This model is a fine-tuned Vision Transformer (ViT) based on facebook/deit-base-patch16-224 for deepfake image classification. It was trained on the Deepfake and Real Images dataset.

This model is intended for research purposes related to deepfake detection and image classification. It can be used to classify images as either 'Fake' or 'Real'.

The model was trained on the Deepfake and Real Images dataset available on Kaggle. The dataset contains both deepfake and real images. Random over-sampling was applied to balance the classes during training.

The model was fine-tuned using the Hugging Face transformers library and Trainer class. The training was performed for 5 epochs with a learning rate of 2e-5 and weight decay of 0.01. Mixed-precision training was enabled.

The model was evaluated on a test set with the following metrics:

A classification report, confusion matrix, and ROC curve are available in the output directory.

You can use this model with the Hugging Face transformers library.