cs221-roberta-large-eng-finetuned-20-epochs-tapt
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: 0.4201
- F1: 0.7943
- Roc Auc: 0.8477
- Accuracy: 0.5095
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: 32
- eval_batch_size: 32
- 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 |
|---|---|---|---|---|---|---|
| 0.5916 | 1.0 | 73 | 0.5657 | 0.4369 | 0.6142 | 0.1386 |
| 0.3931 | 2.0 | 146 | 0.3594 | 0.7503 | 0.8148 | 0.4367 |
| 0.316 | 3.0 | 219 | 0.3356 | 0.7838 | 0.8413 | 0.5078 |
| 0.2292 | 4.0 | 292 | 0.3380 | 0.7800 | 0.8377 | 0.4957 |
| 0.1583 | 5.0 | 365 | 0.3719 | 0.7663 | 0.8255 | 0.4801 |
| 0.1218 | 6.0 | 438 | 0.3702 | 0.7858 | 0.8383 | 0.4957 |
| 0.084 | 7.0 | 511 | 0.3979 | 0.7878 | 0.8396 | 0.5095 |
| 0.0654 | 8.0 | 584 | 0.4201 | 0.7943 | 0.8477 | 0.5095 |
| 0.0428 | 9.0 | 657 | 0.4617 | 0.7834 | 0.8374 | 0.4957 |
| 0.0311 | 10.0 | 730 | 0.4898 | 0.7803 | 0.8353 | 0.4853 |
| 0.0249 | 11.0 | 803 | 0.4736 | 0.7859 | 0.8390 | 0.4957 |
Framework versions
- Transformers 4.48.0
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0