hushem_5x_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: 1.1081
- Accuracy: 0.8293
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 |
|---|---|---|---|---|
| 0.957 | 1.0 | 28 | 0.7236 | 0.7073 |
| 0.3642 | 2.0 | 56 | 0.5185 | 0.8049 |
| 0.1944 | 3.0 | 84 | 0.5546 | 0.8049 |
| 0.0826 | 4.0 | 112 | 0.7838 | 0.7561 |
| 0.027 | 5.0 | 140 | 0.5372 | 0.8049 |
| 0.0125 | 6.0 | 168 | 0.5869 | 0.8293 |
| 0.0034 | 7.0 | 196 | 0.7015 | 0.8293 |
| 0.0012 | 8.0 | 224 | 0.6670 | 0.8049 |
| 0.0008 | 9.0 | 252 | 0.6919 | 0.8293 |
| 0.0006 | 10.0 | 280 | 0.7125 | 0.8293 |
| 0.0004 | 11.0 | 308 | 0.7267 | 0.8293 |
| 0.0004 | 12.0 | 336 | 0.7569 | 0.8293 |
| 0.0003 | 13.0 | 364 | 0.7526 | 0.8293 |
| 0.0003 | 14.0 | 392 | 0.7915 | 0.8293 |
| 0.0002 | 15.0 | 420 | 0.8002 | 0.8293 |
| 0.0002 | 16.0 | 448 | 0.8251 | 0.8293 |
| 0.0002 | 17.0 | 476 | 0.8438 | 0.8293 |
| 0.0001 | 18.0 | 504 | 0.8466 | 0.8293 |
| 0.0001 | 19.0 | 532 | 0.8704 | 0.8293 |
| 0.0001 | 20.0 | 560 | 0.8762 | 0.8293 |
| 0.0001 | 21.0 | 588 | 0.8972 | 0.8293 |
| 0.0001 | 22.0 | 616 | 0.8987 | 0.8293 |
| 0.0001 | 23.0 | 644 | 0.9318 | 0.8293 |
| 0.0001 | 24.0 | 672 | 0.9238 | 0.8293 |
| 0.0001 | 25.0 | 700 | 0.9169 | 0.8293 |
| 0.0 | 26.0 | 728 | 0.9411 | 0.8293 |
| 0.0 | 27.0 | 756 | 0.9447 | 0.8293 |
| 0.0 | 28.0 | 784 | 0.9671 | 0.8293 |
| 0.0 | 29.0 | 812 | 0.9709 | 0.8293 |
| 0.0 | 30.0 | 840 | 0.9844 | 0.8293 |
| 0.0 | 31.0 | 868 | 0.9959 | 0.8293 |
| 0.0 | 32.0 | 896 | 1.0060 | 0.8293 |
| 0.0 | 33.0 | 924 | 1.0055 | 0.8293 |
| 0.0 | 34.0 | 952 | 1.0143 | 0.8293 |
| 0.0 | 35.0 | 980 | 1.0276 | 0.8293 |
| 0.0 | 36.0 | 1008 | 1.0321 | 0.8293 |
| 0.0 | 37.0 | 1036 | 1.0476 | 0.8293 |
| 0.0 | 38.0 | 1064 | 1.0409 | 0.8293 |
| 0.0 | 39.0 | 1092 | 1.0558 | 0.8293 |
| 0.0 | 40.0 | 1120 | 1.0678 | 0.8293 |
| 0.0 | 41.0 | 1148 | 1.0832 | 0.8293 |
| 0.0 | 42.0 | 1176 | 1.0928 | 0.8293 |
| 0.0 | 43.0 | 1204 | 1.0842 | 0.8293 |
| 0.0 | 44.0 | 1232 | 1.0881 | 0.8293 |
| 0.0 | 45.0 | 1260 | 1.0924 | 0.8293 |
| 0.0 | 46.0 | 1288 | 1.1046 | 0.8293 |
| 0.0 | 47.0 | 1316 | 1.1089 | 0.8293 |
| 0.0 | 48.0 | 1344 | 1.1085 | 0.8293 |
| 0.0 | 49.0 | 1372 | 1.1081 | 0.8293 |
| 0.0 | 50.0 | 1400 | 1.1081 | 0.8293 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0