<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
vit-base-driver-drowsiness-detection
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the chbh7051/driver-drowsiness-detection dataset. It achieves the following results on the evaluation set:
- Loss: 0.0800
- Accuracy: 0.9752
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: 0.0002
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.4811 | 0.6 | 2000 | 0.5214 | 0.7636 |
| 0.3339 | 1.2 | 4000 | 0.3437 | 0.8621 |
| 0.284 | 1.8 | 6000 | 0.2679 | 0.8932 |
| 0.2143 | 2.41 | 8000 | 0.2269 | 0.9125 |
| 0.0997 | 3.01 | 10000 | 0.1576 | 0.9444 |
| 0.1168 | 3.61 | 12000 | 0.1214 | 0.9596 |
| 0.0873 | 4.21 | 14000 | 0.1256 | 0.9550 |
| 0.06 | 4.81 | 16000 | 0.0800 | 0.9752 |
Framework versions
- Transformers 4.27.4
- Pytorch 1.13.0
- Datasets 2.1.0
- Tokenizers 0.13.2
