hushem_1x_deit_small_rms_001_fold4
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: 1.2235
- Accuracy: 0.4048
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: 0.001
- 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 |
|---|---|---|---|---|
| No log | 1.0 | 6 | 6.3906 | 0.2381 |
| 3.8063 | 2.0 | 12 | 1.7015 | 0.2619 |
| 3.8063 | 3.0 | 18 | 2.0641 | 0.2619 |
| 1.9221 | 4.0 | 24 | 1.7697 | 0.2381 |
| 1.6782 | 5.0 | 30 | 1.4022 | 0.2619 |
| 1.6782 | 6.0 | 36 | 1.7511 | 0.2381 |
| 1.5442 | 7.0 | 42 | 1.4627 | 0.2381 |
| 1.5442 | 8.0 | 48 | 1.4402 | 0.2619 |
| 1.4869 | 9.0 | 54 | 1.4717 | 0.2619 |
| 1.4572 | 10.0 | 60 | 1.4285 | 0.2381 |
| 1.4572 | 11.0 | 66 | 1.4073 | 0.2619 |
| 1.4861 | 12.0 | 72 | 1.4071 | 0.3095 |
| 1.4861 | 13.0 | 78 | 1.3676 | 0.3095 |
| 1.4283 | 14.0 | 84 | 1.4281 | 0.2381 |
| 1.4135 | 15.0 | 90 | 1.4437 | 0.2381 |
| 1.4135 | 16.0 | 96 | 1.3561 | 0.3095 |
| 1.375 | 17.0 | 102 | 1.3574 | 0.2857 |
| 1.375 | 18.0 | 108 | 1.2368 | 0.2857 |
| 1.3639 | 19.0 | 114 | 1.4601 | 0.2857 |
| 1.2891 | 20.0 | 120 | 1.7927 | 0.2381 |
| 1.2891 | 21.0 | 126 | 1.2451 | 0.4048 |
| 1.3173 | 22.0 | 132 | 1.1578 | 0.4762 |
| 1.3173 | 23.0 | 138 | 1.3222 | 0.3095 |
| 1.2505 | 24.0 | 144 | 1.3748 | 0.2381 |
| 1.263 | 25.0 | 150 | 1.3699 | 0.2857 |
| 1.263 | 26.0 | 156 | 1.2508 | 0.3810 |
| 1.2132 | 27.0 | 162 | 1.1843 | 0.4048 |
| 1.2132 | 28.0 | 168 | 1.4161 | 0.2619 |
| 1.1485 | 29.0 | 174 | 1.1305 | 0.4524 |
| 1.181 | 30.0 | 180 | 1.1818 | 0.4524 |
| 1.181 | 31.0 | 186 | 1.2906 | 0.4048 |
| 1.131 | 32.0 | 192 | 1.1623 | 0.4762 |
| 1.131 | 33.0 | 198 | 1.2826 | 0.4524 |
| 1.164 | 34.0 | 204 | 1.1932 | 0.4524 |
| 1.0879 | 35.0 | 210 | 1.1104 | 0.4286 |
| 1.0879 | 36.0 | 216 | 1.0661 | 0.5714 |
| 1.1012 | 37.0 | 222 | 1.2594 | 0.4048 |
| 1.1012 | 38.0 | 228 | 1.1459 | 0.4286 |
| 1.0505 | 39.0 | 234 | 1.1918 | 0.4524 |
| 1.0052 | 40.0 | 240 | 1.2662 | 0.4286 |
| 1.0052 | 41.0 | 246 | 1.2165 | 0.4048 |
| 0.9631 | 42.0 | 252 | 1.2235 | 0.4048 |
| 0.9631 | 43.0 | 258 | 1.2235 | 0.4048 |
| 0.9397 | 44.0 | 264 | 1.2235 | 0.4048 |
| 0.9545 | 45.0 | 270 | 1.2235 | 0.4048 |
| 0.9545 | 46.0 | 276 | 1.2235 | 0.4048 |
| 0.9591 | 47.0 | 282 | 1.2235 | 0.4048 |
| 0.9591 | 48.0 | 288 | 1.2235 | 0.4048 |
| 0.9579 | 49.0 | 294 | 1.2235 | 0.4048 |
| 0.9362 | 50.0 | 300 | 1.2235 | 0.4048 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1