dinov2_Liveness_detection_v2.1.1

This model is a fine-tuned version of nguyenkhoa/dinov2_Liveness_detection_v2.1 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0419
• Accuracy: 0.9920
• F1: 0.9921
• Recall: 0.9920
• Precision: 0.9920

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 768
• eval_batch_size: 8
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Recall	Precision
0.0487	1.2190	128	0.0476	0.9841	0.9842	0.9841	0.9841
0.0216	2.4381	256	0.0318	0.9902	0.9902	0.9902	0.9903
0.0081	3.6571	384	0.0426	0.9896	0.9896	0.9896	0.9896
0.0012	4.8762	512	0.0419	0.9920	0.9921	0.9920	0.9920

Evaluate results

• Accuracy: 0.8675
• F1: 0.8973
• Recall: 0.9506
• Precision: 0.7105
• APCER: 0.1687
• BPCER: 0.0494
• ACER: 0.1091

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0