Merge branch 'main' of https://github.com/edenartlab/sd-lora-trainer into main

2024-08-23 11:05:53 +02:00
parent ae4f6a0ee1 622303c48d
commit 92e3fe8599
1 changed files with 1 additions and 1 deletions
@@ -70,7 +70,7 @@ Bugs:

 Bigger improvements:
 - integrate Flux / SD3
- Add multi-token training
+- Add multi-concept training (multiple things represented by multiple tokens, trained into a single LoRa)
 - add stronger token regularization (eg CelebBasis spanning basis)
 - implement perfusion ideas (key locking with superclass): https://research.nvidia.com/labs/par/Perfusion/
 - implement prompt-aligned: https://prompt-aligned.github.io/