Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -36,7 +36,7 @@ The three graphs are meant to be driven by a DDIM sampling loop on the JavaScrip
|
|
| 36 |
|
| 37 |
Spatial dimensions are static at 512x512 (64x64 latent): the cross-attention positional embedding is a learned parameter tied to the training resolution.
|
| 38 |
|
| 39 |
-
Reference implementation, export script and sampler: [
|
| 40 |
|
| 41 |
## License and credit
|
| 42 |
|
|
|
|
| 36 |
|
| 37 |
Spatial dimensions are static at 512x512 (64x64 latent): the cross-attention positional embedding is a learned parameter tied to the training resolution.
|
| 38 |
|
| 39 |
+
Reference implementation, export script and sampler: [on-device-eraser demo app](https://github.com/NorbertKlockiewicz/on-device-eraser).
|
| 40 |
|
| 41 |
## License and credit
|
| 42 |
|