vit_itri_64
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: 0.0387
- Accuracy: 0.9933
- Precision: 0.9933
- Recall: 0.9933
- F1: 0.9932
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: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.2675 | 1.0 | 1061 | 0.1884 | 0.9544 | 0.9566 | 0.9544 | 0.9535 |
| 0.1119 | 2.0 | 2122 | 0.1500 | 0.9523 | 0.9580 | 0.9523 | 0.9535 |
| 0.0732 | 3.0 | 3183 | 0.0909 | 0.9784 | 0.9786 | 0.9784 | 0.9778 |
| 0.0562 | 4.0 | 4244 | 0.0629 | 0.9845 | 0.9845 | 0.9845 | 0.9843 |
| 0.042 | 5.0 | 5305 | 0.1149 | 0.9767 | 0.9768 | 0.9767 | 0.9761 |
| 0.0332 | 6.0 | 6366 | 0.0357 | 0.9898 | 0.9896 | 0.9898 | 0.9896 |
| 0.0252 | 7.0 | 7427 | 0.0797 | 0.9770 | 0.9789 | 0.9770 | 0.9772 |
| 0.0214 | 8.0 | 8488 | 0.0491 | 0.9890 | 0.9892 | 0.9890 | 0.9888 |
| 0.0204 | 9.0 | 9549 | 0.0627 | 0.9869 | 0.9869 | 0.9869 | 0.9868 |
| 0.0136 | 10.0 | 10610 | 0.0429 | 0.9898 | 0.9898 | 0.9898 | 0.9897 |
| 0.0126 | 11.0 | 11671 | 0.0439 | 0.9898 | 0.9896 | 0.9898 | 0.9894 |
| 0.0088 | 12.0 | 12732 | 0.0411 | 0.9894 | 0.9894 | 0.9894 | 0.9891 |
| 0.0082 | 13.0 | 13793 | 0.0442 | 0.9901 | 0.9902 | 0.9901 | 0.9900 |
| 0.0033 | 14.0 | 14854 | 0.0457 | 0.9915 | 0.9916 | 0.9915 | 0.9913 |
| 0.0039 | 15.0 | 15915 | 0.0396 | 0.9926 | 0.9926 | 0.9926 | 0.9925 |
| 0.0027 | 16.0 | 16976 | 0.0713 | 0.9905 | 0.9905 | 0.9905 | 0.9903 |
| 0.0015 | 17.0 | 18037 | 0.0331 | 0.9951 | 0.9951 | 0.9951 | 0.9950 |
| 0.0013 | 18.0 | 19098 | 0.0458 | 0.9926 | 0.9926 | 0.9926 | 0.9924 |
| 0.0005 | 19.0 | 20159 | 0.0374 | 0.9943 | 0.9944 | 0.9943 | 0.9943 |
| 0.0009 | 20.0 | 21220 | 0.0387 | 0.9933 | 0.9933 | 0.9933 | 0.9932 |
Framework versions
- Transformers 4.53.0.dev0
- Pytorch 2.7.1+cu126
- Datasets 3.6.0
- Tokenizers 0.21.1