smids_10x_deit_small_rms_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: 1.5041
- Accuracy: 0.8817
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 |
|---|---|---|---|---|
| 0.1992 | 1.0 | 750 | 0.3642 | 0.8683 |
| 0.1164 | 2.0 | 1500 | 0.4047 | 0.88 |
| 0.0544 | 3.0 | 2250 | 0.5775 | 0.8733 |
| 0.0424 | 4.0 | 3000 | 0.7191 | 0.8717 |
| 0.0428 | 5.0 | 3750 | 0.8609 | 0.88 |
| 0.0178 | 6.0 | 4500 | 0.9003 | 0.8767 |
| 0.002 | 7.0 | 5250 | 0.9741 | 0.8883 |
| 0.0106 | 8.0 | 6000 | 1.0810 | 0.88 |
| 0.015 | 9.0 | 6750 | 1.0442 | 0.8767 |
| 0.0 | 10.0 | 7500 | 1.1184 | 0.875 |
| 0.0002 | 11.0 | 8250 | 1.1373 | 0.88 |
| 0.0 | 12.0 | 9000 | 1.0569 | 0.8833 |
| 0.0004 | 13.0 | 9750 | 1.1443 | 0.8783 |
| 0.0211 | 14.0 | 10500 | 1.1371 | 0.8717 |
| 0.0 | 15.0 | 11250 | 1.1887 | 0.8717 |
| 0.074 | 16.0 | 12000 | 1.0864 | 0.8767 |
| 0.0 | 17.0 | 12750 | 1.1543 | 0.8817 |
| 0.0061 | 18.0 | 13500 | 1.0371 | 0.8833 |
| 0.0 | 19.0 | 14250 | 1.2499 | 0.875 |
| 0.006 | 20.0 | 15000 | 1.1799 | 0.8733 |
| 0.0265 | 21.0 | 15750 | 1.2064 | 0.8783 |
| 0.0 | 22.0 | 16500 | 1.2139 | 0.875 |
| 0.0 | 23.0 | 17250 | 1.2392 | 0.8833 |
| 0.0 | 24.0 | 18000 | 1.3220 | 0.8717 |
| 0.0002 | 25.0 | 18750 | 1.3064 | 0.8717 |
| 0.0 | 26.0 | 19500 | 1.3677 | 0.8767 |
| 0.0 | 27.0 | 20250 | 1.2718 | 0.89 |
| 0.0 | 28.0 | 21000 | 1.2639 | 0.8783 |
| 0.0 | 29.0 | 21750 | 1.3047 | 0.8817 |
| 0.0 | 30.0 | 22500 | 1.3577 | 0.885 |
| 0.0 | 31.0 | 23250 | 1.3501 | 0.88 |
| 0.0 | 32.0 | 24000 | 1.3749 | 0.875 |
| 0.0 | 33.0 | 24750 | 1.4487 | 0.875 |
| 0.0 | 34.0 | 25500 | 1.3473 | 0.8867 |
| 0.0 | 35.0 | 26250 | 1.4090 | 0.885 |
| 0.0 | 36.0 | 27000 | 1.4862 | 0.8733 |
| 0.0 | 37.0 | 27750 | 1.3778 | 0.8783 |
| 0.0 | 38.0 | 28500 | 1.4730 | 0.8767 |
| 0.0 | 39.0 | 29250 | 1.4305 | 0.89 |
| 0.0 | 40.0 | 30000 | 1.4453 | 0.8833 |
| 0.0 | 41.0 | 30750 | 1.4782 | 0.8833 |
| 0.0 | 42.0 | 31500 | 1.4838 | 0.88 |
| 0.0 | 43.0 | 32250 | 1.4721 | 0.8867 |
| 0.0 | 44.0 | 33000 | 1.4749 | 0.8867 |
| 0.0 | 45.0 | 33750 | 1.4889 | 0.8867 |
| 0.0 | 46.0 | 34500 | 1.4951 | 0.885 |
| 0.0 | 47.0 | 35250 | 1.4970 | 0.8833 |
| 0.0 | 48.0 | 36000 | 1.4998 | 0.8833 |
| 0.0 | 49.0 | 36750 | 1.5024 | 0.8833 |
| 0.0 | 50.0 | 37500 | 1.5041 | 0.8817 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2