vit_4090_downsample_normal_2class
This model is a fine-tuned version of google/vit-large-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.8639
- Accuracy: 0.8829
- Precision: 0.8895
- Recall: 0.8829
- F1: 0.8857
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: 0.0001
- train_batch_size: 24
- eval_batch_size: 4
- seed: 42
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.1837 | 1.0 | 342 | 0.5253 | 0.8633 | 0.8685 | 0.8633 | 0.8657 |
| 0.0921 | 2.0 | 684 | 0.6153 | 0.8636 | 0.8845 | 0.8636 | 0.8713 |
| 0.0515 | 3.0 | 1026 | 0.4083 | 0.8845 | 0.8942 | 0.8845 | 0.8884 |
| 0.0354 | 4.0 | 1368 | 0.4636 | 0.8916 | 0.8972 | 0.8916 | 0.8940 |
| 0.0243 | 5.0 | 1710 | 0.3429 | 0.9074 | 0.9080 | 0.9074 | 0.9077 |
| 0.0137 | 6.0 | 2052 | 0.5370 | 0.8951 | 0.8913 | 0.8951 | 0.8929 |
| 0.0073 | 7.0 | 2394 | 0.6569 | 0.8832 | 0.9020 | 0.8832 | 0.8897 |
| 0.0015 | 8.0 | 2736 | 0.8353 | 0.8821 | 0.8858 | 0.8821 | 0.8838 |
| 0.0023 | 9.0 | 3078 | 0.8605 | 0.8874 | 0.8860 | 0.8874 | 0.8867 |
| 0.001 | 10.0 | 3420 | 0.8639 | 0.8829 | 0.8895 | 0.8829 | 0.8857 |
Framework versions
- Transformers 4.49.0
- Pytorch 2.5.1
- Datasets 3.2.0
- Tokenizers 0.21.1