Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -13,6 +13,13 @@ library_name: capacitor_diffae
|
|
| 13 |
|
| 14 |
# data-archetype/semdisdiffae
|
| 15 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
**SemDisDiffAE** (**Sem**antically **Dis**entangled **Diff**usion **A**uto**E**ncoder)
|
| 17 |
— a fast image tokenizer with semantically structured 128-channel latents, built
|
| 18 |
on FCDM (Fully Convolutional Diffusion Model) blocks with a VP-parameterized
|
|
|
|
| 13 |
|
| 14 |
# data-archetype/semdisdiffae
|
| 15 |
|
| 16 |
+
### Version History
|
| 17 |
+
|
| 18 |
+
| Date | Change |
|
| 19 |
+
|------|--------|
|
| 20 |
+
| 2026-04-07 | Include latent running stats for whitening/dewhitening in exported weights |
|
| 21 |
+
| 2026-04-06 | Initial release |
|
| 22 |
+
|
| 23 |
**SemDisDiffAE** (**Sem**antically **Dis**entangled **Diff**usion **A**uto**E**ncoder)
|
| 24 |
— a fast image tokenizer with semantically structured 128-channel latents, built
|
| 25 |
on FCDM (Fully Convolutional Diffusion Model) blocks with a VP-parameterized
|