dima806's picture
Update README.md
68e1908 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     0.6774    0.9011    0.7734       445
          02     0.4082    0.5843    0.4806       445
          03     0.5972    0.1933    0.2920       445
          04     0.8293    0.1528    0.2581       445
          05     0.5422    0.9414    0.6881       444
       06-07     0.7629    0.6292    0.6897       445
       08-09     0.6201    0.7483    0.6782       445
       10-12     0.6735    0.6674    0.6704       445
       13-15     0.7411    0.7461    0.7436       445
       16-20     0.6377    0.7079    0.6709       445
       21-25     0.6572    0.6247    0.6406       445
       26-30     0.4299    0.5371    0.4775       445
       31-35     0.5196    0.2382    0.3267       445
       36-40     0.4755    0.3708    0.4167       445
       41-45     0.4634    0.6966    0.5566       445
       46-50     0.4309    0.5270    0.4742       444
       51-55     0.4732    0.4562    0.4645       445
       56-60     0.5097    0.3536    0.4176       444
       61-65     0.6835    0.2427    0.3582       445
       66-70     0.4852    0.7730    0.5962       445
       71-80     0.7740    0.7079    0.7394       445
       81-90     0.8516    0.9955    0.9180       444
         90+     0.9978    1.0000    0.9989       445

    accuracy                         0.5997     10231
   macro avg     0.6192    0.5998    0.5796     10231
weighted avg     0.6192    0.5997    0.5795     10231