smids_10x_deit_small_sgd_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: 0.4121
- Accuracy: 0.8464
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 |
|---|---|---|---|---|
| 0.9962 | 1.0 | 751 | 1.0240 | 0.4741 |
| 0.8829 | 2.0 | 1502 | 0.9630 | 0.5259 |
| 0.8203 | 3.0 | 2253 | 0.8958 | 0.6027 |
| 0.7637 | 4.0 | 3004 | 0.8327 | 0.6578 |
| 0.7321 | 5.0 | 3755 | 0.7782 | 0.6912 |
| 0.7017 | 6.0 | 4506 | 0.7239 | 0.7078 |
| 0.5812 | 7.0 | 5257 | 0.6809 | 0.7212 |
| 0.581 | 8.0 | 6008 | 0.6410 | 0.7346 |
| 0.5592 | 9.0 | 6759 | 0.6086 | 0.7513 |
| 0.5145 | 10.0 | 7510 | 0.5829 | 0.7679 |
| 0.5332 | 11.0 | 8261 | 0.5629 | 0.7746 |
| 0.4756 | 12.0 | 9012 | 0.5433 | 0.7796 |
| 0.4797 | 13.0 | 9763 | 0.5294 | 0.7846 |
| 0.4315 | 14.0 | 10514 | 0.5168 | 0.7930 |
| 0.4112 | 15.0 | 11265 | 0.5056 | 0.8013 |
| 0.4474 | 16.0 | 12016 | 0.4952 | 0.8030 |
| 0.4529 | 17.0 | 12767 | 0.4868 | 0.8097 |
| 0.421 | 18.0 | 13518 | 0.4802 | 0.8130 |
| 0.4112 | 19.0 | 14269 | 0.4730 | 0.8130 |
| 0.4039 | 20.0 | 15020 | 0.4670 | 0.8180 |
| 0.3219 | 21.0 | 15771 | 0.4615 | 0.8164 |
| 0.411 | 22.0 | 16522 | 0.4563 | 0.8180 |
| 0.3769 | 23.0 | 17273 | 0.4528 | 0.8214 |
| 0.4423 | 24.0 | 18024 | 0.4481 | 0.8214 |
| 0.4214 | 25.0 | 18775 | 0.4442 | 0.8230 |
| 0.4588 | 26.0 | 19526 | 0.4419 | 0.8280 |
| 0.3977 | 27.0 | 20277 | 0.4383 | 0.8314 |
| 0.4288 | 28.0 | 21028 | 0.4359 | 0.8297 |
| 0.3842 | 29.0 | 21779 | 0.4331 | 0.8331 |
| 0.38 | 30.0 | 22530 | 0.4307 | 0.8331 |
| 0.3344 | 31.0 | 23281 | 0.4288 | 0.8347 |
| 0.4273 | 32.0 | 24032 | 0.4264 | 0.8347 |
| 0.3923 | 33.0 | 24783 | 0.4244 | 0.8364 |
| 0.3452 | 34.0 | 25534 | 0.4233 | 0.8364 |
| 0.3666 | 35.0 | 26285 | 0.4214 | 0.8381 |
| 0.3806 | 36.0 | 27036 | 0.4199 | 0.8397 |
| 0.4471 | 37.0 | 27787 | 0.4189 | 0.8397 |
| 0.3236 | 38.0 | 28538 | 0.4183 | 0.8414 |
| 0.2974 | 39.0 | 29289 | 0.4171 | 0.8397 |
| 0.4164 | 40.0 | 30040 | 0.4161 | 0.8397 |
| 0.3819 | 41.0 | 30791 | 0.4153 | 0.8431 |
| 0.3798 | 42.0 | 31542 | 0.4146 | 0.8447 |
| 0.3898 | 43.0 | 32293 | 0.4139 | 0.8447 |
| 0.3508 | 44.0 | 33044 | 0.4133 | 0.8447 |
| 0.3647 | 45.0 | 33795 | 0.4128 | 0.8447 |
| 0.4056 | 46.0 | 34546 | 0.4125 | 0.8447 |
| 0.3591 | 47.0 | 35297 | 0.4123 | 0.8464 |
| 0.4233 | 48.0 | 36048 | 0.4121 | 0.8464 |
| 0.3734 | 49.0 | 36799 | 0.4121 | 0.8464 |
| 0.3779 | 50.0 | 37550 | 0.4121 | 0.8464 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2