swin-tiny-patch4-window7-224-finetuned-student_six_classes
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: 0.1187
- Accuracy: 0.9513
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: 5e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.619 | 0.94 | 11 | 1.1587 | 0.4984 |
| 0.841 | 1.96 | 23 | 0.5082 | 0.7689 |
| 0.4154 | 2.98 | 35 | 0.2849 | 0.8868 |
| 0.3476 | 4.0 | 47 | 0.2089 | 0.9418 |
| 0.2414 | 4.94 | 58 | 0.1575 | 0.9450 |
| 0.2128 | 5.96 | 70 | 0.1226 | 0.9497 |
| 0.1783 | 6.98 | 82 | 0.1203 | 0.9481 |
| 0.167 | 8.0 | 94 | 0.1169 | 0.9528 |
| 0.1723 | 8.94 | 105 | 0.1184 | 0.9513 |
| 0.1838 | 9.36 | 110 | 0.1187 | 0.9513 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2