metadata
license: apache-2.0
metrics:
- accuracy
- f1
base_model:
- google/vit-base-patch16-224-in21k
pipeline_tag: image-classification
library_name: transformers
Returns a hand gesture type for smart TV given an image.
See https://www.kaggle.com/code/dima806/smart-tv-hand-gestures-image-detection-vit for details.
Classification report:
precision recall f1-score support
Down 1.0000 1.0000 1.0000 1350
Left Swipe 1.0000 1.0000 1.0000 1350
Right Swipe 1.0000 1.0000 1.0000 1350
Stop 0.9912 1.0000 0.9956 1350
Stop Gesture 1.0000 1.0000 1.0000 1350
Swipe 1.0000 0.9948 0.9974 1350
Thumbs Down 1.0000 1.0000 1.0000 1350
Thumbs Up 1.0000 1.0000 1.0000 1350
Up 1.0000 0.9963 0.9981 1350
accuracy 0.9990 12150
macro avg 0.9990 0.9990 0.9990 12150
weighted avg 0.9990 0.9990 0.9990 12150