hushem_1x_deit_tiny_rms_001_fold4
This model is a fine-tuned version of facebook/deit-tiny-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 1.0712
- Accuracy: 0.4762
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 | 5.0165 | 0.2381 |
| 4.2481 | 2.0 | 12 | 3.3074 | 0.2381 |
| 4.2481 | 3.0 | 18 | 1.5288 | 0.2619 |
| 2.0024 | 4.0 | 24 | 1.5375 | 0.2381 |
| 1.6731 | 5.0 | 30 | 1.4069 | 0.2619 |
| 1.6731 | 6.0 | 36 | 1.8969 | 0.2381 |
| 1.5329 | 7.0 | 42 | 1.4811 | 0.2381 |
| 1.5329 | 8.0 | 48 | 1.4117 | 0.2619 |
| 1.475 | 9.0 | 54 | 1.4704 | 0.2619 |
| 1.4639 | 10.0 | 60 | 1.4459 | 0.2381 |
| 1.4639 | 11.0 | 66 | 1.3572 | 0.4524 |
| 1.4524 | 12.0 | 72 | 1.2630 | 0.4524 |
| 1.4524 | 13.0 | 78 | 1.2843 | 0.4524 |
| 1.4025 | 14.0 | 84 | 1.3420 | 0.2857 |
| 1.3666 | 15.0 | 90 | 1.4060 | 0.2381 |
| 1.3666 | 16.0 | 96 | 1.2621 | 0.3810 |
| 1.3178 | 17.0 | 102 | 1.2969 | 0.2857 |
| 1.3178 | 18.0 | 108 | 1.2881 | 0.3333 |
| 1.3667 | 19.0 | 114 | 1.3980 | 0.2857 |
| 1.3043 | 20.0 | 120 | 1.5195 | 0.2857 |
| 1.3043 | 21.0 | 126 | 1.1841 | 0.4048 |
| 1.2859 | 22.0 | 132 | 1.0567 | 0.5238 |
| 1.2859 | 23.0 | 138 | 1.2258 | 0.2619 |
| 1.2496 | 24.0 | 144 | 1.2372 | 0.2857 |
| 1.252 | 25.0 | 150 | 1.4386 | 0.3333 |
| 1.252 | 26.0 | 156 | 1.1416 | 0.3810 |
| 1.2296 | 27.0 | 162 | 1.0872 | 0.4286 |
| 1.2296 | 28.0 | 168 | 1.4121 | 0.2857 |
| 1.1581 | 29.0 | 174 | 1.0555 | 0.5476 |
| 1.2027 | 30.0 | 180 | 1.1296 | 0.4762 |
| 1.2027 | 31.0 | 186 | 1.2095 | 0.4048 |
| 1.1595 | 32.0 | 192 | 1.0821 | 0.4762 |
| 1.1595 | 33.0 | 198 | 1.1681 | 0.3810 |
| 1.1909 | 34.0 | 204 | 1.1147 | 0.4762 |
| 1.1121 | 35.0 | 210 | 1.0734 | 0.4048 |
| 1.1121 | 36.0 | 216 | 1.0002 | 0.5238 |
| 1.1218 | 37.0 | 222 | 1.1912 | 0.3095 |
| 1.1218 | 38.0 | 228 | 1.0883 | 0.4524 |
| 1.1024 | 39.0 | 234 | 1.1229 | 0.4286 |
| 1.0678 | 40.0 | 240 | 1.0903 | 0.4762 |
| 1.0678 | 41.0 | 246 | 1.0717 | 0.4762 |
| 1.058 | 42.0 | 252 | 1.0712 | 0.4762 |
| 1.058 | 43.0 | 258 | 1.0712 | 0.4762 |
| 1.0512 | 44.0 | 264 | 1.0712 | 0.4762 |
| 1.0743 | 45.0 | 270 | 1.0712 | 0.4762 |
| 1.0743 | 46.0 | 276 | 1.0712 | 0.4762 |
| 1.0691 | 47.0 | 282 | 1.0712 | 0.4762 |
| 1.0691 | 48.0 | 288 | 1.0712 | 0.4762 |
| 1.052 | 49.0 | 294 | 1.0712 | 0.4762 |
| 1.066 | 50.0 | 300 | 1.0712 | 0.4762 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1