mobilevitv2_Liveness_detection_v1.0

This model is a fine-tuned version of apple/mobilevitv2-1.0-imagenet1k-256 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0046
• Accuracy: 0.9988
• F1: 0.9988
• Recall: 0.9988
• Precision: 0.9988

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: 128
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Recall	Precision
0.1093	0.2048	128	0.0679	0.9929	0.9929	0.9929	0.9929
0.0234	0.4096	256	0.0170	0.9962	0.9962	0.9962	0.9962
0.0186	0.6144	384	0.0131	0.9973	0.9973	0.9973	0.9973
0.0068	0.8192	512	0.0089	0.9980	0.9981	0.9980	0.9980
0.0049	1.024	640	0.0067	0.9985	0.9985	0.9985	0.9985
0.0113	1.2288	768	0.0064	0.9983	0.9984	0.9983	0.9983
0.0061	1.4336	896	0.0060	0.9983	0.9983	0.9983	0.9984
0.0025	1.6384	1024	0.0058	0.9983	0.9983	0.9983	0.9984
0.0019	1.8432	1152	0.0053	0.9987	0.9986	0.9987	0.9987
0.0056	2.048	1280	0.0051	0.9987	0.9987	0.9987	0.9987
0.0015	2.2528	1408	0.0050	0.9987	0.9987	0.9987	0.9987
0.0055	2.4576	1536	0.0049	0.9988	0.9987	0.9988	0.9988
0.0023	2.6624	1664	0.0049	0.9989	0.9988	0.9989	0.9989
0.0027	2.8672	1792	0.0046	0.9988	0.9988	0.9988	0.9988

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.2.0
• Tokenizers 0.19.1