vit
This model is a fine-tuned version of VIT on the Mammogram V1 dataset. It achieves the following results on the evaluation set:
- Loss: 0.1157
- Accuracy: 0.9625
- Precision: 0.9745
- Recall: 0.9625
- F1: 0.9682
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.4204 | 1.0 | 1112 | 0.1572 | 0.9797 | 0.9740 | 0.9797 | 0.9767 |
| 0.3987 | 2.0 | 2224 | 0.2308 | 0.9253 | 0.9745 | 0.9253 | 0.9482 |
| 0.2347 | 3.0 | 3336 | 0.1360 | 0.9516 | 0.9737 | 0.9516 | 0.9622 |
| 0.1283 | 4.0 | 4448 | 0.1255 | 0.9564 | 0.9743 | 0.9564 | 0.9649 |
| 0.1304 | 5.0 | 5560 | 0.1157 | 0.9625 | 0.9745 | 0.9625 | 0.9682 |
Framework versions
- Transformers 4.42.4
- Pytorch 2.3.1+cu121
- Datasets 2.21.0
- Tokenizers 0.19.1
- Downloads last month
- 4