smids_3x_deit_small_sgd_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: 1.0200
- Accuracy: 0.4925
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 |
|---|---|---|---|---|
| 1.0919 | 1.0 | 226 | 1.0743 | 0.4407 |
| 1.0742 | 2.0 | 452 | 1.0717 | 0.4391 |
| 1.0524 | 3.0 | 678 | 1.0693 | 0.4357 |
| 1.0603 | 4.0 | 904 | 1.0670 | 0.4391 |
| 1.0627 | 5.0 | 1130 | 1.0647 | 0.4391 |
| 1.0778 | 6.0 | 1356 | 1.0625 | 0.4457 |
| 1.0591 | 7.0 | 1582 | 1.0603 | 0.4441 |
| 1.0519 | 8.0 | 1808 | 1.0582 | 0.4457 |
| 1.0538 | 9.0 | 2034 | 1.0562 | 0.4474 |
| 1.0527 | 10.0 | 2260 | 1.0542 | 0.4491 |
| 1.057 | 11.0 | 2486 | 1.0523 | 0.4491 |
| 1.0703 | 12.0 | 2712 | 1.0505 | 0.4491 |
| 1.0101 | 13.0 | 2938 | 1.0488 | 0.4474 |
| 1.0585 | 14.0 | 3164 | 1.0471 | 0.4491 |
| 1.0541 | 15.0 | 3390 | 1.0455 | 0.4541 |
| 1.0416 | 16.0 | 3616 | 1.0438 | 0.4591 |
| 1.03 | 17.0 | 3842 | 1.0423 | 0.4608 |
| 1.0278 | 18.0 | 4068 | 1.0408 | 0.4624 |
| 1.0509 | 19.0 | 4294 | 1.0394 | 0.4641 |
| 1.0188 | 20.0 | 4520 | 1.0381 | 0.4674 |
| 1.0344 | 21.0 | 4746 | 1.0367 | 0.4674 |
| 1.0343 | 22.0 | 4972 | 1.0355 | 0.4691 |
| 1.0324 | 23.0 | 5198 | 1.0343 | 0.4691 |
| 1.0367 | 24.0 | 5424 | 1.0331 | 0.4725 |
| 0.9995 | 25.0 | 5650 | 1.0320 | 0.4725 |
| 1.0409 | 26.0 | 5876 | 1.0310 | 0.4725 |
| 1.0145 | 27.0 | 6102 | 1.0300 | 0.4758 |
| 1.0333 | 28.0 | 6328 | 1.0290 | 0.4758 |
| 1.0283 | 29.0 | 6554 | 1.0281 | 0.4791 |
| 1.0286 | 30.0 | 6780 | 1.0273 | 0.4808 |
| 1.0197 | 31.0 | 7006 | 1.0265 | 0.4808 |
| 1.0259 | 32.0 | 7232 | 1.0257 | 0.4791 |
| 1.0188 | 33.0 | 7458 | 1.0250 | 0.4808 |
| 1.0201 | 34.0 | 7684 | 1.0244 | 0.4808 |
| 1.0339 | 35.0 | 7910 | 1.0238 | 0.4825 |
| 1.0181 | 36.0 | 8136 | 1.0232 | 0.4875 |
| 1.0181 | 37.0 | 8362 | 1.0227 | 0.4875 |
| 1.0115 | 38.0 | 8588 | 1.0223 | 0.4875 |
| 0.9878 | 39.0 | 8814 | 1.0218 | 0.4891 |
| 1.0096 | 40.0 | 9040 | 1.0215 | 0.4891 |
| 1.024 | 41.0 | 9266 | 1.0212 | 0.4891 |
| 1.0376 | 42.0 | 9492 | 1.0209 | 0.4908 |
| 1.0233 | 43.0 | 9718 | 1.0206 | 0.4925 |
| 1.0195 | 44.0 | 9944 | 1.0204 | 0.4925 |
| 1.0135 | 45.0 | 10170 | 1.0203 | 0.4925 |
| 1.039 | 46.0 | 10396 | 1.0202 | 0.4925 |
| 1.0136 | 47.0 | 10622 | 1.0201 | 0.4925 |
| 1.0247 | 48.0 | 10848 | 1.0200 | 0.4925 |
| 1.0178 | 49.0 | 11074 | 1.0200 | 0.4925 |
| 1.0351 | 50.0 | 11300 | 1.0200 | 0.4925 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2