Deep learning 관련 논문을 볼 때 top-1 와 top-5 error 라는 용어들이 나온다.
top-1와 top-5 error는 이미지 분류성능을 평가하기 위한 것 들이다.
Top-1 accuracy is the conventional accuracy, which means that the model answer (the one with the highest probability) must be exactly the expected answer.
Top-5 accuracy means that any of your model that gives 5 highest probability answers that must match the expected answer.
A picture of a cat is shown, and these are the outputs of your neural network:
예:
- Tiger: 0.4
- Dog: 0.3
- Cat: 0.1
- Lynx: 0.09
- Lion: 0.08
- Bird: 0.02
- Bear: 0.01
In the above-mentioned probabilities:
Using top-1 accuracy, you will count this output as wrong, because it predicted a tiger.
Using top-5 accuracy, you count this output as correct, because the cat is among the top-5 guesses.
top-5 accuracy : 이미지 50개를 분류하는데 , 앞의 5개의 class 를 최대 확률 의 분류를 찾은 다음 , 정확한 label이
앞의 5개에 있는지 확인한다. 있으면 분류 성공했다.
--------------------------------------------------------------------------------------------------------------------------------
예를 들어 :
정답 Ground Truth : cat
예측한 결과 : prediction : 0.4 Tiger
예:
- Tiger: 0.4
- Dog: 0.3
- Cat: 0.1
- Lynx: 0.09
- Lion: 0.08
- Bird: 0.02
- Bear: 0.01
top-1 은 False
top-5 은 Tiger, Dog, Cat, Lynx, Lion 중에서 Cat이 있어서 True이다.
<참고 자료 >:
https://intellipaat.com/community/6715/evaluation-calculate-top-n-accuracy-top-1-and-top-5