Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -77,7 +77,7 @@ Multi-layer tapping lets early layers contribute low-level cues (color, sharpnes
|
|
| 77 |
## Training
|
| 78 |
|
| 79 |
- **Base model**: google/siglip2-so400m-patch16-naflex (pretrained weights)
|
| 80 |
-
- **Training data**: ~134k images with aesthetic scores annotated by Gemini Flash, spanning the following categories:
|
| 81 |
- **Curated photography** — 21 thematic categories: landscape, portrait, wildlife, macro, architecture, fashion, food, street, product, astrophotography, and more
|
| 82 |
- **General photography** — community-shared and web-sourced real photos (both high and low quality)
|
| 83 |
- **AI-generated images** — outputs from multiple text-to-image diffusion model families
|
|
|
|
| 77 |
## Training
|
| 78 |
|
| 79 |
- **Base model**: google/siglip2-so400m-patch16-naflex (pretrained weights)
|
| 80 |
+
- **Training data**: ~134k images hand-curated from diverse web sources, with aesthetic scores annotated by Gemini Flash, spanning the following categories:
|
| 81 |
- **Curated photography** — 21 thematic categories: landscape, portrait, wildlife, macro, architecture, fashion, food, street, product, astrophotography, and more
|
| 82 |
- **General photography** — community-shared and web-sourced real photos (both high and low quality)
|
| 83 |
- **AI-generated images** — outputs from multiple text-to-image diffusion model families
|