smids_1x_deit_tiny_rms_00001_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: 0.8945
- Accuracy: 0.8683
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 |
|---|---|---|---|---|
| 0.4595 | 1.0 | 75 | 0.3430 | 0.875 |
| 0.3075 | 2.0 | 150 | 0.3392 | 0.8667 |
| 0.1885 | 3.0 | 225 | 0.2939 | 0.8833 |
| 0.2401 | 4.0 | 300 | 0.3047 | 0.88 |
| 0.1913 | 5.0 | 375 | 0.2948 | 0.8833 |
| 0.0779 | 6.0 | 450 | 0.3840 | 0.8733 |
| 0.0659 | 7.0 | 525 | 0.3753 | 0.8867 |
| 0.0683 | 8.0 | 600 | 0.4413 | 0.88 |
| 0.0365 | 9.0 | 675 | 0.5012 | 0.875 |
| 0.0284 | 10.0 | 750 | 0.5436 | 0.8783 |
| 0.0339 | 11.0 | 825 | 0.6040 | 0.8883 |
| 0.0022 | 12.0 | 900 | 0.6211 | 0.8767 |
| 0.0421 | 13.0 | 975 | 0.7125 | 0.87 |
| 0.0295 | 14.0 | 1050 | 0.8588 | 0.8567 |
| 0.0522 | 15.0 | 1125 | 0.7259 | 0.875 |
| 0.0002 | 16.0 | 1200 | 0.7288 | 0.87 |
| 0.0239 | 17.0 | 1275 | 0.7745 | 0.865 |
| 0.0349 | 18.0 | 1350 | 0.7880 | 0.8667 |
| 0.0099 | 19.0 | 1425 | 0.7942 | 0.8683 |
| 0.0002 | 20.0 | 1500 | 0.8016 | 0.88 |
| 0.0181 | 21.0 | 1575 | 0.8187 | 0.87 |
| 0.0067 | 22.0 | 1650 | 0.8802 | 0.87 |
| 0.0135 | 23.0 | 1725 | 0.8554 | 0.865 |
| 0.0001 | 24.0 | 1800 | 0.8527 | 0.8733 |
| 0.0254 | 25.0 | 1875 | 0.7951 | 0.87 |
| 0.0001 | 26.0 | 1950 | 0.8033 | 0.87 |
| 0.0127 | 27.0 | 2025 | 0.8239 | 0.8783 |
| 0.0152 | 28.0 | 2100 | 0.9169 | 0.8683 |
| 0.019 | 29.0 | 2175 | 0.8220 | 0.8767 |
| 0.0038 | 30.0 | 2250 | 0.8836 | 0.8683 |
| 0.0001 | 31.0 | 2325 | 0.8686 | 0.875 |
| 0.0 | 32.0 | 2400 | 0.8778 | 0.8717 |
| 0.0121 | 33.0 | 2475 | 1.0279 | 0.8583 |
| 0.0001 | 34.0 | 2550 | 0.9813 | 0.8583 |
| 0.0072 | 35.0 | 2625 | 0.8647 | 0.8767 |
| 0.0 | 36.0 | 2700 | 0.8869 | 0.8683 |
| 0.0027 | 37.0 | 2775 | 0.9010 | 0.865 |
| 0.0 | 38.0 | 2850 | 0.9248 | 0.8633 |
| 0.0 | 39.0 | 2925 | 0.9337 | 0.8617 |
| 0.0 | 40.0 | 3000 | 0.8918 | 0.865 |
| 0.0054 | 41.0 | 3075 | 0.9181 | 0.865 |
| 0.0 | 42.0 | 3150 | 0.8813 | 0.8717 |
| 0.0078 | 43.0 | 3225 | 0.9054 | 0.8633 |
| 0.0 | 44.0 | 3300 | 0.8996 | 0.8667 |
| 0.0016 | 45.0 | 3375 | 0.9082 | 0.8633 |
| 0.0 | 46.0 | 3450 | 0.9083 | 0.865 |
| 0.0068 | 47.0 | 3525 | 0.9001 | 0.8683 |
| 0.0048 | 48.0 | 3600 | 0.8991 | 0.8683 |
| 0.0 | 49.0 | 3675 | 0.8976 | 0.8683 |
| 0.0029 | 50.0 | 3750 | 0.8945 | 0.8683 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0