README: drop venue, note tested transformers versions
Browse files
README.md
CHANGED
|
@@ -10,8 +10,7 @@ pipeline_tag: image-feature-extraction
|
|
| 10 |
|
| 11 |
# PaGE ViT-S Distill
|
| 12 |
|
| 13 |
-
Small / fast distilled student. Part of the [PaGE](https://huggingface.co/Octopus1/PaGE) gaze target estimation family
|
| 14 |
-
(ICLR 2026).
|
| 15 |
|
| 16 |
- **Backbone:** DINOv3 ViT-S (derivative of DINOv3, full-parameter trained)
|
| 17 |
- **Params:** ~25M
|
|
@@ -33,6 +32,8 @@ training of DINOv3. The model code (`modeling_page.py`) is loaded automatically
|
|
| 33 |
pip install torch torchvision timm "transformers>=4.56" safetensors pillow
|
| 34 |
```
|
| 35 |
|
|
|
|
|
|
|
| 36 |
## Usage
|
| 37 |
|
| 38 |
```python
|
|
|
|
| 10 |
|
| 11 |
# PaGE ViT-S Distill
|
| 12 |
|
| 13 |
+
Small / fast distilled student. Part of the [PaGE](https://huggingface.co/Octopus1/PaGE) gaze target estimation family.
|
|
|
|
| 14 |
|
| 15 |
- **Backbone:** DINOv3 ViT-S (derivative of DINOv3, full-parameter trained)
|
| 16 |
- **Params:** ~25M
|
|
|
|
| 32 |
pip install torch torchvision timm "transformers>=4.56" safetensors pillow
|
| 33 |
```
|
| 34 |
|
| 35 |
+
Tested with `transformers` 4.56.0 and 5.6.2.
|
| 36 |
+
|
| 37 |
## Usage
|
| 38 |
|
| 39 |
```python
|