Add model card for dit4sr-replication
Browse files
README.md
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: diffusers
|
| 3 |
+
tags:
|
| 4 |
+
- dit4sr
|
| 5 |
+
- super-resolution
|
| 6 |
+
- diffusion-transformer
|
| 7 |
+
base_model: stabilityai/stable-diffusion-3.5-medium
|
| 8 |
+
---
|
| 9 |
+
# dit4sr-replication
|
| 10 |
+
|
| 11 |
+
This repository contains the `checkpoint-150000` export from the local DiT4SR experiment `dit4sr-replication`.
|
| 12 |
+
|
| 13 |
+
## Included files
|
| 14 |
+
|
| 15 |
+
- `transformer/` contains the publishable model weights and config.
|
| 16 |
+
- `source_checkpoint.json` records the local source path and checkpoint name used for the upload.
|
| 17 |
+
|
| 18 |
+
## Loading
|
| 19 |
+
|
| 20 |
+
```python
|
| 21 |
+
from model_dit4sr.transformer_sd3 import SD3Transformer2DModel
|
| 22 |
+
|
| 23 |
+
model = SD3Transformer2DModel.from_pretrained("NisargUpadhyay/ImageSuperResolution-replication", subfolder="transformer")
|
| 24 |
+
```
|
| 25 |
+
|
| 26 |
+
You still need the rest of the DiT4SR codebase and the base SD3 assets described in this repository's README.
|