swin-tiny-patch4-window7-224-image-classifier
This model is a fine-tuned version of microsoft/swin-tiny-patch4-window7-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.3204
- Accuracy: 0.8081
- F1: 0.7504
- Precision: 0.6851
- Recall: 0.8295
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: 3e-05
- train_batch_size: 64
- eval_batch_size: 64
- 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
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 30
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| 0.6377 | 1.0 | 143 | 0.6156 | 0.6421 | 0.5356 | 0.4881 | 0.5934 |
| 0.4556 | 2.0 | 286 | 0.4928 | 0.7141 | 0.6286 | 0.5734 | 0.6957 |
| 0.3616 | 3.0 | 429 | 0.5772 | 0.6895 | 0.5930 | 0.5450 | 0.6503 |
| 0.3582 | 4.0 | 572 | 0.3441 | 0.7835 | 0.6644 | 0.7208 | 0.6162 |
| 0.3374 | 5.0 | 715 | 0.4094 | 0.7699 | 0.7434 | 0.6072 | 0.9583 |
| 0.3273 | 6.0 | 858 | 0.6065 | 0.7115 | 0.6364 | 0.5665 | 0.7260 |
| 0.3091 | 7.0 | 1001 | 0.3204 | 0.8081 | 0.7504 | 0.6851 | 0.8295 |
| 0.3026 | 8.0 | 1144 | 0.3946 | 0.7694 | 0.6120 | 0.7380 | 0.5227 |
Framework versions
- Transformers 4.46.2
- Pytorch 2.5.0+cu121
- Datasets 3.1.0
- Tokenizers 0.20.3