dima806's picture
Update README.md
e3fe613 verified
|
raw
history blame
1.68 kB
metadata
license: apache-2.0
metrics:
  - accuracy
  - f1

Returns age bin based on a facial image.

See https://www.kaggle.com/code/dima806/facial-age-years-detection-vit for more details.

Classification report:

              precision    recall  f1-score   support

          01     0.7415    0.8851    0.8070       444
          02     0.4386    0.6899    0.5362       445
          03     0.7039    0.2831    0.4038       445
          04     0.8193    0.1532    0.2581       444
          05     0.5747    0.9596    0.7189       445
       06-07     0.6718    0.5843    0.6250       445
       08-09     0.6199    0.8135    0.7036       445
       10-12     0.6840    0.6225    0.6518       445
       13-15     0.7991    0.7685    0.7835       445
       16-20     0.7058    0.7169    0.7113       445
       21-25     0.6793    0.6045    0.6397       445
       26-30     0.4053    0.5820    0.4779       445
       31-35     0.5048    0.2382    0.3237       445
       36-40     0.4748    0.3596    0.4092       445
       41-45     0.5177    0.7213    0.6028       445
       46-50     0.5058    0.5910    0.5451       445
       51-55     0.4960    0.5586    0.5254       444
       56-60     0.5102    0.3371    0.4060       445
       61-65     0.6412    0.1888    0.2917       445
       66-70     0.4871    0.8045    0.6068       445
       71-80     0.7928    0.6966    0.7416       445
       81-90     0.8445    0.9910    0.9119       444
         90+     0.9978    1.0000    0.9989       445

    accuracy                         0.6152     10231
   macro avg     0.6355    0.6152    0.5948     10231
weighted avg     0.6354    0.6152    0.5948     10231