fix unidiffuser

This commit is contained in:
Vladimir Mandic
2024-01-27 08:56:45 -05:00
parent e6a1946e06
commit ffee8d65af
2 changed files with 2 additions and 2 deletions
+1 -1
View File
@@ -129,7 +129,7 @@ As of this release, default backend is set to **diffusers** as its more feature
- greatly improves temporal consistency of generated outputs
- all options are available in animateddiff script
- [SalesForce BlipDiffusion](https://huggingface.co/docs/diffusers/api/pipelines/blip_diffusion)
- model can be used to replace anything in an image
- model can be used to place subject in a different context
- requires input image
- last word in prompt and negative prompt will be used as source and target subjects
- sampler must be set to default before loading the model
@@ -154,7 +154,7 @@ class EmbeddingDatabase:
tokenizer_2 = getattr(pipe, 'tokenizer_2', None)
clip_l = getattr(pipe, 'text_encoder', None) # clip_l
clip_g = getattr(pipe, 'text_encoder_2', None) # clip_g
if clip_l is None and tokenizer is None:
if clip_l is None or tokenizer is None:
return 0
filenames = (