smids_3x_deit_small_adamax_0001_fold1
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: 0.6646
- Accuracy: 0.9115
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 |
|---|---|---|---|---|
| 0.3694 | 1.0 | 226 | 0.3043 | 0.8765 |
| 0.1918 | 2.0 | 452 | 0.2837 | 0.8948 |
| 0.1232 | 3.0 | 678 | 0.5268 | 0.8464 |
| 0.0814 | 4.0 | 904 | 0.3209 | 0.9048 |
| 0.0289 | 5.0 | 1130 | 0.4828 | 0.8948 |
| 0.009 | 6.0 | 1356 | 0.6672 | 0.8932 |
| 0.057 | 7.0 | 1582 | 0.5442 | 0.9032 |
| 0.0008 | 8.0 | 1808 | 0.5498 | 0.9082 |
| 0.037 | 9.0 | 2034 | 0.4929 | 0.9182 |
| 0.0001 | 10.0 | 2260 | 0.5526 | 0.9082 |
| 0.0002 | 11.0 | 2486 | 0.5981 | 0.9048 |
| 0.0001 | 12.0 | 2712 | 0.5439 | 0.9115 |
| 0.0 | 13.0 | 2938 | 0.5693 | 0.9082 |
| 0.017 | 14.0 | 3164 | 0.5385 | 0.9149 |
| 0.003 | 15.0 | 3390 | 0.6151 | 0.9165 |
| 0.0 | 16.0 | 3616 | 0.5918 | 0.9149 |
| 0.0028 | 17.0 | 3842 | 0.5844 | 0.9115 |
| 0.0 | 18.0 | 4068 | 0.5936 | 0.9149 |
| 0.0042 | 19.0 | 4294 | 0.5972 | 0.9032 |
| 0.0 | 20.0 | 4520 | 0.5981 | 0.9098 |
| 0.0 | 21.0 | 4746 | 0.6811 | 0.9082 |
| 0.0168 | 22.0 | 4972 | 0.5972 | 0.9115 |
| 0.0028 | 23.0 | 5198 | 0.6127 | 0.9132 |
| 0.0 | 24.0 | 5424 | 0.6115 | 0.9149 |
| 0.0 | 25.0 | 5650 | 0.6222 | 0.9149 |
| 0.0 | 26.0 | 5876 | 0.6074 | 0.9215 |
| 0.0 | 27.0 | 6102 | 0.6129 | 0.9165 |
| 0.0 | 28.0 | 6328 | 0.6078 | 0.9165 |
| 0.0 | 29.0 | 6554 | 0.6211 | 0.9215 |
| 0.0 | 30.0 | 6780 | 0.6156 | 0.9149 |
| 0.0 | 31.0 | 7006 | 0.6329 | 0.9149 |
| 0.0 | 32.0 | 7232 | 0.6211 | 0.9215 |
| 0.0031 | 33.0 | 7458 | 0.6254 | 0.9165 |
| 0.0 | 34.0 | 7684 | 0.6575 | 0.9149 |
| 0.0 | 35.0 | 7910 | 0.6417 | 0.9132 |
| 0.0 | 36.0 | 8136 | 0.6653 | 0.9199 |
| 0.0 | 37.0 | 8362 | 0.6525 | 0.9149 |
| 0.0 | 38.0 | 8588 | 0.6394 | 0.9149 |
| 0.0 | 39.0 | 8814 | 0.6447 | 0.9165 |
| 0.0 | 40.0 | 9040 | 0.6473 | 0.9132 |
| 0.0027 | 41.0 | 9266 | 0.6511 | 0.9149 |
| 0.0028 | 42.0 | 9492 | 0.6536 | 0.9149 |
| 0.0 | 43.0 | 9718 | 0.6481 | 0.9149 |
| 0.0 | 44.0 | 9944 | 0.6547 | 0.9132 |
| 0.0 | 45.0 | 10170 | 0.6581 | 0.9132 |
| 0.0 | 46.0 | 10396 | 0.6606 | 0.9132 |
| 0.0 | 47.0 | 10622 | 0.6627 | 0.9132 |
| 0.0 | 48.0 | 10848 | 0.6635 | 0.9132 |
| 0.0 | 49.0 | 11074 | 0.6649 | 0.9115 |
| 0.0 | 50.0 | 11300 | 0.6646 | 0.9115 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2