smids_5x_deit_tiny_rms_00001_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: 0.9495
- Accuracy: 0.8982
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 |
|---|---|---|---|---|
| 0.2593 | 1.0 | 376 | 0.3151 | 0.8765 |
| 0.1718 | 2.0 | 752 | 0.2623 | 0.8998 |
| 0.1615 | 3.0 | 1128 | 0.2861 | 0.8965 |
| 0.0982 | 4.0 | 1504 | 0.3444 | 0.8881 |
| 0.0553 | 5.0 | 1880 | 0.3784 | 0.9098 |
| 0.0747 | 6.0 | 2256 | 0.5204 | 0.8881 |
| 0.0183 | 7.0 | 2632 | 0.5683 | 0.8948 |
| 0.0068 | 8.0 | 3008 | 0.6428 | 0.8998 |
| 0.0727 | 9.0 | 3384 | 0.7962 | 0.8815 |
| 0.0001 | 10.0 | 3760 | 0.7940 | 0.8965 |
| 0.001 | 11.0 | 4136 | 0.9819 | 0.8681 |
| 0.0 | 12.0 | 4512 | 0.8908 | 0.8848 |
| 0.0018 | 13.0 | 4888 | 0.8621 | 0.8865 |
| 0.0198 | 14.0 | 5264 | 0.8948 | 0.8881 |
| 0.0291 | 15.0 | 5640 | 0.9361 | 0.8915 |
| 0.0001 | 16.0 | 6016 | 0.7825 | 0.8948 |
| 0.0 | 17.0 | 6392 | 0.8996 | 0.8815 |
| 0.0001 | 18.0 | 6768 | 0.8212 | 0.8948 |
| 0.0026 | 19.0 | 7144 | 0.8543 | 0.8831 |
| 0.0145 | 20.0 | 7520 | 0.8936 | 0.8881 |
| 0.004 | 21.0 | 7896 | 0.9825 | 0.8815 |
| 0.0 | 22.0 | 8272 | 0.9004 | 0.8932 |
| 0.0001 | 23.0 | 8648 | 0.8961 | 0.8965 |
| 0.0 | 24.0 | 9024 | 1.0000 | 0.8915 |
| 0.0 | 25.0 | 9400 | 0.9507 | 0.8865 |
| 0.079 | 26.0 | 9776 | 1.0040 | 0.8865 |
| 0.0 | 27.0 | 10152 | 0.9365 | 0.8998 |
| 0.0 | 28.0 | 10528 | 0.9689 | 0.8815 |
| 0.0089 | 29.0 | 10904 | 0.9542 | 0.8898 |
| 0.0105 | 30.0 | 11280 | 0.9853 | 0.8898 |
| 0.0 | 31.0 | 11656 | 0.9962 | 0.8965 |
| 0.0 | 32.0 | 12032 | 0.9324 | 0.8982 |
| 0.0 | 33.0 | 12408 | 1.0542 | 0.8881 |
| 0.0 | 34.0 | 12784 | 0.9887 | 0.8932 |
| 0.0 | 35.0 | 13160 | 0.8827 | 0.9082 |
| 0.0 | 36.0 | 13536 | 0.8957 | 0.8982 |
| 0.0 | 37.0 | 13912 | 0.9316 | 0.8932 |
| 0.0 | 38.0 | 14288 | 0.9562 | 0.8915 |
| 0.0 | 39.0 | 14664 | 0.9229 | 0.8982 |
| 0.0 | 40.0 | 15040 | 0.9352 | 0.8932 |
| 0.0 | 41.0 | 15416 | 0.9221 | 0.8915 |
| 0.0 | 42.0 | 15792 | 0.9253 | 0.8965 |
| 0.0 | 43.0 | 16168 | 0.9330 | 0.8881 |
| 0.0 | 44.0 | 16544 | 0.9447 | 0.8965 |
| 0.0 | 45.0 | 16920 | 0.9432 | 0.8965 |
| 0.0047 | 46.0 | 17296 | 0.9445 | 0.8965 |
| 0.0 | 47.0 | 17672 | 0.9464 | 0.8948 |
| 0.0 | 48.0 | 18048 | 0.9465 | 0.8948 |
| 0.0 | 49.0 | 18424 | 0.9475 | 0.8982 |
| 0.0039 | 50.0 | 18800 | 0.9495 | 0.8982 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.1+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2