Videomae Large Finetuned Deepfake Subset

🚀 videomae-large-finetuned-deepfake-subset

This model is a fine - tuned version of [MCG - NJU/videomae - large](https://huggingface.co/MCG - NJU/videomae - large) on the [Deepfake Detection Challenge](https://www.kaggle.com/competitions/deepfake - detection - challenge) dataset. It offers enhanced performance for deepfake detection tasks, bringing more accurate results to the field of video authenticity verification.

🚀 Quick Start

This model is a fine - tuned version of [MCG - NJU/videomae - large](https://huggingface.co/MCG - NJU/videomae - large) on the [Deepfake Detection Challenge](https://www.kaggle.com/competitions/deepfake - detection - challenge) dataset. It achieves the following results on the evaluation set:

• Loss: 0.2588

📚 Documentation

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

No specific data details provided in the original document.

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e - 05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon = 1e - 08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 4470
• mixed_precision_training: Native AMP

Training results

Framework versions

• Transformers 4.44.2
• Pytorch 2.5.0+cu121
• Datasets 3.1.0
• Tokenizers 0.19.1

📄 License

This model is released under the CC - BY - NC - 4.0 license.