zero out channels that aren't normazlied#1156
Merged
Merged
Conversation
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This PR resolves an issue where the transformer model behaved inconsistently depending on how extra input channels were padded. Specifically, CP4 failed to recognize some cells that CP3 detected, due to differences in input normalization when additional channels were present.
Previously, extra channels (e.g., the 2nd and 3rd channels in grayscale images or the 3rd in RGB) were padded with ones after normalization.
This change ensures that all extra channels are consistently padded with zeros, preventing the transformer from interpreting the padding as meaningful signal.