metadata
license: apache-2.0
metrics:
- accuracy
- f1
pipeline_tag: image-classification
library_name: transformers
base_model:
- google/vit-base-patch16-224-in21k
Returns the orange fruit common disease (melanose or citrus canker) with about 98% accuracy based on an image.
See https://www.kaggle.com/code/dima806/orange-fruit-disease-detection-vit for details.
Classification report:
precision recall f1-score support
citrus canker 0.9806 0.9700 0.9753 1200
healthy 0.9795 0.9933 0.9863 1200
melanose 0.9783 0.9750 0.9766 1200
accuracy 0.9794 3600
macro avg 0.9794 0.9794 0.9794 3600
weighted avg 0.9794 0.9794 0.9794 3600