smids_1x_deit_tiny_rms_0001_fold5
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: 1.2046
- Accuracy: 0.8517
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 |
|---|---|---|---|---|
| 0.7851 | 1.0 | 75 | 0.8659 | 0.64 |
| 0.6518 | 2.0 | 150 | 0.7541 | 0.6467 |
| 0.4507 | 3.0 | 225 | 0.6126 | 0.755 |
| 0.4597 | 4.0 | 300 | 0.4698 | 0.805 |
| 0.3528 | 5.0 | 375 | 0.4309 | 0.835 |
| 0.2717 | 6.0 | 450 | 0.4110 | 0.8517 |
| 0.2211 | 7.0 | 525 | 0.5132 | 0.8283 |
| 0.1873 | 8.0 | 600 | 0.5255 | 0.835 |
| 0.1509 | 9.0 | 675 | 0.5409 | 0.85 |
| 0.06 | 10.0 | 750 | 0.7466 | 0.8333 |
| 0.1297 | 11.0 | 825 | 0.8027 | 0.835 |
| 0.0789 | 12.0 | 900 | 0.7518 | 0.8417 |
| 0.1522 | 13.0 | 975 | 0.7901 | 0.8533 |
| 0.0628 | 14.0 | 1050 | 0.8326 | 0.845 |
| 0.0732 | 15.0 | 1125 | 0.9433 | 0.8317 |
| 0.0276 | 16.0 | 1200 | 0.9028 | 0.845 |
| 0.0402 | 17.0 | 1275 | 0.8882 | 0.8617 |
| 0.0561 | 18.0 | 1350 | 0.9516 | 0.8367 |
| 0.0072 | 19.0 | 1425 | 1.0341 | 0.8467 |
| 0.0251 | 20.0 | 1500 | 1.0436 | 0.8433 |
| 0.0171 | 21.0 | 1575 | 0.8887 | 0.855 |
| 0.0141 | 22.0 | 1650 | 0.9265 | 0.8517 |
| 0.0297 | 23.0 | 1725 | 1.1359 | 0.8383 |
| 0.0008 | 24.0 | 1800 | 1.0337 | 0.8567 |
| 0.0322 | 25.0 | 1875 | 0.8913 | 0.87 |
| 0.0416 | 26.0 | 1950 | 0.9175 | 0.84 |
| 0.0268 | 27.0 | 2025 | 0.9551 | 0.86 |
| 0.0237 | 28.0 | 2100 | 1.0150 | 0.8533 |
| 0.0252 | 29.0 | 2175 | 0.8872 | 0.8617 |
| 0.0035 | 30.0 | 2250 | 0.9489 | 0.8633 |
| 0.0155 | 31.0 | 2325 | 1.0473 | 0.8417 |
| 0.0007 | 32.0 | 2400 | 0.9648 | 0.8533 |
| 0.0102 | 33.0 | 2475 | 1.0603 | 0.8517 |
| 0.0 | 34.0 | 2550 | 1.0445 | 0.8533 |
| 0.0057 | 35.0 | 2625 | 1.0369 | 0.8467 |
| 0.0 | 36.0 | 2700 | 1.0577 | 0.8517 |
| 0.004 | 37.0 | 2775 | 1.0782 | 0.845 |
| 0.0033 | 38.0 | 2850 | 1.1658 | 0.8433 |
| 0.0001 | 39.0 | 2925 | 1.0942 | 0.8533 |
| 0.0 | 40.0 | 3000 | 1.1718 | 0.8467 |
| 0.0038 | 41.0 | 3075 | 1.1726 | 0.855 |
| 0.0 | 42.0 | 3150 | 1.1472 | 0.85 |
| 0.008 | 43.0 | 3225 | 1.1850 | 0.8517 |
| 0.0008 | 44.0 | 3300 | 1.1576 | 0.845 |
| 0.0022 | 45.0 | 3375 | 1.1935 | 0.855 |
| 0.0 | 46.0 | 3450 | 1.1973 | 0.8533 |
| 0.0056 | 47.0 | 3525 | 1.2032 | 0.8533 |
| 0.0051 | 48.0 | 3600 | 1.2041 | 0.8533 |
| 0.0 | 49.0 | 3675 | 1.2053 | 0.8517 |
| 0.0043 | 50.0 | 3750 | 1.2046 | 0.8517 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0