metadata
library_name: transformers
license: apache-2.0
base_model: facebook/dinov2-small
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
- recall
- precision
model-index:
- name: dinov2_Liveness_detection_v2.2.1
results: []
dinov2_Liveness_detection_v2.2.1
This model is a fine-tuned version of facebook/dinov2-small on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0671
- Accuracy: 0.9869
- F1: 0.9868
- Recall: 0.9869
- Precision: 0.9870
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: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Recall | Precision |
|---|---|---|---|---|---|---|---|
| 0.183 | 0.8153 | 128 | 0.2473 | 0.9016 | 0.9039 | 0.9016 | 0.9123 |
| 0.1022 | 1.6306 | 256 | 0.0750 | 0.9729 | 0.9727 | 0.9729 | 0.9737 |
| 0.0432 | 2.4459 | 384 | 0.0575 | 0.9820 | 0.9820 | 0.9820 | 0.9823 |
| 0.0247 | 3.2611 | 512 | 0.0507 | 0.9832 | 0.9832 | 0.9832 | 0.9833 |
| 0.0115 | 4.0764 | 640 | 0.0536 | 0.9865 | 0.9864 | 0.9865 | 0.9866 |
| 0.002 | 4.8917 | 768 | 0.0671 | 0.9869 | 0.9868 | 0.9869 | 0.9870 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.1+cu121
- Datasets 3.2.0
- Tokenizers 0.19.1