Fix hardcoded action dim in pi0 pytorch model #855
+3
−3
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
abhaybd
requested review from
Michael-Equi,
jimmyt857 and
uzhilinsky
as code owners
Author
|
Fixes #714 |
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Outdated suggestions cannot be applied.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.
As seen here, the jax version of the pi0 modeling code correctly uses
config.action_dimfor the linear layer shapes. However, the corresponding pytorch code hardcodes this as 32.This works fine for pi0/pi05 since they use an action dim of 32, but if users want to e.g. train a paligemma model with a different action dimension the pytorch code will error. This PR fixes this bug and uses the action dimension from the config.