swin-tiny-patch4-window7-224-category-classification
This model is a fine-tuned version of microsoft/swin-tiny-patch4-window7-224 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.1410
- Accuracy: 0.6445
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: 64
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 256
- 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 |
|---|---|---|---|---|
| 1.89 | 0.9948 | 96 | 1.5075 | 0.5441 |
| 1.6059 | 2.0 | 193 | 1.3103 | 0.5983 |
| 1.4844 | 2.9948 | 289 | 1.2595 | 0.6107 |
| 1.4392 | 4.0 | 386 | 1.2414 | 0.6152 |
| 1.3431 | 4.9948 | 482 | 1.1954 | 0.6285 |
| 1.2897 | 6.0 | 579 | 1.1611 | 0.6384 |
| 1.2222 | 6.9948 | 675 | 1.1575 | 0.6417 |
| 1.212 | 8.0 | 772 | 1.1474 | 0.6421 |
| 1.2087 | 8.9948 | 868 | 1.1410 | 0.6445 |
| 1.1897 | 9.9482 | 960 | 1.1434 | 0.6432 |
Framework versions
- Transformers 4.41.2
- Pytorch 1.13.1+cu117
- Datasets 2.19.2
- Tokenizers 0.19.1