CS221-bert-base-uncased-finetuned-semeval-NT-amh
This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4393
- F1: 0.5978
- Roc Auc: 0.7647
- Accuracy: 0.5620
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 | Roc Auc | Accuracy |
|---|---|---|---|---|---|---|
| 0.2818 | 1.0 | 355 | 0.2680 | 0.4060 | 0.6633 | 0.5789 |
| 0.1748 | 2.0 | 710 | 0.2676 | 0.4931 | 0.6932 | 0.5648 |
| 0.1946 | 3.0 | 1065 | 0.2721 | 0.5730 | 0.7283 | 0.5746 |
| 0.0735 | 4.0 | 1420 | 0.3130 | 0.5593 | 0.7182 | 0.5789 |
| 0.046 | 5.0 | 1775 | 0.3659 | 0.5816 | 0.7418 | 0.5718 |
| 0.0135 | 6.0 | 2130 | 0.3934 | 0.5682 | 0.7352 | 0.5634 |
| 0.023 | 7.0 | 2485 | 0.4187 | 0.5784 | 0.7508 | 0.5465 |
| 0.0151 | 8.0 | 2840 | 0.4393 | 0.5978 | 0.7647 | 0.5620 |
| 0.0056 | 9.0 | 3195 | 0.4637 | 0.5602 | 0.7282 | 0.5648 |
| 0.0045 | 10.0 | 3550 | 0.4791 | 0.5931 | 0.7442 | 0.5662 |
| 0.0024 | 11.0 | 3905 | 0.4835 | 0.5886 | 0.7442 | 0.5592 |
Framework versions
- Transformers 4.47.1
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0