fix: Handle missing and unexpected keys during LLMEncoder state dict load #3841
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
geoffreyangus
approved these changes
Dec 20, 2023
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Outdated suggestions cannot be applied.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.
When using an adapter, the LLMEncoder uses custom
state_dict()ansload_state_dict()logic to only save/load adapter parameters, however loading would fail when using strict mode due tostate_dict(missing_keys)unexpected_keys)None of these keys are missing or unexpected under the custom logic, though, so the lists of missing and unexpected keys must be updated to reflect our expected state. This PR adds a post-load hook that updates the
missing_keysandunexpected_keyslists to reflect our custom logic, and it re-enables a unit test that checks loading adapter state.