smids_10x_deit_small_rms_00001_fold2
This model is a fine-tuned version of facebook/deit-small-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 1.1223
- Accuracy: 0.8952
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: 1e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 50
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.2125 | 1.0 | 750 | 0.2971 | 0.8636 |
| 0.1007 | 2.0 | 1500 | 0.3569 | 0.8902 |
| 0.033 | 3.0 | 2250 | 0.4786 | 0.8852 |
| 0.0414 | 4.0 | 3000 | 0.6308 | 0.8719 |
| 0.0169 | 5.0 | 3750 | 0.7881 | 0.8769 |
| 0.0209 | 6.0 | 4500 | 0.8756 | 0.8802 |
| 0.0232 | 7.0 | 5250 | 0.7942 | 0.8785 |
| 0.0001 | 8.0 | 6000 | 0.8024 | 0.8885 |
| 0.0037 | 9.0 | 6750 | 0.9766 | 0.8852 |
| 0.0663 | 10.0 | 7500 | 0.9288 | 0.8785 |
| 0.0416 | 11.0 | 8250 | 1.0051 | 0.8835 |
| 0.0257 | 12.0 | 9000 | 1.1036 | 0.8752 |
| 0.0003 | 13.0 | 9750 | 0.9284 | 0.8835 |
| 0.0007 | 14.0 | 10500 | 0.9766 | 0.8752 |
| 0.0009 | 15.0 | 11250 | 1.0060 | 0.8869 |
| 0.024 | 16.0 | 12000 | 0.9566 | 0.8918 |
| 0.0002 | 17.0 | 12750 | 0.9308 | 0.8985 |
| 0.0226 | 18.0 | 13500 | 0.9878 | 0.8952 |
| 0.0002 | 19.0 | 14250 | 1.0344 | 0.8802 |
| 0.0 | 20.0 | 15000 | 1.0012 | 0.8902 |
| 0.0 | 21.0 | 15750 | 1.0757 | 0.8852 |
| 0.0197 | 22.0 | 16500 | 1.1327 | 0.8918 |
| 0.0059 | 23.0 | 17250 | 1.1959 | 0.8785 |
| 0.014 | 24.0 | 18000 | 0.9244 | 0.8918 |
| 0.0 | 25.0 | 18750 | 1.0134 | 0.8952 |
| 0.0001 | 26.0 | 19500 | 1.2273 | 0.8735 |
| 0.0081 | 27.0 | 20250 | 1.2216 | 0.8735 |
| 0.0 | 28.0 | 21000 | 1.1304 | 0.8769 |
| 0.0 | 29.0 | 21750 | 0.9499 | 0.8902 |
| 0.0 | 30.0 | 22500 | 1.0368 | 0.8885 |
| 0.0 | 31.0 | 23250 | 1.0392 | 0.8852 |
| 0.0038 | 32.0 | 24000 | 1.2288 | 0.8835 |
| 0.0 | 33.0 | 24750 | 1.1678 | 0.8952 |
| 0.0 | 34.0 | 25500 | 1.0162 | 0.8918 |
| 0.0 | 35.0 | 26250 | 1.0770 | 0.8918 |
| 0.0 | 36.0 | 27000 | 1.0678 | 0.8902 |
| 0.0067 | 37.0 | 27750 | 1.0739 | 0.8935 |
| 0.0 | 38.0 | 28500 | 1.1577 | 0.8935 |
| 0.0 | 39.0 | 29250 | 1.1277 | 0.8935 |
| 0.0 | 40.0 | 30000 | 1.1396 | 0.8918 |
| 0.0 | 41.0 | 30750 | 1.1296 | 0.8952 |
| 0.0 | 42.0 | 31500 | 1.1324 | 0.8935 |
| 0.0 | 43.0 | 32250 | 1.1390 | 0.8918 |
| 0.0 | 44.0 | 33000 | 1.1307 | 0.8952 |
| 0.0025 | 45.0 | 33750 | 1.1302 | 0.8918 |
| 0.0 | 46.0 | 34500 | 1.1293 | 0.8935 |
| 0.0 | 47.0 | 35250 | 1.1264 | 0.8935 |
| 0.0 | 48.0 | 36000 | 1.1267 | 0.8952 |
| 0.0 | 49.0 | 36750 | 1.1233 | 0.8952 |
| 0.0 | 50.0 | 37500 | 1.1223 | 0.8952 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2