seeai_vit_itri_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: 0.3552
- Accuracy: 0.9256
- Precision: 0.9253
- Recall: 0.9256
- F1: 0.9253
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 |
|---|---|---|---|---|---|---|---|
| 0.4297 | 1.0 | 555 | 0.4344 | 0.8127 | 0.8539 | 0.8127 | 0.7881 |
| 0.3102 | 2.0 | 1110 | 0.2700 | 0.8925 | 0.8968 | 0.8925 | 0.8887 |
| 0.2604 | 3.0 | 1665 | 0.2374 | 0.9094 | 0.9104 | 0.9094 | 0.9075 |
| 0.2334 | 4.0 | 2220 | 0.3126 | 0.9026 | 0.9088 | 0.9026 | 0.9039 |
| 0.1979 | 5.0 | 2775 | 0.2212 | 0.9121 | 0.9117 | 0.9121 | 0.9111 |
| 0.1596 | 6.0 | 3330 | 0.2598 | 0.9013 | 0.9024 | 0.9013 | 0.9017 |
| 0.1299 | 7.0 | 3885 | 0.2788 | 0.9168 | 0.9163 | 0.9168 | 0.9163 |
| 0.1003 | 8.0 | 4440 | 0.3066 | 0.9189 | 0.9184 | 0.9189 | 0.9184 |
| 0.0696 | 9.0 | 4995 | 0.2737 | 0.9249 | 0.9246 | 0.9249 | 0.9243 |
| 0.0448 | 10.0 | 5550 | 0.3552 | 0.9256 | 0.9253 | 0.9256 | 0.9253 |
Framework versions
- Transformers 4.53.0.dev0
- Pytorch 2.7.1+cu126
- Datasets 3.6.0
- Tokenizers 0.21.1