Fix load_checkpoint_in_model behavior when unexpected keys are in the checkpoint
#2588
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
fxmarty
commented
Mar 26, 2024
| buffer_names = [name for name, _ in model.named_buffers()] | ||
| for checkpoint_file in checkpoint_files: | ||
| checkpoint = load_state_dict(checkpoint_file, device_map=device_map) | ||
| loaded_checkpoint = load_state_dict(checkpoint_file, device_map=device_map) |
There was a problem hiding this comment.
This variable is renamed because it overrides the function input checkpoint needed later.
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
SunMarc
approved these changes
Mar 26, 2024
This was referenced Apr 9, 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Outdated suggestions cannot be applied.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.
Up to now,
load_checkpoint_in_modelwould behave differently when the checkpoint to be loaded has unexpected keys compared to the model state_dict keys.With
device_map=Noneinput, we are currently not erroring out, while withdevice_mapnot None, we are erroring out.This PR aligns the behavior of both, defaulting to
strict=False, and logs the unloaded keys.