Hugging Face's logo

Severian-Void
/
Starsector-Portraits

Text-to-Image
Diffusers
English
StableDiffusionPipeline
stable-diffusion
stable-diffusion-diffusers
Inference Endpoints
Model card Files Community
7
Severian-Void commited on
Commit
c9a6b2d
•
1 Parent(s): ec092d7

Update images and readme

Browse files
Files changed (5) hide show
  1. README.md +11 -5
  2. RandomPortraits.png → images/RandomPortraits.png +0 -0
  3. patherflightschool.png → images/patherflightschool.png +0 -0
  4. images/rc_v1_cfg_compare.png +0 -0
  5. images/rc_v1_step_compare.png +0 -0
README.md CHANGED
@@ -28,10 +28,16 @@ Even with the training I found it was still useful to include some other phrases
28
  **conte** if brush strokes is not doing enough or adding colorful artifacts to images this is another option
29
 
30
  # Sample images from the model:
31
- ![output Samples](https://huggingface.co/Severian-Void/Starsector-Portraits/resolve/main/patherflightschool.png)
32
- ![output Samples](https://huggingface.co/Severian-Void/Starsector-Portraits/resolve/main/RandomPortraits.png)
33
 
34
- # File Version
35
- All version were trained using a hand labeled image set of 556 images with black/white backgrounds as well as flipped version of the same images
36
 
37
- Any checkpoint version with pploss in the name were trained with prior preservation loss enabled and a regularization set of 8000 portrait images comprised of photos/paintings/digitalart of humans in modern day attire.
 
 
 
 
 
 
 
28
  **conte** if brush strokes is not doing enough or adding colorful artifacts to images this is another option
29
 
30
  # Sample images from the model:
31
+ ![output Samples](https://huggingface.co/Severian-Void/Starsector-Portraits/resolve/main/images/patherflightschool.png)
32
+ ![output Samples](https://huggingface.co/Severian-Void/Starsector-Portraits/resolve/main/images/RandomPortraits.png)
33
 
34
+ # File Versions and Comparison Details
35
+ All version are trained using a hand labeled image set of 556 images with black/white backgrounds as well as flipped version of the same images
36
 
37
+ Any checkpoint version with pploss in the name were trained with prior preservation loss enabled and a regularization set of 8000 portrait images comprised of photos/paintings/digitalart of humans in modern day attire.
38
+
39
+ CFG and SAMPLE STEP comparisons between rc_v1 model files (prompt: "a starsectorportrait of a person" sampler: euler_a)
40
+ ![output Samples](https://huggingface.co/Severian-Void/Starsector-Portraits/resolve/main/images/rc_v1_step_compare.png)
41
+ ![output Samples](https://huggingface.co/Severian-Void/Starsector-Portraits/resolve/main/images/rc_v1_cfg_compare.png)
42
+ Adaptability comparisons between rc_v1 model files (prompt: "a starsectorportrait of an alien cat" sampler: euler_a)
43
+ ![output Samples](https://huggingface.co/Severian-Void/Starsector-Portraits/resolve/main/images/rc_v1_adaptability.png)
RandomPortraits.png → images/RandomPortraits.png RENAMED
File without changes
patherflightschool.png → images/patherflightschool.png RENAMED
File without changes
images/rc_v1_cfg_compare.png ADDED
images/rc_v1_step_compare.png ADDED