smids_10x_deit_tiny_rms_001_fold1
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: 1.8993
- Accuracy: 0.8047
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 |
|---|---|---|---|---|
| 0.8539 | 1.0 | 751 | 0.7969 | 0.6144 |
| 0.672 | 2.0 | 1502 | 0.7985 | 0.6194 |
| 0.6698 | 3.0 | 2253 | 0.6650 | 0.6962 |
| 0.6307 | 4.0 | 3004 | 0.6193 | 0.7379 |
| 0.6465 | 5.0 | 3755 | 0.8387 | 0.6244 |
| 0.6942 | 6.0 | 4506 | 0.5940 | 0.7329 |
| 0.5576 | 7.0 | 5257 | 0.5999 | 0.7312 |
| 0.5549 | 8.0 | 6008 | 0.6970 | 0.6795 |
| 0.5488 | 9.0 | 6759 | 0.5984 | 0.7229 |
| 0.5437 | 10.0 | 7510 | 0.5960 | 0.7412 |
| 0.5332 | 11.0 | 8261 | 0.5776 | 0.7362 |
| 0.5215 | 12.0 | 9012 | 0.5476 | 0.7880 |
| 0.5333 | 13.0 | 9763 | 0.5233 | 0.7713 |
| 0.4806 | 14.0 | 10514 | 0.5633 | 0.7629 |
| 0.4615 | 15.0 | 11265 | 0.5133 | 0.7830 |
| 0.535 | 16.0 | 12016 | 0.5400 | 0.7362 |
| 0.4592 | 17.0 | 12767 | 0.5660 | 0.7546 |
| 0.4186 | 18.0 | 13518 | 0.5267 | 0.7913 |
| 0.4986 | 19.0 | 14269 | 0.5193 | 0.7846 |
| 0.4777 | 20.0 | 15020 | 0.5053 | 0.7880 |
| 0.4305 | 21.0 | 15771 | 0.5032 | 0.7846 |
| 0.4268 | 22.0 | 16522 | 0.5151 | 0.7830 |
| 0.3969 | 23.0 | 17273 | 0.5356 | 0.7846 |
| 0.4042 | 24.0 | 18024 | 0.5574 | 0.7780 |
| 0.3889 | 25.0 | 18775 | 0.5262 | 0.7679 |
| 0.4127 | 26.0 | 19526 | 0.5352 | 0.7696 |
| 0.3984 | 27.0 | 20277 | 0.5102 | 0.7796 |
| 0.4155 | 28.0 | 21028 | 0.5170 | 0.8030 |
| 0.2889 | 29.0 | 21779 | 0.5194 | 0.8080 |
| 0.2798 | 30.0 | 22530 | 0.5140 | 0.8063 |
| 0.3345 | 31.0 | 23281 | 0.5542 | 0.8097 |
| 0.3266 | 32.0 | 24032 | 0.5467 | 0.7763 |
| 0.3139 | 33.0 | 24783 | 0.5408 | 0.8013 |
| 0.2727 | 34.0 | 25534 | 0.6157 | 0.7997 |
| 0.2326 | 35.0 | 26285 | 0.6453 | 0.8080 |
| 0.2862 | 36.0 | 27036 | 0.6184 | 0.8080 |
| 0.2315 | 37.0 | 27787 | 0.5988 | 0.8097 |
| 0.2292 | 38.0 | 28538 | 0.5807 | 0.8164 |
| 0.1347 | 39.0 | 29289 | 0.6678 | 0.8247 |
| 0.2469 | 40.0 | 30040 | 0.7259 | 0.8080 |
| 0.1108 | 41.0 | 30791 | 0.8996 | 0.7813 |
| 0.1298 | 42.0 | 31542 | 0.8857 | 0.8247 |
| 0.1746 | 43.0 | 32293 | 0.9061 | 0.8197 |
| 0.0873 | 44.0 | 33044 | 1.0379 | 0.8230 |
| 0.087 | 45.0 | 33795 | 1.2137 | 0.8130 |
| 0.1277 | 46.0 | 34546 | 1.3333 | 0.8197 |
| 0.069 | 47.0 | 35297 | 1.4653 | 0.8164 |
| 0.0206 | 48.0 | 36048 | 1.6979 | 0.8063 |
| 0.0279 | 49.0 | 36799 | 1.8534 | 0.8063 |
| 0.02 | 50.0 | 37550 | 1.8993 | 0.8047 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.1+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2