swin-tiny-patch4-window7-224-finetuned
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: 0.5803
- Accuracy: 0.845
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: 1
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.2602 | 0.14 | 10 | 1.9846 | 0.365 |
| 1.8138 | 0.28 | 20 | 1.4061 | 0.614 |
| 1.3707 | 0.43 | 30 | 0.9769 | 0.702 |
| 1.1183 | 0.57 | 40 | 0.7637 | 0.78 |
| 0.9688 | 0.71 | 50 | 0.6487 | 0.812 |
| 0.9258 | 0.85 | 60 | 0.6009 | 0.838 |
| 0.8917 | 0.99 | 70 | 0.5803 | 0.845 |
Framework versions
- Transformers 4.31.0
- Pytorch 1.10.1+cu111
- Datasets 2.14.6
- Tokenizers 0.13.3