---
license: apache-2.0
metrics:
- accuracy
- f1
base_model:
- google/vit-base-patch16-224-in21k
---
Returns horse breed given an image with about 91% accuracy.

See https://www.kaggle.com/code/dima806/horse-breed-image-detection-vit for more details.

```
Classification report:

                      precision    recall  f1-score   support

            Friesian     0.8889    1.0000    0.9412        24
             Arabian     0.8571    0.9600    0.9057        25
           Percheron     1.0000    0.6400    0.7805        25
       Orlov Trotter     0.7931    0.9200    0.8519        25
          Akhal-Teke     1.0000    0.9200    0.9583        25
Vladimir Heavy Draft     0.9200    0.9583    0.9388        24
           Appaloosa     1.0000    1.0000    1.0000        25

            accuracy                         0.9133       173
           macro avg     0.9227    0.9140    0.9109       173
        weighted avg     0.9229    0.9133    0.9106       173
```