add strict arg to load_checkpoint_and_dispatch #2641
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
|
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. |
muellerzr
approved these changes
Apr 9, 2024
There was a problem hiding this comment.
Thanks I agree its worth doing a patch, we also should probably run some diffusers stuff in our integration tests that we can run as part of the CI to check for breaking. @sayakpaul do you know an easy subset/file of tests we can run just for sanity testing on main? For transformers we run the test_trainer.py file
|
Do the tests and our |
|
@sayakpaul if they broke due to what this PR is fixing, yes :) |
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Outdated suggestions cannot be applied.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.
What does this PR do ?
This PR adds the strict arg in
load_checkpoint_and_dispatch. We need this since this PR introduced a breaking change in the way to error out when the keys in the checkpoint state_dict do not match the keys of the model's state_dict.Fixes #2640 (comment) cc @yiyixuxu
maybe worth doing a patch @muellerzr ? There are other issues in diffusers so let's wait a bit if we indeed do a patch.