smids_5x_deit_tiny_adamax_0001_fold4
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: 1.2292
- Accuracy: 0.88
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 |
|---|---|---|---|---|
| 0.2019 | 1.0 | 375 | 0.3616 | 0.8683 |
| 0.2348 | 2.0 | 750 | 0.5390 | 0.7983 |
| 0.0464 | 3.0 | 1125 | 0.5043 | 0.88 |
| 0.0924 | 4.0 | 1500 | 0.5883 | 0.8833 |
| 0.0137 | 5.0 | 1875 | 0.7305 | 0.8783 |
| 0.0256 | 6.0 | 2250 | 0.8161 | 0.8783 |
| 0.0006 | 7.0 | 2625 | 0.7997 | 0.8833 |
| 0.0263 | 8.0 | 3000 | 0.8542 | 0.885 |
| 0.0002 | 9.0 | 3375 | 0.9159 | 0.87 |
| 0.0 | 10.0 | 3750 | 0.9248 | 0.8833 |
| 0.0181 | 11.0 | 4125 | 1.0824 | 0.8633 |
| 0.0031 | 12.0 | 4500 | 0.9537 | 0.89 |
| 0.0115 | 13.0 | 4875 | 1.0751 | 0.8667 |
| 0.0169 | 14.0 | 5250 | 0.8764 | 0.8867 |
| 0.0 | 15.0 | 5625 | 0.9541 | 0.8817 |
| 0.0 | 16.0 | 6000 | 1.0324 | 0.87 |
| 0.0003 | 17.0 | 6375 | 1.0424 | 0.8733 |
| 0.0131 | 18.0 | 6750 | 1.0393 | 0.8767 |
| 0.0 | 19.0 | 7125 | 1.0119 | 0.8867 |
| 0.0 | 20.0 | 7500 | 0.9792 | 0.8833 |
| 0.0 | 21.0 | 7875 | 1.0247 | 0.88 |
| 0.0 | 22.0 | 8250 | 1.0061 | 0.885 |
| 0.0 | 23.0 | 8625 | 1.0234 | 0.8867 |
| 0.0 | 24.0 | 9000 | 1.0734 | 0.8733 |
| 0.0 | 25.0 | 9375 | 1.0638 | 0.8867 |
| 0.0 | 26.0 | 9750 | 1.0711 | 0.88 |
| 0.0 | 27.0 | 10125 | 1.1175 | 0.88 |
| 0.0 | 28.0 | 10500 | 1.0879 | 0.8867 |
| 0.0 | 29.0 | 10875 | 1.1361 | 0.8817 |
| 0.0 | 30.0 | 11250 | 1.1028 | 0.89 |
| 0.0 | 31.0 | 11625 | 1.1478 | 0.8817 |
| 0.0 | 32.0 | 12000 | 1.1406 | 0.8833 |
| 0.0 | 33.0 | 12375 | 1.1490 | 0.8833 |
| 0.0 | 34.0 | 12750 | 1.1669 | 0.8817 |
| 0.0 | 35.0 | 13125 | 1.1635 | 0.8833 |
| 0.0 | 36.0 | 13500 | 1.1789 | 0.8817 |
| 0.0 | 37.0 | 13875 | 1.1756 | 0.8833 |
| 0.0029 | 38.0 | 14250 | 1.1808 | 0.8833 |
| 0.0 | 39.0 | 14625 | 1.1891 | 0.8833 |
| 0.0 | 40.0 | 15000 | 1.1976 | 0.8833 |
| 0.0 | 41.0 | 15375 | 1.2036 | 0.8817 |
| 0.0 | 42.0 | 15750 | 1.2058 | 0.88 |
| 0.0 | 43.0 | 16125 | 1.2107 | 0.8817 |
| 0.0 | 44.0 | 16500 | 1.2163 | 0.88 |
| 0.0 | 45.0 | 16875 | 1.2201 | 0.8783 |
| 0.0 | 46.0 | 17250 | 1.2238 | 0.8783 |
| 0.0 | 47.0 | 17625 | 1.2266 | 0.88 |
| 0.0 | 48.0 | 18000 | 1.2286 | 0.88 |
| 0.0 | 49.0 | 18375 | 1.2293 | 0.88 |
| 0.0 | 50.0 | 18750 | 1.2292 | 0.88 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.1+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2