CS221-roberta-base-finetuned-semeval-NT
This model is a fine-tuned version of FacebookAI/roberta-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5269
- F1: 0.7240
- Roc Auc: 0.7938
- Accuracy: 0.4639
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 |
|---|---|---|---|---|---|---|
| 0.4269 | 1.0 | 277 | 0.3920 | 0.6532 | 0.7374 | 0.4025 |
| 0.3157 | 2.0 | 554 | 0.3683 | 0.6965 | 0.7692 | 0.4152 |
| 0.2287 | 3.0 | 831 | 0.3818 | 0.6849 | 0.7667 | 0.4314 |
| 0.1779 | 4.0 | 1108 | 0.4116 | 0.6927 | 0.7689 | 0.4097 |
| 0.1274 | 5.0 | 1385 | 0.4471 | 0.6991 | 0.7729 | 0.4314 |
| 0.1036 | 6.0 | 1662 | 0.4658 | 0.7166 | 0.7848 | 0.4549 |
| 0.0684 | 7.0 | 1939 | 0.5065 | 0.7133 | 0.7840 | 0.4422 |
| 0.055 | 8.0 | 2216 | 0.5269 | 0.7240 | 0.7938 | 0.4639 |
| 0.0156 | 9.0 | 2493 | 0.5896 | 0.7157 | 0.7920 | 0.4513 |
| 0.0173 | 10.0 | 2770 | 0.6118 | 0.7215 | 0.7868 | 0.4477 |
| 0.0171 | 11.0 | 3047 | 0.6322 | 0.7234 | 0.7941 | 0.4513 |
Framework versions
- Transformers 4.47.1
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0