leaftype-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: 0.1713
- Accuracy: 0.9263
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: 30
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.7327 | 0.94 | 12 | 0.6155 | 0.6044 |
| 0.5479 | 1.96 | 25 | 0.4489 | 0.7936 |
| 0.3205 | 2.98 | 38 | 0.2164 | 0.9115 |
| 0.2386 | 4.0 | 51 | 0.2124 | 0.9115 |
| 0.1937 | 4.94 | 63 | 0.2349 | 0.9091 |
| 0.1884 | 5.96 | 76 | 0.1752 | 0.9214 |
| 0.2186 | 6.98 | 89 | 0.3518 | 0.8698 |
| 0.1833 | 8.0 | 102 | 0.2443 | 0.9017 |
| 0.1856 | 8.94 | 114 | 0.2492 | 0.9017 |
| 0.1605 | 9.96 | 127 | 0.2005 | 0.9189 |
| 0.1934 | 10.98 | 140 | 0.1713 | 0.9263 |
| 0.186 | 12.0 | 153 | 0.1573 | 0.9238 |
| 0.1407 | 12.94 | 165 | 0.1658 | 0.9214 |
| 0.1655 | 13.96 | 178 | 0.1570 | 0.9214 |
| 0.1448 | 14.98 | 191 | 0.1622 | 0.9238 |
| 0.1525 | 16.0 | 204 | 0.2110 | 0.9165 |
| 0.1226 | 16.94 | 216 | 0.2109 | 0.9165 |
| 0.1556 | 17.96 | 229 | 0.1914 | 0.9165 |
| 0.1264 | 18.98 | 242 | 0.2251 | 0.9115 |
| 0.1249 | 20.0 | 255 | 0.2863 | 0.8993 |
| 0.1099 | 20.94 | 267 | 0.2447 | 0.9066 |
| 0.1223 | 21.96 | 280 | 0.2358 | 0.9042 |
| 0.1084 | 22.98 | 293 | 0.1713 | 0.9238 |
| 0.1054 | 24.0 | 306 | 0.2085 | 0.9115 |
| 0.1055 | 24.94 | 318 | 0.2002 | 0.9115 |
| 0.1167 | 25.96 | 331 | 0.2289 | 0.9140 |
| 0.1156 | 26.98 | 344 | 0.1889 | 0.9115 |
| 0.0919 | 28.0 | 357 | 0.2015 | 0.9115 |
| 0.0827 | 28.24 | 360 | 0.2021 | 0.9091 |
Framework versions
- Transformers 4.38.1
- Pytorch 1.10.0+cu111
- Datasets 2.17.1
- Tokenizers 0.15.2