fix sd 2.0/2.1 lora fine-tuning text encoder params mismatch issue by updating model_util.py#636
Open
miracleyoo wants to merge 1 commit intoAkegarasu:mainfrom
Open
fix sd 2.0/2.1 lora fine-tuning text encoder params mismatch issue by updating model_util.py#636miracleyoo wants to merge 1 commit intoAkegarasu:mainfrom
miracleyoo wants to merge 1 commit intoAkegarasu:mainfrom
Conversation
Fix sd 2.0/2.1 lora fine-tuning text encoder params mismatch issue. With the original text encoder and its corresponding parser, stable diffusion 2.0/2.1/2.1-unclip all cannot be properly loaded due to the text encoder transformer version difference and checkpoints' state_dict key name difference. After this fix, all these three versions are tested to work well when executing lora fine tuning. Release it here as SD 2.x version has a 1024 text embedding, which is unique and a must for many related projects, like ImageBind-related works.
miracleyoo
changed the title
Author
|
Passed the Lora fine-tuning trial using the following SD base models: |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Fix sd 2.0/2.1 lora fine-tuning text encoder params mismatch issue. With the original text encoder and its corresponding parser, stable diffusion 2.0/2.1/2.1-unclip all cannot be properly loaded due to the text encoder transformer version difference and checkpoints' state_dict key name difference. After this fix, all these three versions are tested to work well when executing lora fine tuning. Release it here as SD 2.x version has a 1024 text embedding, which is unique and a must for many related projects, like ImageBind-related works.