceRegressionTasknow overrides_make_normalizers, adding a check for the case when energy normalization is specified and not force. In this scenario, we automatically add the correct force normalizer based on copying the energy standard deviation, which should ensure the correct behavior.