vit_itri_2class_downsample
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: 0.8733
- Accuracy: 0.8678
- Precision: 0.9081
- Recall: 0.8678
- F1: 0.8832
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 |
|---|---|---|---|---|---|---|---|
| 0.2762 | 1.0 | 197 | 0.2583 | 0.9268 | 0.9254 | 0.9268 | 0.9261 |
| 0.0972 | 2.0 | 394 | 0.6026 | 0.8398 | 0.9182 | 0.8398 | 0.8663 |
| 0.05 | 3.0 | 591 | 0.3871 | 0.9175 | 0.9248 | 0.9175 | 0.9207 |
| 0.0323 | 4.0 | 788 | 0.3336 | 0.9112 | 0.9187 | 0.9112 | 0.9145 |
| 0.0194 | 5.0 | 985 | 0.5212 | 0.9153 | 0.9193 | 0.9153 | 0.9171 |
| 0.0211 | 6.0 | 1182 | 0.4201 | 0.9125 | 0.9167 | 0.9125 | 0.9145 |
| 0.0074 | 7.0 | 1379 | 0.4826 | 0.9151 | 0.9141 | 0.9151 | 0.9146 |
| 0.0014 | 8.0 | 1576 | 0.5316 | 0.9075 | 0.9190 | 0.9075 | 0.9124 |
| 0.003 | 9.0 | 1773 | 0.9022 | 0.8623 | 0.9073 | 0.8623 | 0.8794 |
| 0.0001 | 10.0 | 1970 | 0.8733 | 0.8678 | 0.9081 | 0.8678 | 0.8832 |
Framework versions
- Transformers 4.53.0.dev0
- Pytorch 2.7.1+cu126
- Datasets 3.6.0
- Tokenizers 0.21.1