smids_10x_deit_small_adamax_00001_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: 1.3483
- Accuracy: 0.8783
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 |
|---|---|---|---|---|
| 0.2714 | 1.0 | 750 | 0.3483 | 0.875 |
| 0.1928 | 2.0 | 1500 | 0.3347 | 0.8833 |
| 0.1383 | 3.0 | 2250 | 0.3802 | 0.87 |
| 0.0835 | 4.0 | 3000 | 0.4083 | 0.8833 |
| 0.07 | 5.0 | 3750 | 0.4749 | 0.8833 |
| 0.0338 | 6.0 | 4500 | 0.5541 | 0.8767 |
| 0.0133 | 7.0 | 5250 | 0.6527 | 0.8783 |
| 0.0087 | 8.0 | 6000 | 0.7143 | 0.88 |
| 0.0145 | 9.0 | 6750 | 0.7738 | 0.88 |
| 0.0002 | 10.0 | 7500 | 0.8388 | 0.8767 |
| 0.0004 | 11.0 | 8250 | 0.9053 | 0.8817 |
| 0.0065 | 12.0 | 9000 | 0.9720 | 0.8783 |
| 0.0 | 13.0 | 9750 | 1.0304 | 0.8767 |
| 0.0 | 14.0 | 10500 | 1.0771 | 0.8717 |
| 0.0 | 15.0 | 11250 | 1.0764 | 0.8783 |
| 0.0326 | 16.0 | 12000 | 1.0955 | 0.8833 |
| 0.0001 | 17.0 | 12750 | 1.0921 | 0.8817 |
| 0.0 | 18.0 | 13500 | 1.1024 | 0.8817 |
| 0.0 | 19.0 | 14250 | 1.1225 | 0.8817 |
| 0.0 | 20.0 | 15000 | 1.1467 | 0.88 |
| 0.0 | 21.0 | 15750 | 1.1711 | 0.88 |
| 0.0 | 22.0 | 16500 | 1.1842 | 0.8783 |
| 0.0 | 23.0 | 17250 | 1.1878 | 0.8783 |
| 0.0 | 24.0 | 18000 | 1.2170 | 0.8817 |
| 0.0 | 25.0 | 18750 | 1.2183 | 0.88 |
| 0.0 | 26.0 | 19500 | 1.2367 | 0.88 |
| 0.0 | 27.0 | 20250 | 1.2535 | 0.8783 |
| 0.0 | 28.0 | 21000 | 1.2655 | 0.8833 |
| 0.0 | 29.0 | 21750 | 1.2701 | 0.8783 |
| 0.0 | 30.0 | 22500 | 1.2647 | 0.8783 |
| 0.0 | 31.0 | 23250 | 1.2884 | 0.8783 |
| 0.0 | 32.0 | 24000 | 1.2899 | 0.8733 |
| 0.0 | 33.0 | 24750 | 1.3073 | 0.8817 |
| 0.0 | 34.0 | 25500 | 1.3112 | 0.8833 |
| 0.0 | 35.0 | 26250 | 1.3094 | 0.8817 |
| 0.0 | 36.0 | 27000 | 1.3116 | 0.88 |
| 0.0 | 37.0 | 27750 | 1.3157 | 0.88 |
| 0.0 | 38.0 | 28500 | 1.3213 | 0.88 |
| 0.0 | 39.0 | 29250 | 1.3285 | 0.8767 |
| 0.0 | 40.0 | 30000 | 1.3297 | 0.8767 |
| 0.0 | 41.0 | 30750 | 1.3323 | 0.8783 |
| 0.0 | 42.0 | 31500 | 1.3346 | 0.8767 |
| 0.0 | 43.0 | 32250 | 1.3389 | 0.8783 |
| 0.0 | 44.0 | 33000 | 1.3404 | 0.8783 |
| 0.0 | 45.0 | 33750 | 1.3431 | 0.8783 |
| 0.0 | 46.0 | 34500 | 1.3453 | 0.8783 |
| 0.0 | 47.0 | 35250 | 1.3463 | 0.8783 |
| 0.0 | 48.0 | 36000 | 1.3478 | 0.8783 |
| 0.0 | 49.0 | 36750 | 1.3483 | 0.8783 |
| 0.0 | 50.0 | 37500 | 1.3483 | 0.8783 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2