--- library_name: peft base_model: nguyenkhoa/dinov2_Liveness_detection_v2.1 tags: - generated_from_trainer metrics: - accuracy - f1 - recall - precision model-index: - name: dinov2_LoRA_Liveness_detection_v1.1 results: [] --- [Visualize in Weights & Biases](https://wandb.ai/nguyenkhoaht002/liveness_detection/runs/an3h6pu9) # dinov2_LoRA_Liveness_detection_v1.1 This model is a fine-tuned version of [nguyenkhoa/dinov2_Liveness_detection_v2.1](https://huggingface.co/nguyenkhoa/dinov2_Liveness_detection_v2.1) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.0243 - Accuracy: 0.9917 - F1: 0.9917 - Recall: 0.9917 - Precision: 0.9917 ## 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: 512 - 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 | |:-------------:|:------:|:----:|:---------------:|:--------:|:------:|:------:|:---------:| | 0.0501 | 0.4076 | 64 | 0.0458 | 0.9841 | 0.9841 | 0.9841 | 0.9842 | | 0.0355 | 0.8153 | 128 | 0.0342 | 0.9880 | 0.9880 | 0.9880 | 0.9880 | | 0.0253 | 1.2229 | 192 | 0.0308 | 0.9901 | 0.9901 | 0.9901 | 0.9901 | | 0.0226 | 1.6306 | 256 | 0.0267 | 0.9910 | 0.9910 | 0.9910 | 0.9910 | | 0.0264 | 2.0382 | 320 | 0.0255 | 0.9912 | 0.9912 | 0.9912 | 0.9912 | | 0.0168 | 2.4459 | 384 | 0.0245 | 0.9922 | 0.9922 | 0.9922 | 0.9922 | | 0.0144 | 2.8535 | 448 | 0.0243 | 0.9917 | 0.9917 | 0.9917 | 0.9917 | ### Framework versions - PEFT 0.14.0 - Transformers 4.44.2 - Pytorch 2.4.1+cu121 - Datasets 3.2.0 - Tokenizers 0.19.1