vit-base-patch16-224-finetuned-kidney_v4
This model is a fine-tuned version of google/vit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.2255
- Accuracy: 0.9108
- F1: 0.9094
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: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 0.4503 | 1.0 | 734 | 0.3135 | 0.8862 | 0.8840 |
| 0.2869 | 2.0 | 1468 | 0.2501 | 0.9019 | 0.9019 |
| 0.3429 | 3.0 | 2202 | 0.2211 | 0.9074 | 0.9077 |
| 0.3433 | 4.0 | 2936 | 0.2414 | 0.8944 | 0.8895 |
| 0.3379 | 5.0 | 3670 | 0.2332 | 0.8971 | 0.8934 |
| 0.2536 | 6.0 | 4404 | 0.2254 | 0.9087 | 0.9071 |
| 0.2976 | 7.0 | 5138 | 0.2255 | 0.9108 | 0.9094 |
| 0.2516 | 8.0 | 5872 | 0.2178 | 0.9101 | 0.9087 |
| 0.2559 | 9.0 | 6606 | 0.2162 | 0.9101 | 0.9090 |
| 0.2685 | 10.0 | 7340 | 0.2177 | 0.9087 | 0.9076 |
Framework versions
- Transformers 5.0.0
- Pytorch 2.7.1+cu118
- Datasets 4.8.3
- Tokenizers 0.22.2