dima806's picture
Update README.md
a44f3ec verified
|
raw
history blame
787 Bytes
metadata
license: apache-2.0
datasets:
  - nateraw/fairface
metrics:
  - accuracy
  - f1
base_model:
  - google/vit-base-patch16-224-in21k
pipeline_tag: image-classification

Detects gender with about 88% accuracy based on an image.

See https://www.kaggle.com/code/dima806/gender-image-classification-vit for details.

image/png

Classification report:

              precision    recall  f1-score   support

      Female     0.8563    0.8938    0.8746       480
        Male     0.8978    0.8615    0.8793       520

    accuracy                         0.8770      1000
   macro avg     0.8770    0.8776    0.8770      1000
weighted avg     0.8779    0.8770    0.8770      1000