This repository has been archived by the owner on Oct 22, 2023. It is now read-only.
Repeat concept, and separate bucket concept options #111
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
…component freely, because that component is now being randomized ~10 times faster than for the base distribution https://www.crosslabs.org/blog/diffusion-with-offset-noise
This option puts selected concepts in their own buckets so they do not get batched together with the rest of the images of the same resolution. This should prevent some bleeding over between concepts.
Buckzor
changed the title
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.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.
Global repeat is useful, but when training multiple concepts it's a lot better to have the ability to repeat certain concepts to increase their representation in the overall dataset.
This PR adds two new concept options:
1: Repeat concept:
Global repeat is useful, but when training multiple concepts it's a lot better to have the ability to repeat certain concepts to increase their representation in the overall dataset.
2: Separate buckets:
Places the selected concepts in their own separate buckets, so they do not get mixed in with the rest of the images of the same resolution when training with higher batch sizes. This might help prevent some bleeding between styles/subjects.