smids_5x_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: 0.9977
- Accuracy: 0.5159
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 |
|---|---|---|---|---|
| 1.0592 | 1.0 | 376 | 1.0737 | 0.4391 |
| 1.0782 | 2.0 | 752 | 1.0706 | 0.4374 |
| 1.0408 | 3.0 | 1128 | 1.0676 | 0.4391 |
| 1.0826 | 4.0 | 1504 | 1.0646 | 0.4441 |
| 1.0586 | 5.0 | 1880 | 1.0616 | 0.4407 |
| 1.0333 | 6.0 | 2256 | 1.0588 | 0.4424 |
| 1.0668 | 7.0 | 2632 | 1.0559 | 0.4424 |
| 1.0617 | 8.0 | 3008 | 1.0531 | 0.4441 |
| 1.0464 | 9.0 | 3384 | 1.0504 | 0.4457 |
| 1.0296 | 10.0 | 3760 | 1.0477 | 0.4474 |
| 1.0219 | 11.0 | 4136 | 1.0452 | 0.4524 |
| 1.036 | 12.0 | 4512 | 1.0426 | 0.4558 |
| 1.0086 | 13.0 | 4888 | 1.0402 | 0.4591 |
| 1.0374 | 14.0 | 5264 | 1.0378 | 0.4591 |
| 1.0308 | 15.0 | 5640 | 1.0354 | 0.4624 |
| 1.0138 | 16.0 | 6016 | 1.0332 | 0.4641 |
| 1.039 | 17.0 | 6392 | 1.0310 | 0.4674 |
| 1.0251 | 18.0 | 6768 | 1.0289 | 0.4691 |
| 1.0132 | 19.0 | 7144 | 1.0268 | 0.4674 |
| 1.0078 | 20.0 | 7520 | 1.0248 | 0.4674 |
| 1.0073 | 21.0 | 7896 | 1.0229 | 0.4741 |
| 0.9973 | 22.0 | 8272 | 1.0210 | 0.4775 |
| 0.9979 | 23.0 | 8648 | 1.0192 | 0.4791 |
| 0.9943 | 24.0 | 9024 | 1.0175 | 0.4791 |
| 0.9653 | 25.0 | 9400 | 1.0159 | 0.4841 |
| 0.9982 | 26.0 | 9776 | 1.0143 | 0.4841 |
| 1.0041 | 27.0 | 10152 | 1.0128 | 0.4875 |
| 1.0054 | 28.0 | 10528 | 1.0114 | 0.4908 |
| 0.9643 | 29.0 | 10904 | 1.0101 | 0.4925 |
| 0.9735 | 30.0 | 11280 | 1.0088 | 0.4958 |
| 1.0 | 31.0 | 11656 | 1.0076 | 0.4958 |
| 0.998 | 32.0 | 12032 | 1.0064 | 0.4975 |
| 0.9763 | 33.0 | 12408 | 1.0054 | 0.4975 |
| 0.9704 | 34.0 | 12784 | 1.0044 | 0.4992 |
| 0.9948 | 35.0 | 13160 | 1.0035 | 0.5008 |
| 0.9708 | 36.0 | 13536 | 1.0026 | 0.5008 |
| 0.9711 | 37.0 | 13912 | 1.0019 | 0.5025 |
| 0.999 | 38.0 | 14288 | 1.0012 | 0.5042 |
| 0.9534 | 39.0 | 14664 | 1.0005 | 0.5042 |
| 0.9776 | 40.0 | 15040 | 1.0000 | 0.5058 |
| 1.0022 | 41.0 | 15416 | 0.9995 | 0.5058 |
| 0.9618 | 42.0 | 15792 | 0.9991 | 0.5058 |
| 0.9978 | 43.0 | 16168 | 0.9987 | 0.5109 |
| 0.9845 | 44.0 | 16544 | 0.9984 | 0.5142 |
| 0.9783 | 45.0 | 16920 | 0.9982 | 0.5159 |
| 0.99 | 46.0 | 17296 | 0.9980 | 0.5159 |
| 0.9708 | 47.0 | 17672 | 0.9979 | 0.5159 |
| 1.0004 | 48.0 | 18048 | 0.9978 | 0.5159 |
| 0.9871 | 49.0 | 18424 | 0.9977 | 0.5159 |
| 0.9947 | 50.0 | 18800 | 0.9977 | 0.5159 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2