smids_1x_deit_tiny_rms_0001_fold2
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: 1.4112
- Accuracy: 0.8569
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 |
|---|---|---|---|---|
| 0.8146 | 1.0 | 75 | 0.7861 | 0.5524 |
| 0.5727 | 2.0 | 150 | 0.4984 | 0.8103 |
| 0.5359 | 3.0 | 225 | 0.4227 | 0.8286 |
| 0.4099 | 4.0 | 300 | 0.4531 | 0.8270 |
| 0.3762 | 5.0 | 375 | 0.4703 | 0.8236 |
| 0.2518 | 6.0 | 450 | 0.5236 | 0.8453 |
| 0.2143 | 7.0 | 525 | 0.5529 | 0.8353 |
| 0.1716 | 8.0 | 600 | 0.6424 | 0.8519 |
| 0.1064 | 9.0 | 675 | 1.0458 | 0.7987 |
| 0.1362 | 10.0 | 750 | 0.7319 | 0.8336 |
| 0.1448 | 11.0 | 825 | 0.9729 | 0.8236 |
| 0.0254 | 12.0 | 900 | 0.9267 | 0.8436 |
| 0.0822 | 13.0 | 975 | 1.0041 | 0.8336 |
| 0.0792 | 14.0 | 1050 | 1.1093 | 0.8220 |
| 0.0861 | 15.0 | 1125 | 1.1399 | 0.8153 |
| 0.049 | 16.0 | 1200 | 1.3759 | 0.8103 |
| 0.0209 | 17.0 | 1275 | 1.1868 | 0.8303 |
| 0.0314 | 18.0 | 1350 | 1.3024 | 0.8353 |
| 0.0371 | 19.0 | 1425 | 1.1958 | 0.8303 |
| 0.0408 | 20.0 | 1500 | 1.0595 | 0.8469 |
| 0.0443 | 21.0 | 1575 | 1.2918 | 0.8353 |
| 0.0161 | 22.0 | 1650 | 1.3270 | 0.8270 |
| 0.002 | 23.0 | 1725 | 1.3561 | 0.8369 |
| 0.0119 | 24.0 | 1800 | 1.3471 | 0.8353 |
| 0.021 | 25.0 | 1875 | 1.3114 | 0.8403 |
| 0.0001 | 26.0 | 1950 | 1.2789 | 0.8453 |
| 0.0215 | 27.0 | 2025 | 1.3801 | 0.8253 |
| 0.0117 | 28.0 | 2100 | 1.3311 | 0.8353 |
| 0.0064 | 29.0 | 2175 | 1.5354 | 0.8153 |
| 0.0497 | 30.0 | 2250 | 1.2007 | 0.8419 |
| 0.0245 | 31.0 | 2325 | 1.2452 | 0.8586 |
| 0.0 | 32.0 | 2400 | 1.2980 | 0.8586 |
| 0.0 | 33.0 | 2475 | 1.3038 | 0.8586 |
| 0.0 | 34.0 | 2550 | 1.3062 | 0.8552 |
| 0.0104 | 35.0 | 2625 | 1.3421 | 0.8519 |
| 0.0001 | 36.0 | 2700 | 1.3682 | 0.8369 |
| 0.0027 | 37.0 | 2775 | 1.4409 | 0.8419 |
| 0.0 | 38.0 | 2850 | 1.3923 | 0.8519 |
| 0.0017 | 39.0 | 2925 | 1.4064 | 0.8536 |
| 0.0 | 40.0 | 3000 | 1.4003 | 0.8519 |
| 0.0027 | 41.0 | 3075 | 1.4111 | 0.8519 |
| 0.0 | 42.0 | 3150 | 1.4021 | 0.8519 |
| 0.0025 | 43.0 | 3225 | 1.4193 | 0.8519 |
| 0.0029 | 44.0 | 3300 | 1.3989 | 0.8552 |
| 0.0 | 45.0 | 3375 | 1.4257 | 0.8536 |
| 0.0 | 46.0 | 3450 | 1.4244 | 0.8536 |
| 0.0027 | 47.0 | 3525 | 1.4185 | 0.8536 |
| 0.0 | 48.0 | 3600 | 1.4177 | 0.8569 |
| 0.0023 | 49.0 | 3675 | 1.4124 | 0.8569 |
| 0.0021 | 50.0 | 3750 | 1.4112 | 0.8569 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0