<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
wbc-50-pretrain-20-epochs
This model is a fine-tuned version of koobear/masked-50-pretraining-20-epoch on the image_folder dataset. It achieves the following results on the evaluation set:
- Loss: 0.1141
- Accuracy: 0.9826
- F1: 0.9827
- Precision: 0.9829
- Recall: 0.9826
- Balanced Acc: 0.9737
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: 2e-05
- train_batch_size: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall | Balanced Acc |
---|---|---|---|---|---|---|---|---|
1.2239 | 1.0 | 33 | 0.8645 | 0.9219 | 0.9268 | 0.9404 | 0.9219 | 0.9370 |
0.7434 | 2.0 | 66 | 0.5164 | 0.9549 | 0.9560 | 0.9589 | 0.9549 | 0.9614 |
0.4834 | 3.0 | 99 | 0.3333 | 0.9711 | 0.9713 | 0.9718 | 0.9711 | 0.9645 |
0.3492 | 4.0 | 132 | 0.2470 | 0.9797 | 0.9797 | 0.9797 | 0.9797 | 0.9683 |
0.292 | 5.0 | 165 | 0.2056 | 0.9792 | 0.9793 | 0.9795 | 0.9792 | 0.9758 |
0.2612 | 6.0 | 198 | 0.1960 | 0.9745 | 0.9746 | 0.9750 | 0.9745 | 0.9659 |
0.2424 | 7.0 | 231 | 0.1660 | 0.9792 | 0.9796 | 0.9809 | 0.9792 | 0.9801 |
0.2157 | 8.0 | 264 | 0.1528 | 0.9803 | 0.9805 | 0.9813 | 0.9803 | 0.9793 |
0.2087 | 9.0 | 297 | 0.1511 | 0.9774 | 0.9776 | 0.9781 | 0.9774 | 0.9763 |
0.1967 | 10.0 | 330 | 0.1359 | 0.9809 | 0.9810 | 0.9814 | 0.9809 | 0.9749 |
0.1851 | 11.0 | 363 | 0.1353 | 0.9821 | 0.9821 | 0.9821 | 0.9821 | 0.9743 |
0.1859 | 12.0 | 396 | 0.1317 | 0.9815 | 0.9815 | 0.9816 | 0.9815 | 0.9738 |
0.1679 | 13.0 | 429 | 0.1291 | 0.9815 | 0.9816 | 0.9820 | 0.9815 | 0.9769 |
0.1583 | 14.0 | 462 | 0.1195 | 0.9832 | 0.9833 | 0.9836 | 0.9832 | 0.9791 |
0.1429 | 15.0 | 495 | 0.1191 | 0.9815 | 0.9816 | 0.9819 | 0.9815 | 0.9749 |
0.1624 | 16.0 | 528 | 0.1177 | 0.9826 | 0.9828 | 0.9831 | 0.9826 | 0.9825 |
0.1515 | 17.0 | 561 | 0.1159 | 0.9821 | 0.9822 | 0.9824 | 0.9821 | 0.9781 |
0.1517 | 18.0 | 594 | 0.1158 | 0.9826 | 0.9827 | 0.9828 | 0.9826 | 0.9737 |
0.1389 | 19.0 | 627 | 0.1140 | 0.9832 | 0.9833 | 0.9835 | 0.9832 | 0.9758 |
0.1325 | 20.0 | 660 | 0.1141 | 0.9826 | 0.9827 | 0.9829 | 0.9826 | 0.9737 |
Framework versions
- Transformers 4.33.0
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.13.3