smids_1x_deit_small_rms_00001_fold3
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: 0.7182
- Accuracy: 0.905
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 |
|---|---|---|---|---|
| 0.3259 | 1.0 | 75 | 0.3001 | 0.89 |
| 0.2426 | 2.0 | 150 | 0.3217 | 0.8717 |
| 0.1676 | 3.0 | 225 | 0.2596 | 0.9083 |
| 0.1287 | 4.0 | 300 | 0.2827 | 0.895 |
| 0.0316 | 5.0 | 375 | 0.3452 | 0.885 |
| 0.0237 | 6.0 | 450 | 0.3793 | 0.9017 |
| 0.0244 | 7.0 | 525 | 0.4128 | 0.8967 |
| 0.0233 | 8.0 | 600 | 0.4590 | 0.8883 |
| 0.0286 | 9.0 | 675 | 0.4790 | 0.8983 |
| 0.0295 | 10.0 | 750 | 0.4835 | 0.8917 |
| 0.0562 | 11.0 | 825 | 0.4705 | 0.9067 |
| 0.0087 | 12.0 | 900 | 0.5035 | 0.9033 |
| 0.0083 | 13.0 | 975 | 0.5418 | 0.9017 |
| 0.0001 | 14.0 | 1050 | 0.5563 | 0.9 |
| 0.0012 | 15.0 | 1125 | 0.5874 | 0.8983 |
| 0.0001 | 16.0 | 1200 | 0.5698 | 0.8967 |
| 0.0001 | 17.0 | 1275 | 0.5930 | 0.9033 |
| 0.0062 | 18.0 | 1350 | 0.5972 | 0.9017 |
| 0.0048 | 19.0 | 1425 | 0.5918 | 0.9033 |
| 0.0089 | 20.0 | 1500 | 0.6518 | 0.9017 |
| 0.0001 | 21.0 | 1575 | 0.7835 | 0.885 |
| 0.0001 | 22.0 | 1650 | 0.6700 | 0.9 |
| 0.0031 | 23.0 | 1725 | 0.6679 | 0.8983 |
| 0.0 | 24.0 | 1800 | 0.6364 | 0.9033 |
| 0.0001 | 25.0 | 1875 | 0.6464 | 0.8983 |
| 0.003 | 26.0 | 1950 | 0.6535 | 0.8967 |
| 0.0 | 27.0 | 2025 | 0.6525 | 0.8983 |
| 0.0 | 28.0 | 2100 | 0.6526 | 0.8983 |
| 0.0 | 29.0 | 2175 | 0.6663 | 0.895 |
| 0.0 | 30.0 | 2250 | 0.6645 | 0.8983 |
| 0.0 | 31.0 | 2325 | 0.6717 | 0.9 |
| 0.0 | 32.0 | 2400 | 0.6659 | 0.8983 |
| 0.0 | 33.0 | 2475 | 0.6774 | 0.9017 |
| 0.0051 | 34.0 | 2550 | 0.6726 | 0.905 |
| 0.0059 | 35.0 | 2625 | 0.7209 | 0.8933 |
| 0.0031 | 36.0 | 2700 | 0.6818 | 0.9067 |
| 0.0022 | 37.0 | 2775 | 0.6938 | 0.8967 |
| 0.0 | 38.0 | 2850 | 0.6968 | 0.8967 |
| 0.0 | 39.0 | 2925 | 0.7122 | 0.8983 |
| 0.0 | 40.0 | 3000 | 0.7008 | 0.8983 |
| 0.0 | 41.0 | 3075 | 0.7070 | 0.8983 |
| 0.0026 | 42.0 | 3150 | 0.7002 | 0.9 |
| 0.0025 | 43.0 | 3225 | 0.7107 | 0.9 |
| 0.0 | 44.0 | 3300 | 0.7106 | 0.9033 |
| 0.0025 | 45.0 | 3375 | 0.7116 | 0.905 |
| 0.0025 | 46.0 | 3450 | 0.7142 | 0.905 |
| 0.0047 | 47.0 | 3525 | 0.7163 | 0.9033 |
| 0.0 | 48.0 | 3600 | 0.7169 | 0.9033 |
| 0.0 | 49.0 | 3675 | 0.7178 | 0.9033 |
| 0.0045 | 50.0 | 3750 | 0.7182 | 0.905 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0