swin-tiny-patch4-window7-224-finetuned-plant
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: 0.5060
- Accuracy: 0.8291
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: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.7171 | 1.0 | 17 | 0.9830 | 0.7094 |
| 0.8722 | 2.0 | 34 | 0.6553 | 0.7564 |
| 0.6241 | 3.0 | 51 | 0.5716 | 0.7991 |
| 0.6185 | 4.0 | 68 | 0.5342 | 0.8162 |
| 0.538 | 5.0 | 85 | 0.6078 | 0.7863 |
| 0.5488 | 6.0 | 102 | 0.5294 | 0.8162 |
| 0.5151 | 7.0 | 119 | 0.5060 | 0.8291 |
| 0.4742 | 8.0 | 136 | 0.5111 | 0.8120 |
| 0.4554 | 9.0 | 153 | 0.4852 | 0.8291 |
| 0.4492 | 10.0 | 170 | 0.4928 | 0.8205 |
Framework versions
- Transformers 4.57.1
- Pytorch 2.9.0+cu126
- Datasets 4.0.0
- Tokenizers 0.22.1