smids_10x_deit_small_rms_0001_fold1
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: 1.0119
- Accuracy: 0.9015
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: 0.0001
- 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 |
|---|---|---|---|---|
| 0.3304 | 1.0 | 751 | 0.3159 | 0.8881 |
| 0.1492 | 2.0 | 1502 | 0.3656 | 0.8798 |
| 0.137 | 3.0 | 2253 | 0.4619 | 0.8831 |
| 0.0713 | 4.0 | 3004 | 0.5428 | 0.8798 |
| 0.0396 | 5.0 | 3755 | 0.4807 | 0.9098 |
| 0.041 | 6.0 | 4506 | 0.6616 | 0.8881 |
| 0.0505 | 7.0 | 5257 | 0.6229 | 0.9032 |
| 0.016 | 8.0 | 6008 | 0.6188 | 0.8865 |
| 0.0115 | 9.0 | 6759 | 0.6016 | 0.8915 |
| 0.0369 | 10.0 | 7510 | 0.5294 | 0.8965 |
| 0.0298 | 11.0 | 8261 | 0.5896 | 0.8865 |
| 0.0282 | 12.0 | 9012 | 0.6477 | 0.8815 |
| 0.0159 | 13.0 | 9763 | 0.5597 | 0.9032 |
| 0.0145 | 14.0 | 10514 | 0.5006 | 0.9032 |
| 0.0274 | 15.0 | 11265 | 0.6365 | 0.8932 |
| 0.0233 | 16.0 | 12016 | 0.5679 | 0.9032 |
| 0.0005 | 17.0 | 12767 | 0.7523 | 0.8815 |
| 0.0253 | 18.0 | 13518 | 0.6785 | 0.8915 |
| 0.0256 | 19.0 | 14269 | 0.6246 | 0.8815 |
| 0.0099 | 20.0 | 15020 | 0.6349 | 0.9082 |
| 0.0038 | 21.0 | 15771 | 0.5512 | 0.9115 |
| 0.0001 | 22.0 | 16522 | 0.6204 | 0.9032 |
| 0.0062 | 23.0 | 17273 | 0.7652 | 0.8932 |
| 0.0 | 24.0 | 18024 | 0.6455 | 0.9048 |
| 0.0 | 25.0 | 18775 | 0.8288 | 0.8932 |
| 0.0004 | 26.0 | 19526 | 0.7865 | 0.8982 |
| 0.0072 | 27.0 | 20277 | 0.8381 | 0.8965 |
| 0.0124 | 28.0 | 21028 | 0.6706 | 0.9082 |
| 0.0027 | 29.0 | 21779 | 0.7345 | 0.9115 |
| 0.0001 | 30.0 | 22530 | 0.8086 | 0.9032 |
| 0.0007 | 31.0 | 23281 | 0.9133 | 0.8948 |
| 0.0 | 32.0 | 24032 | 0.9315 | 0.8948 |
| 0.0005 | 33.0 | 24783 | 0.9041 | 0.8865 |
| 0.0002 | 34.0 | 25534 | 0.7984 | 0.9082 |
| 0.0 | 35.0 | 26285 | 0.7336 | 0.9199 |
| 0.0 | 36.0 | 27036 | 0.7739 | 0.9082 |
| 0.0 | 37.0 | 27787 | 0.7374 | 0.9149 |
| 0.0 | 38.0 | 28538 | 0.8856 | 0.8998 |
| 0.0 | 39.0 | 29289 | 0.7863 | 0.9115 |
| 0.0 | 40.0 | 30040 | 0.8280 | 0.9065 |
| 0.0 | 41.0 | 30791 | 0.8525 | 0.9065 |
| 0.0 | 42.0 | 31542 | 0.8579 | 0.9015 |
| 0.0 | 43.0 | 32293 | 0.9128 | 0.9048 |
| 0.0 | 44.0 | 33044 | 0.9440 | 0.8998 |
| 0.0 | 45.0 | 33795 | 0.9799 | 0.8998 |
| 0.0 | 46.0 | 34546 | 0.9704 | 0.9015 |
| 0.0 | 47.0 | 35297 | 0.9969 | 0.9015 |
| 0.0 | 48.0 | 36048 | 1.0043 | 0.9015 |
| 0.0 | 49.0 | 36799 | 1.0140 | 0.8998 |
| 0.0 | 50.0 | 37550 | 1.0119 | 0.9015 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2