cat-breed-60-classes

ferdifdi/cat-breed-60-classes

cat-breed-60-classes is a machine learning model from ferdifdi. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as mit.

Detects cat breed (from the list of 60 common breeds) with about 75% accuracy based on image.

Downloads · 30 days

30% of all-time downloads

All-time downloads

Public

Parameters

85.8M

343 MB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors343 MB · 100%

At a glance

License: mit
Model type: vit
Access: Public
Created: Dec 12, 2025
Updated: Dec 12, 2025
SHA: 86c96206

Base models

dima806/cat_breed_image_detection

Model card

From the Hugging Face model README

Detects cat breed (from the list of 60 common breeds) with about 75% accuracy based on image.

See https://github.com/ferdifdi/cat_breeds_classifier_extended_breeds for more details.

This is for the web app (BISCUTE: Breed Identification & Specific Care Utility Tool Engine): https://github.com/ferdifdi/BISCUTE

Classification Report

PER-BREED ACCURACY - ALL BREEDS

ORIGINAL BREEDS

Breed	Accuracy	Correct/Total
Abyssinian	94.33%	2130/2258
American Bobtail	38.80%	876/2258
American Curl	87.29%	1971/2258
American Shorthair	4.38%	99/2258
Applehead Siamese	95.44%	2155/2258
Balinese	92.69%	2093/2258
Bengal	78.17%	1765/2258
Birman	89.95%	2031/2258
Bombay	84.10%	1899/2258
British Shorthair	83.13%	1877/2258
Burmese	88.22%	1992/2258
Calico	62.98%	1422/2258
Cornish Rex	96.86%	2187/2258
Devon Rex	94.77%	2140/2258
Dilute Calico	56.42%	1274/2258
Dilute Tortoiseshell	67.05%	1514/2258
Domestic Long Hair	33.04%	746/2258
Domestic Medium Hair	12.49%	282/2258
Domestic Short Hair	97.65%	2205/2258
Egyptian Mau	89.15%	2013/2258
Exotic Shorthair	78.92%	1782/2258
Extra-Toes Cat - Hemingway Polydactyl	23.52%	531/2258
Havana	95.84%	2164/2258
Himalayan	73.56%	1661/2258
Japanese Bobtail	96.59%	2181/2258
Maine Coon	58.72%	1326/2258
Manx	9.03%	204/2258
Munchkin	91.45%	2065/2258
Nebelung	97.39%	2199/2258
Norwegian Forest	75.29%	1700/2258
Oriental Short Hair	76.13%	1719/2258
Persian	73.43%	1658/2258
Ragamuffin	99.42%	2245/2258
Ragdoll	51.77%	1169/2258
Russian Blue	89.90%	2030/2258
Scottish Fold	89.42%	2019/2258
Siamese	65.01%	1468/2258
Siberian	93.67%	2115/2258
Snowshoe	80.20%	1811/2258
Sphynx	98.45%	2223/2258
Tabby	40.48%	914/2258
Tiger	46.63%	1053/2258
Tonkinese	93.89%	2120/2258
Torbie	61.38%	1386/2258
Tortoiseshell	77.15%	1742/2258
Turkish Angora	77.50%	1750/2258
Turkish Van	84.99%	1919/2258
Tuxedo	79.81%	1802/2258

NEW BREEDS

Breed	Accuracy	Correct/Total
Chartreux	94.51%	2134/2258
Chausie	83.08%	1876/2258
Khao Manee	85.12%	1922/2258
Kurilian Bobtail	86.05%	1943/2258
LaPerm	64.57%	1458/2258
Neva Masquerade	91.90%	2075/2258
Ocicat	82.95%	1873/2258
Peterbald	76.97%	1738/2258
Savannah	77.41%	1748/2258
Selkirk Rex	88.09%	1989/2258
Serengeti	64.13%	1448/2258
Singapura	78.17%	1765/2258

SUMMARY STATISTICS

ORIGINAL BREEDS (48 breeds)

Metric	Value
Average accuracy	73.47%
Min accuracy	4.38%
Max accuracy	99.42%

NEW BREEDS (12 breeds)

Metric	Value
Average accuracy	81.08%
Min accuracy	64.13%
Max accuracy	94.51%

OVERALL (All 60 breeds)

Metric	Value
Average accuracy	74.99%
Min accuracy	4.38%
Max accuracy	99.42%

BREEDS WITH POOR PERFORMANCE (< 20% accuracy)

Breed	Accuracy	Type	Correct/Total
American Shorthair	4.38%	ORIGINAL	99/2258
Domestic Medium Hair	12.49%	ORIGINAL	282/2258
Manx	9.03%	ORIGINAL	204/2258

Total: 3 breeds (3 original, 0 new)

Type: vit
License: mit
Created: Dec 12, 2025
Updated: Dec 12, 2025