smids_3x_deit_small_rms_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: 0.8635
- Accuracy: 0.9065
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 |
|---|---|---|---|---|
| 0.3954 | 1.0 | 226 | 0.3893 | 0.8280 |
| 0.2465 | 2.0 | 452 | 0.3428 | 0.8881 |
| 0.1728 | 3.0 | 678 | 0.3745 | 0.8831 |
| 0.1347 | 4.0 | 904 | 0.4194 | 0.8848 |
| 0.0709 | 5.0 | 1130 | 0.4431 | 0.8898 |
| 0.0807 | 6.0 | 1356 | 0.4957 | 0.8781 |
| 0.0722 | 7.0 | 1582 | 0.4748 | 0.8898 |
| 0.036 | 8.0 | 1808 | 0.6178 | 0.8831 |
| 0.0808 | 9.0 | 2034 | 0.5850 | 0.8831 |
| 0.0404 | 10.0 | 2260 | 0.5350 | 0.9015 |
| 0.016 | 11.0 | 2486 | 0.5574 | 0.8831 |
| 0.0147 | 12.0 | 2712 | 0.5709 | 0.8865 |
| 0.0113 | 13.0 | 2938 | 0.6888 | 0.8815 |
| 0.0209 | 14.0 | 3164 | 0.4757 | 0.9149 |
| 0.0245 | 15.0 | 3390 | 0.6913 | 0.8815 |
| 0.0203 | 16.0 | 3616 | 0.6653 | 0.8865 |
| 0.0109 | 17.0 | 3842 | 0.7353 | 0.8898 |
| 0.0341 | 18.0 | 4068 | 0.7660 | 0.8865 |
| 0.0053 | 19.0 | 4294 | 0.6013 | 0.8965 |
| 0.0015 | 20.0 | 4520 | 0.6073 | 0.8965 |
| 0.0003 | 21.0 | 4746 | 0.7366 | 0.8965 |
| 0.0274 | 22.0 | 4972 | 0.7587 | 0.8798 |
| 0.0019 | 23.0 | 5198 | 0.6702 | 0.8998 |
| 0.0001 | 24.0 | 5424 | 0.7767 | 0.8815 |
| 0.0008 | 25.0 | 5650 | 0.6634 | 0.8998 |
| 0.023 | 26.0 | 5876 | 0.7380 | 0.8915 |
| 0.0 | 27.0 | 6102 | 0.8025 | 0.8898 |
| 0.0797 | 28.0 | 6328 | 0.7171 | 0.8948 |
| 0.0492 | 29.0 | 6554 | 0.6827 | 0.8982 |
| 0.0 | 30.0 | 6780 | 0.7690 | 0.9048 |
| 0.0 | 31.0 | 7006 | 0.7411 | 0.9048 |
| 0.0 | 32.0 | 7232 | 0.7425 | 0.8965 |
| 0.0032 | 33.0 | 7458 | 0.7178 | 0.9115 |
| 0.0006 | 34.0 | 7684 | 0.7893 | 0.9082 |
| 0.0 | 35.0 | 7910 | 0.8185 | 0.8932 |
| 0.0181 | 36.0 | 8136 | 0.8745 | 0.8932 |
| 0.0003 | 37.0 | 8362 | 0.8672 | 0.8932 |
| 0.0 | 38.0 | 8588 | 0.8314 | 0.8982 |
| 0.0 | 39.0 | 8814 | 0.8333 | 0.8965 |
| 0.0 | 40.0 | 9040 | 0.7854 | 0.9065 |
| 0.0036 | 41.0 | 9266 | 0.8828 | 0.9015 |
| 0.0028 | 42.0 | 9492 | 0.8402 | 0.9065 |
| 0.0 | 43.0 | 9718 | 0.8689 | 0.8982 |
| 0.0 | 44.0 | 9944 | 0.8390 | 0.9065 |
| 0.0 | 45.0 | 10170 | 0.8434 | 0.9082 |
| 0.0 | 46.0 | 10396 | 0.8531 | 0.9132 |
| 0.0 | 47.0 | 10622 | 0.8589 | 0.9098 |
| 0.0 | 48.0 | 10848 | 0.8632 | 0.9065 |
| 0.0 | 49.0 | 11074 | 0.8630 | 0.9065 |
| 0.0 | 50.0 | 11300 | 0.8635 | 0.9065 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2