metadata
license: apache-2.0
pipeline_tag: image-classification
library_name: transformers
tags:
- deep-fake
- ViT
- detection
- Image
- transformers-4.49.0.dev0
- precision-92.12
base_model:
- google/vit-base-patch16-224-in21k
Classification report:
precision recall f1-score support
Realism 0.9683 0.8708 0.9170 28001
Deepfake 0.8826 0.9715 0.9249 28000
accuracy 0.9212 56001
macro avg 0.9255 0.9212 0.9210 56001
weighted avg 0.9255 0.9212 0.9210 56001
Machine training time : 247.9998 Minute / 4.13333 Hour
