Update README.md
Browse files
README.md
CHANGED
|
@@ -4,4 +4,22 @@ tags:
|
|
| 4 |
- 'vit '
|
| 5 |
- image classification
|
| 6 |
- ggml
|
| 7 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4 |
- 'vit '
|
| 5 |
- image classification
|
| 6 |
- ggml
|
| 7 |
+
---
|
| 8 |
+
|
| 9 |
+
# Vision Transformer (ViT) models for image classification converted to ggml format
|
| 10 |
+
|
| 11 |
+
[Available models](https://github.com/staghado/vit.cpp)
|
| 12 |
+
|
| 13 |
+
| Model | Disk | Mem | SHA |
|
| 14 |
+
| --- | --- | --- | --- |
|
| 15 |
+
| tiny | 12 MB | ~20 MB | `25ce65ff60e08a1a5b486685b533d79718e74c0f` |
|
| 16 |
+
| small | 45 MB | ~52 MB | `7a9f85340bd1a3dcd4275f46d5ee1db66649700e` |
|
| 17 |
+
| base | 174 MB | ~179 MB | `a10d29628977fe27691edf55b7238f899b8c02eb` |
|
| 18 |
+
| large | 610 MB | ~597 MB | `5f27087930f21987050188f9dc9eea75ac607214` |
|
| 19 |
+
|
| 20 |
+
The models are pre-trained on ImageNet21k then finetuned on ImageNet1k
|
| 21 |
+
with a patch size of 16 and an image size of 224.
|
| 22 |
+
|
| 23 |
+
For more information, visit:
|
| 24 |
+
|
| 25 |
+
https://github.com/staghado/vit.cpp
|