smids_5x_deit_tiny_rms_0001_fold1
This model is a fine-tuned version of facebook/deit-tiny-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.9972
- Accuracy: 0.9048
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 |
|---|---|---|---|---|
| 0.2949 | 1.0 | 376 | 0.4792 | 0.7896 |
| 0.1877 | 2.0 | 752 | 0.3869 | 0.8631 |
| 0.1943 | 3.0 | 1128 | 0.4273 | 0.8514 |
| 0.1151 | 4.0 | 1504 | 0.4170 | 0.8932 |
| 0.1309 | 5.0 | 1880 | 0.4159 | 0.8748 |
| 0.0937 | 6.0 | 2256 | 0.5222 | 0.8831 |
| 0.0299 | 7.0 | 2632 | 0.5974 | 0.8932 |
| 0.0659 | 8.0 | 3008 | 0.6171 | 0.8715 |
| 0.0586 | 9.0 | 3384 | 0.7200 | 0.8781 |
| 0.0715 | 10.0 | 3760 | 0.9149 | 0.8664 |
| 0.0752 | 11.0 | 4136 | 0.7964 | 0.8765 |
| 0.0401 | 12.0 | 4512 | 0.6968 | 0.8831 |
| 0.0094 | 13.0 | 4888 | 0.6898 | 0.8865 |
| 0.0111 | 14.0 | 5264 | 0.7411 | 0.8932 |
| 0.0334 | 15.0 | 5640 | 0.8411 | 0.8798 |
| 0.0369 | 16.0 | 6016 | 0.7849 | 0.8798 |
| 0.0017 | 17.0 | 6392 | 0.7191 | 0.8898 |
| 0.0026 | 18.0 | 6768 | 0.8047 | 0.8815 |
| 0.0265 | 19.0 | 7144 | 0.6550 | 0.8982 |
| 0.0527 | 20.0 | 7520 | 0.7590 | 0.8798 |
| 0.0052 | 21.0 | 7896 | 0.7860 | 0.8881 |
| 0.001 | 22.0 | 8272 | 0.8487 | 0.8965 |
| 0.0432 | 23.0 | 8648 | 0.8524 | 0.8865 |
| 0.0032 | 24.0 | 9024 | 0.8174 | 0.9015 |
| 0.0001 | 25.0 | 9400 | 0.8214 | 0.8815 |
| 0.0146 | 26.0 | 9776 | 0.9080 | 0.8765 |
| 0.0 | 27.0 | 10152 | 0.8028 | 0.9032 |
| 0.0001 | 28.0 | 10528 | 0.9579 | 0.8915 |
| 0.0043 | 29.0 | 10904 | 0.8349 | 0.8982 |
| 0.0053 | 30.0 | 11280 | 0.9140 | 0.8831 |
| 0.0204 | 31.0 | 11656 | 0.9273 | 0.8898 |
| 0.0001 | 32.0 | 12032 | 0.9480 | 0.8848 |
| 0.0006 | 33.0 | 12408 | 1.0366 | 0.8865 |
| 0.0042 | 34.0 | 12784 | 1.0682 | 0.8798 |
| 0.0025 | 35.0 | 13160 | 0.9542 | 0.8932 |
| 0.0006 | 36.0 | 13536 | 0.8930 | 0.9048 |
| 0.0001 | 37.0 | 13912 | 0.9451 | 0.8932 |
| 0.0112 | 38.0 | 14288 | 1.0303 | 0.8848 |
| 0.0 | 39.0 | 14664 | 1.0298 | 0.8932 |
| 0.0 | 40.0 | 15040 | 0.9996 | 0.8932 |
| 0.0 | 41.0 | 15416 | 0.9909 | 0.8998 |
| 0.0 | 42.0 | 15792 | 0.9652 | 0.9015 |
| 0.0 | 43.0 | 16168 | 0.9547 | 0.9032 |
| 0.0 | 44.0 | 16544 | 0.9994 | 0.8982 |
| 0.0 | 45.0 | 16920 | 0.9802 | 0.9015 |
| 0.003 | 46.0 | 17296 | 0.9911 | 0.9032 |
| 0.0 | 47.0 | 17672 | 0.9936 | 0.9048 |
| 0.0 | 48.0 | 18048 | 0.9937 | 0.9048 |
| 0.0 | 49.0 | 18424 | 0.9932 | 0.9048 |
| 0.0025 | 50.0 | 18800 | 0.9972 | 0.9048 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.1+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2