Update README.md
Browse files
README.md
CHANGED
|
@@ -12,9 +12,10 @@ pipeline_tag: text-to-image
|
|
| 12 |
|
| 13 |
This is a latent unconditional diffusion model to demonstrate how U-Net training affects the generated images.
|
| 14 |
|
| 15 |
-
- Text Encoder is removed
|
| 16 |
- VAE is from [Mitsua Diffusion One](https://huggingface.co/Mitsua/mitsua-diffusion-one), Mitsua Open RAIL-M License, Training Data: Public Domain/CC0 + Licensed
|
| 17 |
- U-Net is trained from scratch using full version of [VRoid Image Dataset Lite](https://huggingface.co/datasets/Mitsua/vroid-image-dataset-lite) with some modifications.
|
|
|
|
| 18 |
- VRoid is a trademark or registered trademark of Pixiv inc. in Japan and other regions.
|
| 19 |
|
| 20 |
## Model variant
|
|
|
|
| 12 |
|
| 13 |
This is a latent unconditional diffusion model to demonstrate how U-Net training affects the generated images.
|
| 14 |
|
| 15 |
+
- Pretrained Text Encoder (OpenCLIP) is removed, but an empty text encoder is included for compatibility with `StableDiffusionPipeline`.
|
| 16 |
- VAE is from [Mitsua Diffusion One](https://huggingface.co/Mitsua/mitsua-diffusion-one), Mitsua Open RAIL-M License, Training Data: Public Domain/CC0 + Licensed
|
| 17 |
- U-Net is trained from scratch using full version of [VRoid Image Dataset Lite](https://huggingface.co/datasets/Mitsua/vroid-image-dataset-lite) with some modifications.
|
| 18 |
+
- The architecture of the U-Net model was modified to conform to unconditional image generation. Cross-attention blocks are replaced by self-attention blocks.
|
| 19 |
- VRoid is a trademark or registered trademark of Pixiv inc. in Japan and other regions.
|
| 20 |
|
| 21 |
## Model variant
|