smids_5x_deit_tiny_adamax_0001_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: 0.9118
- Accuracy: 0.8983
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 |
|---|---|---|---|---|
| 0.2637 | 1.0 | 375 | 0.4028 | 0.845 |
| 0.1351 | 2.0 | 750 | 0.3297 | 0.8967 |
| 0.0831 | 3.0 | 1125 | 0.4271 | 0.895 |
| 0.0388 | 4.0 | 1500 | 0.5488 | 0.8933 |
| 0.0329 | 5.0 | 1875 | 0.5460 | 0.91 |
| 0.015 | 6.0 | 2250 | 0.7045 | 0.8983 |
| 0.0097 | 7.0 | 2625 | 0.6091 | 0.9083 |
| 0.0008 | 8.0 | 3000 | 0.6613 | 0.9 |
| 0.0051 | 9.0 | 3375 | 0.6977 | 0.9 |
| 0.0095 | 10.0 | 3750 | 0.7759 | 0.8983 |
| 0.0182 | 11.0 | 4125 | 0.8338 | 0.8983 |
| 0.012 | 12.0 | 4500 | 0.7866 | 0.91 |
| 0.0163 | 13.0 | 4875 | 0.7705 | 0.9067 |
| 0.0 | 14.0 | 5250 | 0.8292 | 0.9033 |
| 0.0 | 15.0 | 5625 | 0.7633 | 0.9167 |
| 0.0002 | 16.0 | 6000 | 0.7628 | 0.9033 |
| 0.0 | 17.0 | 6375 | 0.8121 | 0.91 |
| 0.0 | 18.0 | 6750 | 0.7784 | 0.905 |
| 0.0 | 19.0 | 7125 | 0.8065 | 0.9067 |
| 0.0 | 20.0 | 7500 | 0.8530 | 0.9083 |
| 0.0 | 21.0 | 7875 | 0.7859 | 0.9083 |
| 0.0 | 22.0 | 8250 | 0.7653 | 0.9133 |
| 0.0079 | 23.0 | 8625 | 0.8464 | 0.91 |
| 0.0 | 24.0 | 9000 | 0.8244 | 0.9067 |
| 0.0 | 25.0 | 9375 | 0.8204 | 0.9067 |
| 0.0 | 26.0 | 9750 | 0.8559 | 0.91 |
| 0.0 | 27.0 | 10125 | 0.8371 | 0.9117 |
| 0.0 | 28.0 | 10500 | 0.8892 | 0.9067 |
| 0.0 | 29.0 | 10875 | 0.8646 | 0.9083 |
| 0.0054 | 30.0 | 11250 | 0.8291 | 0.9083 |
| 0.0 | 31.0 | 11625 | 0.8780 | 0.895 |
| 0.0 | 32.0 | 12000 | 0.8742 | 0.9033 |
| 0.0 | 33.0 | 12375 | 0.8764 | 0.9017 |
| 0.0 | 34.0 | 12750 | 0.8694 | 0.9017 |
| 0.0 | 35.0 | 13125 | 0.8829 | 0.9017 |
| 0.0 | 36.0 | 13500 | 0.8880 | 0.905 |
| 0.0 | 37.0 | 13875 | 0.8817 | 0.9 |
| 0.0 | 38.0 | 14250 | 0.8874 | 0.9033 |
| 0.0 | 39.0 | 14625 | 0.8863 | 0.8983 |
| 0.0028 | 40.0 | 15000 | 0.8961 | 0.9017 |
| 0.0 | 41.0 | 15375 | 0.8936 | 0.9017 |
| 0.0 | 42.0 | 15750 | 0.9013 | 0.9017 |
| 0.0 | 43.0 | 16125 | 0.9020 | 0.8983 |
| 0.0 | 44.0 | 16500 | 0.9052 | 0.9 |
| 0.0 | 45.0 | 16875 | 0.9070 | 0.9 |
| 0.0 | 46.0 | 17250 | 0.9083 | 0.9 |
| 0.0 | 47.0 | 17625 | 0.9101 | 0.9 |
| 0.0 | 48.0 | 18000 | 0.9113 | 0.8983 |
| 0.0 | 49.0 | 18375 | 0.9121 | 0.8983 |
| 0.0 | 50.0 | 18750 | 0.9118 | 0.8983 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.1+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2