CS221-roberta-large-finetuned-semeval-NT
This model is a fine-tuned version of FacebookAI/roberta-large on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6250
- F1: 0.7461
- Roc Auc: 0.8116
- Accuracy: 0.4657
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 |
|---|---|---|---|---|---|---|
| 0.4247 | 1.0 | 277 | 0.3771 | 0.6157 | 0.7185 | 0.4025 |
| 0.3123 | 2.0 | 554 | 0.3756 | 0.6707 | 0.7597 | 0.4495 |
| 0.2477 | 3.0 | 831 | 0.3577 | 0.7215 | 0.7856 | 0.4982 |
| 0.153 | 4.0 | 1108 | 0.4303 | 0.7345 | 0.8017 | 0.4711 |
| 0.0938 | 5.0 | 1385 | 0.4975 | 0.7334 | 0.7961 | 0.4657 |
| 0.0761 | 6.0 | 1662 | 0.5342 | 0.7427 | 0.8027 | 0.4819 |
| 0.0475 | 7.0 | 1939 | 0.5857 | 0.7441 | 0.7987 | 0.4458 |
| 0.0165 | 8.0 | 2216 | 0.6250 | 0.7461 | 0.8116 | 0.4657 |
| 0.0077 | 9.0 | 2493 | 0.6812 | 0.7355 | 0.7937 | 0.4567 |
| 0.0065 | 10.0 | 2770 | 0.6681 | 0.7368 | 0.7974 | 0.4874 |
| 0.0093 | 11.0 | 3047 | 0.7421 | 0.7393 | 0.7981 | 0.4603 |
Framework versions
- Transformers 4.47.1
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0