smids_3x_deit_tiny_rms_00001_fold4
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: 1.3270
- Accuracy: 0.8783
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 |
|---|---|---|---|---|
| 0.3199 | 1.0 | 225 | 0.3628 | 0.8533 |
| 0.1542 | 2.0 | 450 | 0.3942 | 0.8633 |
| 0.1186 | 3.0 | 675 | 0.4019 | 0.8783 |
| 0.1068 | 4.0 | 900 | 0.4521 | 0.8667 |
| 0.0629 | 5.0 | 1125 | 0.5068 | 0.8733 |
| 0.0347 | 6.0 | 1350 | 0.6175 | 0.8733 |
| 0.0294 | 7.0 | 1575 | 0.7043 | 0.8683 |
| 0.0692 | 8.0 | 1800 | 0.7322 | 0.8683 |
| 0.0709 | 9.0 | 2025 | 0.8957 | 0.8683 |
| 0.0668 | 10.0 | 2250 | 0.9618 | 0.8633 |
| 0.0043 | 11.0 | 2475 | 1.0931 | 0.855 |
| 0.0053 | 12.0 | 2700 | 1.0669 | 0.8617 |
| 0.0043 | 13.0 | 2925 | 1.2308 | 0.8533 |
| 0.0002 | 14.0 | 3150 | 1.1863 | 0.86 |
| 0.0021 | 15.0 | 3375 | 1.1504 | 0.8633 |
| 0.0 | 16.0 | 3600 | 1.1831 | 0.8683 |
| 0.0542 | 17.0 | 3825 | 1.2418 | 0.8633 |
| 0.0002 | 18.0 | 4050 | 1.2334 | 0.8567 |
| 0.0 | 19.0 | 4275 | 1.2760 | 0.8717 |
| 0.0018 | 20.0 | 4500 | 1.3796 | 0.8567 |
| 0.0 | 21.0 | 4725 | 1.2762 | 0.8683 |
| 0.0 | 22.0 | 4950 | 1.3847 | 0.8583 |
| 0.0001 | 23.0 | 5175 | 1.2718 | 0.8733 |
| 0.0 | 24.0 | 5400 | 1.2359 | 0.8717 |
| 0.0 | 25.0 | 5625 | 1.1935 | 0.8783 |
| 0.0 | 26.0 | 5850 | 1.2054 | 0.8783 |
| 0.0 | 27.0 | 6075 | 1.3706 | 0.86 |
| 0.0 | 28.0 | 6300 | 1.3639 | 0.8683 |
| 0.0 | 29.0 | 6525 | 1.2958 | 0.865 |
| 0.0 | 30.0 | 6750 | 1.3600 | 0.8617 |
| 0.0001 | 31.0 | 6975 | 1.4078 | 0.865 |
| 0.0 | 32.0 | 7200 | 1.3542 | 0.8667 |
| 0.0 | 33.0 | 7425 | 1.3120 | 0.865 |
| 0.0 | 34.0 | 7650 | 1.2757 | 0.8717 |
| 0.0 | 35.0 | 7875 | 1.2928 | 0.8667 |
| 0.0 | 36.0 | 8100 | 1.3460 | 0.8667 |
| 0.0002 | 37.0 | 8325 | 1.3775 | 0.8633 |
| 0.0 | 38.0 | 8550 | 1.3704 | 0.8617 |
| 0.0 | 39.0 | 8775 | 1.3338 | 0.87 |
| 0.0 | 40.0 | 9000 | 1.3261 | 0.8717 |
| 0.0 | 41.0 | 9225 | 1.3217 | 0.8717 |
| 0.0068 | 42.0 | 9450 | 1.3066 | 0.8733 |
| 0.0 | 43.0 | 9675 | 1.3141 | 0.875 |
| 0.0059 | 44.0 | 9900 | 1.3176 | 0.8767 |
| 0.0 | 45.0 | 10125 | 1.3464 | 0.865 |
| 0.0 | 46.0 | 10350 | 1.3263 | 0.8733 |
| 0.0 | 47.0 | 10575 | 1.3244 | 0.875 |
| 0.0 | 48.0 | 10800 | 1.3245 | 0.8767 |
| 0.0 | 49.0 | 11025 | 1.3277 | 0.8783 |
| 0.0 | 50.0 | 11250 | 1.3270 | 0.8783 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.1+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2