metadata
license: apache-2.0
metrics:
- accuracy
- f1
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 0.7593 0.8301 0.7931 612
A400M 0.8385 0.6618 0.7397 612
AG600 0.9045 0.9444 0.9241 612
AV8B 0.4674 0.8676 0.6076 612
B1 0.8211 0.8023 0.8116 612
B2 0.8460 0.9428 0.8918 612
B52 0.9259 0.6127 0.7375 612
Be200 0.8349 0.8758 0.8549 612
C130 0.8843 0.3873 0.5386 612
C17 0.7860 0.3840 0.5159 612
C2 0.7546 0.8693 0.8079 612
C5 0.6237 0.7042 0.6616 612
E2 0.8295 0.9542 0.8875 612
E7 0.7003 0.9853 0.8187 612
EF2000 0.7548 0.1912 0.3051 612
F117 0.7472 0.9657 0.8425 612
F14 0.6215 0.5474 0.5821 612
F15 0.4713 0.1209 0.1925 612
F16 0.5069 0.2402 0.3259 612
F18 0.6530 0.4673 0.5448 612
F22 0.5633 0.6324 0.5958 612
F35 0.5055 0.4510 0.4767 612
F4 0.6870 0.2582 0.3753 612
H6 0.8816 0.9003 0.8909 612
J10 0.5703 0.4902 0.5272 612
J20 0.7433 0.4967 0.5955 612
JAS39 0.4681 0.4314 0.4490 612
JF17 0.3657 0.9297 0.5249 612
KC135 0.6563 0.8611 0.7449 612
MQ9 0.8416 0.9118 0.8753 612
Mig31 0.7485 0.5931 0.6618 612
Mirage2000 0.7489 0.2729 0.4000 612
P3 0.7738 0.9167 0.8392 612
RQ4 0.8146 0.9118 0.8604 612
Rafale 0.3744 0.4771 0.4195 612
SR71 0.8404 0.9118 0.8746 612
Su24 0.6591 0.4739 0.5513 612
Su25 0.7762 0.7255 0.7500 612
Su34 0.7154 0.8709 0.7856 612
Su57 0.5959 0.8578 0.7033 612
Tornado 0.5558 0.4886 0.5200 612
Tu160 0.7494 0.9575 0.8407 612
Tu22M 0.7414 0.8431 0.7890 612
Tu95 0.8777 0.8791 0.8784 612
U2 0.8855 0.7075 0.7866 612
US2 0.7507 0.8807 0.8105 612
V22 0.8502 0.9183 0.8830 612
Vulcan 0.8696 0.8497 0.8595 612
XB70 0.8252 0.9951 0.9022 612
YF23 0.7581 0.9984 0.8618 612
accuracy 0.7049 30600
macro avg 0.7185 0.7049 0.6883 30600
weighted avg 0.7185 0.7049 0.6883 30600