results
This model is a fine-tuned version of xlm-roberta-large on the Tass dataset. It achieves the following results on the evaluation set:
- Loss: 0.8187
- Accuracy: 0.6848
- F1: 0.6754
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: 1e-05
- train_batch_size: 4
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 0.8864 | 1.0 | 300 | 0.9594 | 0.4893 | 0.3957 |
| 0.7847 | 2.0 | 600 | 0.7880 | 0.6525 | 0.6432 |
| 0.6791 | 3.0 | 900 | 0.7441 | 0.6709 | 0.6646 |
| 0.5428 | 4.0 | 1200 | 0.8187 | 0.6848 | 0.6754 |
| 0.3746 | 5.0 | 1500 | 0.9587 | 0.6807 | 0.6729 |
| 0.3164 | 6.0 | 1800 | 1.0413 | 0.6713 | 0.6685 |
Framework versions
- Transformers 4.56.1
- Pytorch 2.10.0+cu128
- Datasets 5.0.0
- Tokenizers 0.22.2