Returns military aircraft type given cropped image.
See https://www.kaggle.com/code/dima806/military-aircraft-detection-vit for more details.
Classification report:
precision recall f1-score support
A10 0.6250 0.4966 0.5535 594
A400M 0.5721 0.4411 0.4981 594
AG600 0.6266 0.8249 0.7122 594
AV8B 0.4511 0.6212 0.5227 594
B1 0.6286 0.5926 0.6101 594
B2 0.8817 0.5774 0.6979 594
B52 0.7801 0.4360 0.5594 594
Be200 0.8311 0.6296 0.7165 594
C130 0.5257 0.2407 0.3303 594
C17 0.2842 0.0455 0.0784 594
C2 0.6514 0.5505 0.5967 594
C5 0.3356 0.2458 0.2838 594
E2 0.7451 0.7088 0.7265 594
E7 0.3257 0.9630 0.4868 594
EF2000 0.5333 0.0673 0.1196 594
F117 0.7260 0.8165 0.7686 594
F14 0.4187 0.3687 0.3921 594
F15 0.1818 0.0842 0.1151 594
F16 0.5082 0.0522 0.0947 594
F18 0.6618 0.2306 0.3421 594
F22 0.3601 0.4461 0.3985 594
F35 0.3929 0.0370 0.0677 594
F4 0.3051 0.0303 0.0551 594
H6 0.8991 0.5253 0.6631 594
J10 0.4393 0.4327 0.4360 594
J20 0.3782 0.3266 0.3505 594
JAS39 0.2936 0.2407 0.2646 594
JF17 0.1583 0.8838 0.2685 594
KC135 0.2764 0.7710 0.4069 594
MQ9 0.5835 0.8350 0.6870 594
Mig31 0.7051 0.3704 0.4857 594
Mirage2000 0.4085 0.1465 0.2156 594
P3 0.4364 0.5774 0.4971 594
RQ4 0.7155 0.5842 0.6432 594
Rafale 0.2446 0.0572 0.0928 594
SR71 0.6041 0.5421 0.5714 594
Su24 0.5887 0.3687 0.4534 594
Su25 0.6599 0.4377 0.5263 594
Su34 0.6296 0.7155 0.6698 594
Su57 0.7877 0.4310 0.5571 594
Tornado 0.5531 0.1667 0.2561 594
Tu160 0.5890 0.8300 0.6890 594
Tu95 0.6147 0.7761 0.6860 594
U2 0.8426 0.1532 0.2593 594
US2 0.7644 0.5135 0.6143 594
V22 0.7368 0.8013 0.7677 594
Vulcan 0.6860 0.4966 0.5762 594
XB70 0.5531 0.9125 0.6887 594
YF23 0.2503 0.9444 0.3958 594
accuracy 0.4765 29106
macro avg 0.5459 0.4765 0.4581 29106
weighted avg 0.5459 0.4765 0.4581 29106
- Downloads last month
- 137