dima806's picture
Update README.md
ccbb363 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     0.6281    0.9054    0.7417       222
          02     0.3854    0.5000    0.4353       222
          03     0.7121    0.2117    0.3264       222
          04     0.8462    0.0495    0.0936       222
          05     0.4567    0.8559    0.5956       222
       06-07     0.6808    0.6502    0.6651       223
       08-09     0.6111    0.6413    0.6258       223
       10-12     0.5833    0.5676    0.5753       222
       13-15     0.6577    0.7668    0.7081       223
       16-20     0.5935    0.6577    0.6239       222
       21-25     0.5514    0.6009    0.5751       223
       26-30     0.4274    0.4640    0.4449       222
       31-35     0.5376    0.2252    0.3175       222
       36-40     0.4370    0.2646    0.3296       223
       41-45     0.3746    0.5830    0.4561       223
       46-50     0.3828    0.4978    0.4327       223
       51-55     0.4061    0.4189    0.4124       222
       56-60     0.3791    0.2613    0.3093       222
       61-65     0.4706    0.1794    0.2597       223
       66-70     0.3811    0.4550    0.4148       222
       71-80     0.5468    0.6577    0.5971       222
       81-90     0.7985    0.9596    0.8717       223
         90+     0.9781    1.0000    0.9889       223

    accuracy                         0.5381      5116
   macro avg     0.5577    0.5380    0.5131      5116
weighted avg     0.5577    0.5381    0.5132      5116