smids_10x_deit_small_rms_00001_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: 0.8415
- Accuracy: 0.9132
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 |
|---|---|---|---|---|
| 0.2066 | 1.0 | 751 | 0.2884 | 0.8948 |
| 0.1058 | 2.0 | 1502 | 0.3397 | 0.8898 |
| 0.0676 | 3.0 | 2253 | 0.3801 | 0.9098 |
| 0.0667 | 4.0 | 3004 | 0.5510 | 0.8998 |
| 0.0253 | 5.0 | 3755 | 0.5205 | 0.9048 |
| 0.0216 | 6.0 | 4506 | 0.6002 | 0.9082 |
| 0.0362 | 7.0 | 5257 | 0.8486 | 0.8815 |
| 0.0357 | 8.0 | 6008 | 0.6658 | 0.9082 |
| 0.035 | 9.0 | 6759 | 0.5803 | 0.9065 |
| 0.001 | 10.0 | 7510 | 0.6719 | 0.9082 |
| 0.0231 | 11.0 | 8261 | 0.7324 | 0.9098 |
| 0.0001 | 12.0 | 9012 | 0.6703 | 0.9082 |
| 0.0008 | 13.0 | 9763 | 0.8412 | 0.8965 |
| 0.0582 | 14.0 | 10514 | 0.7418 | 0.9115 |
| 0.0077 | 15.0 | 11265 | 0.8736 | 0.8998 |
| 0.0 | 16.0 | 12016 | 0.7725 | 0.9115 |
| 0.0019 | 17.0 | 12767 | 0.7981 | 0.9065 |
| 0.0064 | 18.0 | 13518 | 1.0317 | 0.8848 |
| 0.0 | 19.0 | 14269 | 0.8686 | 0.9048 |
| 0.0 | 20.0 | 15020 | 0.8083 | 0.9082 |
| 0.0 | 21.0 | 15771 | 0.6862 | 0.9115 |
| 0.0 | 22.0 | 16522 | 0.7317 | 0.9132 |
| 0.0 | 23.0 | 17273 | 0.6937 | 0.9182 |
| 0.0 | 24.0 | 18024 | 0.7497 | 0.9199 |
| 0.0 | 25.0 | 18775 | 0.8180 | 0.9098 |
| 0.0001 | 26.0 | 19526 | 0.8680 | 0.9098 |
| 0.0 | 27.0 | 20277 | 0.8268 | 0.9115 |
| 0.0 | 28.0 | 21028 | 0.8126 | 0.9082 |
| 0.0042 | 29.0 | 21779 | 0.8397 | 0.9048 |
| 0.024 | 30.0 | 22530 | 0.8418 | 0.9098 |
| 0.0 | 31.0 | 23281 | 0.8800 | 0.9065 |
| 0.0 | 32.0 | 24032 | 0.8577 | 0.9065 |
| 0.0 | 33.0 | 24783 | 0.7988 | 0.9048 |
| 0.0 | 34.0 | 25534 | 0.8415 | 0.9032 |
| 0.0 | 35.0 | 26285 | 0.8311 | 0.9132 |
| 0.0 | 36.0 | 27036 | 0.8203 | 0.9149 |
| 0.0 | 37.0 | 27787 | 0.8136 | 0.9165 |
| 0.0 | 38.0 | 28538 | 0.8129 | 0.9199 |
| 0.0 | 39.0 | 29289 | 0.8165 | 0.9199 |
| 0.0 | 40.0 | 30040 | 0.8151 | 0.9149 |
| 0.0 | 41.0 | 30791 | 0.8242 | 0.9115 |
| 0.0 | 42.0 | 31542 | 0.8225 | 0.9115 |
| 0.0 | 43.0 | 32293 | 0.8294 | 0.9132 |
| 0.0 | 44.0 | 33044 | 0.8299 | 0.9132 |
| 0.0 | 45.0 | 33795 | 0.8337 | 0.9132 |
| 0.0 | 46.0 | 34546 | 0.8327 | 0.9132 |
| 0.0 | 47.0 | 35297 | 0.8358 | 0.9132 |
| 0.0 | 48.0 | 36048 | 0.8374 | 0.9132 |
| 0.0 | 49.0 | 36799 | 0.8388 | 0.9115 |
| 0.0 | 50.0 | 37550 | 0.8415 | 0.9132 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2