Hugging Face's logo

Disty0
/
sd3_vae_anime_ft

Text-to-Image
Diffusers
Safetensors
Model card Files
xet
 Community
Disty0 commited on
Commit
fdd4e69
·
verified ·
1 Parent(s): b4d3fa0

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -8,7 +8,7 @@ pipeline_tag: text-to-image
8
  library_name: diffusers
9
  ---
10
 
11
- This VAE was finetuned on PNG only anime illustrations for 512 steps.
12
  Used fp32 weights + fp16 mixed precision with learning rate 4e-6 and effective batch size of 16.
13
 
14
  This training was to test my VAE decoder training code and 512 step model turned out to be better than i expected and fixes the color shifting issues of the original SD3 VAE pretty well.
 
8
  library_name: diffusers
9
  ---
10
 
11
+ This VAE was finetuned on PNG only anime illustrations for 512 steps with MAE Loss.
12
  Used fp32 weights + fp16 mixed precision with learning rate 4e-6 and effective batch size of 16.
13
 
14
  This training was to test my VAE decoder training code and 512 step model turned out to be better than i expected and fixes the color shifting issues of the original SD3 VAE pretty well.