smids_1x_deit_small_sgd_00001_fold2
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: 1.0336
- Accuracy: 0.5042
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 |
|---|---|---|---|---|
| 1.0925 | 1.0 | 75 | 1.0713 | 0.4343 |
| 1.0735 | 2.0 | 150 | 1.0692 | 0.4343 |
| 1.0724 | 3.0 | 225 | 1.0673 | 0.4393 |
| 1.0873 | 4.0 | 300 | 1.0654 | 0.4426 |
| 1.1019 | 5.0 | 375 | 1.0637 | 0.4426 |
| 1.0577 | 6.0 | 450 | 1.0620 | 0.4459 |
| 1.0861 | 7.0 | 525 | 1.0604 | 0.4493 |
| 1.0644 | 8.0 | 600 | 1.0588 | 0.4542 |
| 1.0424 | 9.0 | 675 | 1.0573 | 0.4509 |
| 1.0503 | 10.0 | 750 | 1.0559 | 0.4509 |
| 1.0641 | 11.0 | 825 | 1.0545 | 0.4493 |
| 1.0679 | 12.0 | 900 | 1.0532 | 0.4526 |
| 1.0629 | 13.0 | 975 | 1.0520 | 0.4542 |
| 1.0438 | 14.0 | 1050 | 1.0508 | 0.4542 |
| 1.061 | 15.0 | 1125 | 1.0497 | 0.4509 |
| 1.0498 | 16.0 | 1200 | 1.0486 | 0.4509 |
| 1.0521 | 17.0 | 1275 | 1.0475 | 0.4559 |
| 1.0469 | 18.0 | 1350 | 1.0466 | 0.4576 |
| 1.047 | 19.0 | 1425 | 1.0456 | 0.4609 |
| 1.0592 | 20.0 | 1500 | 1.0447 | 0.4659 |
| 1.0668 | 21.0 | 1575 | 1.0439 | 0.4709 |
| 1.0281 | 22.0 | 1650 | 1.0431 | 0.4725 |
| 1.0356 | 23.0 | 1725 | 1.0423 | 0.4775 |
| 1.026 | 24.0 | 1800 | 1.0416 | 0.4775 |
| 1.0466 | 25.0 | 1875 | 1.0409 | 0.4792 |
| 1.0451 | 26.0 | 1950 | 1.0402 | 0.4809 |
| 1.0338 | 27.0 | 2025 | 1.0396 | 0.4859 |
| 1.0199 | 28.0 | 2100 | 1.0390 | 0.4842 |
| 1.0289 | 29.0 | 2175 | 1.0384 | 0.4875 |
| 1.0316 | 30.0 | 2250 | 1.0379 | 0.4908 |
| 1.0446 | 31.0 | 2325 | 1.0374 | 0.4925 |
| 1.0407 | 32.0 | 2400 | 1.0369 | 0.4925 |
| 1.0163 | 33.0 | 2475 | 1.0365 | 0.4925 |
| 1.0508 | 34.0 | 2550 | 1.0361 | 0.4925 |
| 1.024 | 35.0 | 2625 | 1.0358 | 0.4942 |
| 1.0435 | 36.0 | 2700 | 1.0354 | 0.4958 |
| 1.0618 | 37.0 | 2775 | 1.0351 | 0.4958 |
| 1.0365 | 38.0 | 2850 | 1.0349 | 0.4975 |
| 1.0269 | 39.0 | 2925 | 1.0346 | 0.4992 |
| 1.0291 | 40.0 | 3000 | 1.0344 | 0.5008 |
| 1.0505 | 41.0 | 3075 | 1.0342 | 0.5008 |
| 1.0316 | 42.0 | 3150 | 1.0340 | 0.5008 |
| 1.0295 | 43.0 | 3225 | 1.0339 | 0.5008 |
| 1.049 | 44.0 | 3300 | 1.0338 | 0.5008 |
| 1.0556 | 45.0 | 3375 | 1.0337 | 0.5008 |
| 1.0458 | 46.0 | 3450 | 1.0336 | 0.5008 |
| 1.0348 | 47.0 | 3525 | 1.0336 | 0.5008 |
| 1.0496 | 48.0 | 3600 | 1.0336 | 0.5025 |
| 1.0321 | 49.0 | 3675 | 1.0336 | 0.5042 |
| 1.0497 | 50.0 | 3750 | 1.0336 | 0.5042 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0