smids_1x_deit_tiny_rms_00001_fold1
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: 0.8877
- Accuracy: 0.8631
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 |
|---|---|---|---|---|
| 0.4093 | 1.0 | 76 | 0.4066 | 0.8264 |
| 0.3822 | 2.0 | 152 | 0.3553 | 0.8614 |
| 0.1979 | 3.0 | 228 | 0.3399 | 0.8631 |
| 0.1648 | 4.0 | 304 | 0.3252 | 0.8815 |
| 0.0965 | 5.0 | 380 | 0.3551 | 0.8531 |
| 0.072 | 6.0 | 456 | 0.4036 | 0.8631 |
| 0.0292 | 7.0 | 532 | 0.4208 | 0.8598 |
| 0.0237 | 8.0 | 608 | 0.5314 | 0.8497 |
| 0.0407 | 9.0 | 684 | 0.5484 | 0.8497 |
| 0.0074 | 10.0 | 760 | 0.5780 | 0.8715 |
| 0.0366 | 11.0 | 836 | 0.5799 | 0.8631 |
| 0.0022 | 12.0 | 912 | 0.8054 | 0.8414 |
| 0.0514 | 13.0 | 988 | 0.5849 | 0.8748 |
| 0.0003 | 14.0 | 1064 | 0.6713 | 0.8664 |
| 0.0448 | 15.0 | 1140 | 0.6921 | 0.8715 |
| 0.0014 | 16.0 | 1216 | 0.6848 | 0.8631 |
| 0.0001 | 17.0 | 1292 | 0.7084 | 0.8648 |
| 0.0152 | 18.0 | 1368 | 0.8109 | 0.8681 |
| 0.0001 | 19.0 | 1444 | 0.7361 | 0.8698 |
| 0.004 | 20.0 | 1520 | 0.7743 | 0.8664 |
| 0.0035 | 21.0 | 1596 | 0.7272 | 0.8748 |
| 0.0282 | 22.0 | 1672 | 0.7515 | 0.8731 |
| 0.0001 | 23.0 | 1748 | 0.8060 | 0.8581 |
| 0.0001 | 24.0 | 1824 | 0.7763 | 0.8581 |
| 0.0156 | 25.0 | 1900 | 0.7302 | 0.8831 |
| 0.0068 | 26.0 | 1976 | 0.8763 | 0.8514 |
| 0.0045 | 27.0 | 2052 | 0.8144 | 0.8664 |
| 0.0058 | 28.0 | 2128 | 0.7716 | 0.8614 |
| 0.009 | 29.0 | 2204 | 0.8016 | 0.8664 |
| 0.0 | 30.0 | 2280 | 0.8234 | 0.8631 |
| 0.0087 | 31.0 | 2356 | 0.8420 | 0.8631 |
| 0.0102 | 32.0 | 2432 | 0.8218 | 0.8698 |
| 0.0 | 33.0 | 2508 | 0.8439 | 0.8564 |
| 0.0 | 34.0 | 2584 | 0.8448 | 0.8598 |
| 0.0154 | 35.0 | 2660 | 0.8638 | 0.8631 |
| 0.0044 | 36.0 | 2736 | 0.8664 | 0.8715 |
| 0.0088 | 37.0 | 2812 | 0.8649 | 0.8598 |
| 0.0 | 38.0 | 2888 | 0.8771 | 0.8598 |
| 0.0028 | 39.0 | 2964 | 0.8789 | 0.8631 |
| 0.0 | 40.0 | 3040 | 0.8645 | 0.8648 |
| 0.0044 | 41.0 | 3116 | 0.8681 | 0.8664 |
| 0.0 | 42.0 | 3192 | 0.8746 | 0.8631 |
| 0.0056 | 43.0 | 3268 | 0.8786 | 0.8664 |
| 0.0 | 44.0 | 3344 | 0.8858 | 0.8648 |
| 0.0 | 45.0 | 3420 | 0.8848 | 0.8648 |
| 0.0 | 46.0 | 3496 | 0.8858 | 0.8648 |
| 0.0 | 47.0 | 3572 | 0.8868 | 0.8631 |
| 0.0023 | 48.0 | 3648 | 0.8879 | 0.8631 |
| 0.0 | 49.0 | 3724 | 0.8884 | 0.8631 |
| 0.0 | 50.0 | 3800 | 0.8877 | 0.8631 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0