xlnet-large-cased-finetuned-augmentation-LUNAR-TAPT-MICRO
This model is a fine-tuned version of xlnet/xlnet-large-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5171
- F1: 0.8326
- Roc Auc: 0.8759
- Accuracy: 0.6052
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 | Roc Auc | Accuracy |
|---|---|---|---|---|---|---|
| 0.4028 | 1.0 | 318 | 0.3883 | 0.7185 | 0.7928 | 0.4082 |
| 0.3124 | 2.0 | 636 | 0.3370 | 0.7842 | 0.8459 | 0.5051 |
| 0.2128 | 3.0 | 954 | 0.3211 | 0.7902 | 0.8422 | 0.5493 |
| 0.1418 | 4.0 | 1272 | 0.3509 | 0.8048 | 0.8562 | 0.5437 |
| 0.0971 | 5.0 | 1590 | 0.3839 | 0.8036 | 0.8524 | 0.5642 |
| 0.0538 | 6.0 | 1908 | 0.4280 | 0.8232 | 0.8744 | 0.5800 |
| 0.038 | 7.0 | 2226 | 0.4600 | 0.8251 | 0.8766 | 0.5831 |
| 0.0251 | 8.0 | 2544 | 0.4612 | 0.8184 | 0.8670 | 0.5808 |
| 0.0251 | 9.0 | 2862 | 0.5169 | 0.8228 | 0.8746 | 0.5729 |
| 0.016 | 10.0 | 3180 | 0.5178 | 0.8258 | 0.8731 | 0.5902 |
| 0.0119 | 11.0 | 3498 | 0.5177 | 0.8305 | 0.8757 | 0.5997 |
| 0.0063 | 12.0 | 3816 | 0.5171 | 0.8326 | 0.8759 | 0.6052 |
| 0.0021 | 13.0 | 4134 | 0.5385 | 0.8325 | 0.8792 | 0.5934 |
| 0.0012 | 14.0 | 4452 | 0.5370 | 0.8284 | 0.8736 | 0.5950 |
| 0.0039 | 15.0 | 4770 | 0.5462 | 0.8324 | 0.8786 | 0.5997 |
| 0.0027 | 16.0 | 5088 | 0.5458 | 0.8320 | 0.8767 | 0.6044 |
Framework versions
- Transformers 4.45.1
- Pytorch 2.4.0
- Datasets 3.0.1
- Tokenizers 0.20.0