Detects cat breed (from the list of 48 common breeds) with about 77% accuracy based on image.
See https://www.kaggle.com/code/dima806/cat-breed-image-detection-vit for more details.
Classification report:
precision recall f1-score support
Abyssinian 0.9771 0.9628 0.9699 2258
American Bobtail 0.3837 0.4956 0.4325 2258
American Curl 0.9434 0.9376 0.9405 2258
American Shorthair 0.2603 0.0558 0.0919 2258
Applehead Siamese 0.9825 0.9920 0.9872 2258
Balinese 0.9191 0.9606 0.9394 2258
Bengal 0.7666 0.7724 0.7695 2258
Birman 0.8913 0.9371 0.9136 2258
Bombay 0.6471 0.8893 0.7491 2258
British Shorthair 0.7879 0.8720 0.8278 2258
Burmese 0.9124 0.9544 0.9329 2258
Calico 0.6144 0.6492 0.6314 2258
Cornish Rex 0.9960 0.9849 0.9904 2258
Devon Rex 0.9856 1.0000 0.9927 2258
Dilute Calico 0.6094 0.5540 0.5804 2258
Dilute Tortoiseshell 0.5747 0.7037 0.6327 2258
Domestic Long Hair 0.4103 0.3392 0.3714 2258
Domestic Medium Hair 0.2805 0.1120 0.1601 2258
Domestic Short Hair 0.9490 0.9810 0.9647 2258
Egyptian Mau 0.8900 0.9243 0.9068 2258
Exotic Shorthair 0.8720 0.8233 0.8469 2258
Extra-Toes Cat - Hemingway Polydactyl 0.3899 0.2055 0.2691 2258
Havana 0.9811 0.9889 0.9850 2258
Himalayan 0.8022 0.7201 0.7589 2258
Japanese Bobtail 0.9799 0.9725 0.9762 2258
Maine Coon 0.5321 0.6833 0.5983 2258
Manx 0.3942 0.1147 0.1777 2258
Munchkin 0.9325 0.9544 0.9433 2258
Nebelung 0.9639 0.9938 0.9786 2258
Norwegian Forest 0.8280 0.8463 0.8371 2258
Oriental Short Hair 0.8772 0.8096 0.8420 2258
Persian 0.8302 0.8012 0.8154 2258
Ragamuffin 0.9641 0.9872 0.9755 2258
Ragdoll 0.6866 0.6501 0.6679 2258
Russian Blue 0.8215 0.9256 0.8705 2258
Scottish Fold 0.9728 0.9181 0.9446 2258
Siamese 0.7649 0.7551 0.7600 2258
Siberian 0.9636 0.9619 0.9628 2258
Snowshoe 0.8389 0.8140 0.8263 2258
Sphynx 0.9942 0.9903 0.9922 2258
Tabby 0.3533 0.3441 0.3487 2258
Tiger 0.4071 0.6076 0.4876 2258
Tonkinese 0.9613 0.9464 0.9538 2258
Torbie 0.5544 0.6568 0.6013 2258
Tortoiseshell 0.6667 0.8122 0.7323 2258
Turkish Angora 0.7328 0.8308 0.7787 2258
Turkish Van 0.7658 0.9225 0.8369 2258
Tuxedo 0.6491 0.8357 0.7307 2258
accuracy 0.7698 108384
macro avg 0.7554 0.7698 0.7559 108384
weighted avg 0.7554 0.7698 0.7559 108384
- Downloads last month
- 99
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Model tree for dima806/cat_breed_image_detection
Base model
google/vit-base-patch16-224-in21k