Update README.md
Browse files
README.md
CHANGED
|
@@ -2,6 +2,7 @@
|
|
| 2 |
tags:
|
| 3 |
- image-classification
|
| 4 |
- birder
|
|
|
|
| 5 |
library_name: birder
|
| 6 |
license: apache-2.0
|
| 7 |
---
|
|
@@ -19,6 +20,8 @@ The species list is derived from data available at <https://www.israbirding.com/
|
|
| 19 |
- Params (M): 86.1
|
| 20 |
- Input image size: 256 x 256
|
| 21 |
- **Dataset:** il-common (371 classes)
|
|
|
|
|
|
|
| 22 |
|
| 23 |
- **Papers:**
|
| 24 |
- An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale: <https://arxiv.org/abs/2010.11929>
|
|
|
|
| 2 |
tags:
|
| 3 |
- image-classification
|
| 4 |
- birder
|
| 5 |
+
- pytorch
|
| 6 |
library_name: birder
|
| 7 |
license: apache-2.0
|
| 8 |
---
|
|
|
|
| 20 |
- Params (M): 86.1
|
| 21 |
- Input image size: 256 x 256
|
| 22 |
- **Dataset:** il-common (371 classes)
|
| 23 |
+
- Intermediate training involved ~5000 species from asia, europe and eastern africa
|
| 24 |
+
- Epoch 200 checkpoint of [vitreg4_b16_mim](https://huggingface.co/birder-project/vitreg4_b16_mim)
|
| 25 |
|
| 26 |
- **Papers:**
|
| 27 |
- An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale: <https://arxiv.org/abs/2010.11929>
|