CS221-xlm-roberta-base-finetuned-semeval
This model is a fine-tuned version of FacebookAI/xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4974
- F1: 0.7079
- Roc Auc: 0.7818
- Accuracy: 0.3791
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 |
|---|---|---|---|---|---|---|
| 0.5792 | 1.0 | 277 | 0.5662 | 0.4622 | 0.6287 | 0.1516 |
| 0.4577 | 2.0 | 554 | 0.4709 | 0.6128 | 0.7167 | 0.3321 |
| 0.3852 | 3.0 | 831 | 0.4541 | 0.6619 | 0.7480 | 0.3628 |
| 0.3082 | 4.0 | 1108 | 0.4574 | 0.6739 | 0.7562 | 0.3809 |
| 0.2796 | 5.0 | 1385 | 0.4827 | 0.6949 | 0.7722 | 0.3664 |
| 0.2216 | 6.0 | 1662 | 0.4974 | 0.7079 | 0.7818 | 0.3791 |
| 0.1233 | 7.0 | 1939 | 0.5549 | 0.7032 | 0.7785 | 0.3736 |
| 0.0894 | 8.0 | 2216 | 0.6017 | 0.6754 | 0.7573 | 0.3718 |
| 0.07 | 9.0 | 2493 | 0.6214 | 0.7031 | 0.7779 | 0.3863 |
Framework versions
- Transformers 4.47.1
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0