Apply LoRas to refiner #98
Open
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
The stage for applying LoRas currently only applies selected LoRas to the base model. The more steps run using a refiner model, the less the effects of the selected LoRas can be seen in the output image because the refiner model never has the LoRas applied to it. This change makes it so that, when a refiner model is selected, the LoRas are applied to that along with the base model.
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Outdated suggestions cannot be applied.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.
The stage for applying LoRas currently only applies selected LoRas to the base model. The more steps run using a refiner model, the less the effects of the selected LoRas can be seen in the output image because the refiner model never has the LoRas applied to it. This change makes it so that, when a refiner model is selected, the LoRas are applied to that along with the base model.
If you would prefer making this an option the user chooses, I am open to having this code changed to allow for that.