XLM_CITA
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: 0.5616
- Accuracy: 0.7705
- F1: 0.7698
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_with_restarts
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 0.6485 | 1.0 | 250 | 0.6020 | 0.6645 | 0.6490 |
| 0.5652 | 2.0 | 500 | 0.5210 | 0.7395 | 0.7397 |
| 0.5122 | 3.0 | 750 | 0.5111 | 0.7495 | 0.7496 |
| 0.4661 | 4.0 | 1000 | 0.5370 | 0.7685 | 0.7684 |
| 0.4244 | 5.0 | 1250 | 0.5206 | 0.7635 | 0.7636 |
| 0.3942 | 6.0 | 1500 | 0.5299 | 0.762 | 0.7621 |
| 0.3611 | 7.0 | 1750 | 0.5380 | 0.7695 | 0.7686 |
| 0.3421 | 8.0 | 2000 | 0.5595 | 0.7745 | 0.7736 |
| 0.3362 | 9.0 | 2250 | 0.5596 | 0.7715 | 0.7708 |
| 0.3274 | 10.0 | 2500 | 0.5616 | 0.7705 | 0.7698 |
Framework versions
- Transformers 4.48.0
- Pytorch 2.1.2
- Datasets 2.19.2
- Tokenizers 0.21.0