multilingual-xlm-roberta-large-hausa-ner-v1
This model is a fine-tuned version of xlm-roberta-large on the Beijuka/Multilingual_PII_NER_dataset dataset. It achieves the following results on the evaluation set:
- Loss: 0.0660
- Precision: 0.9485
- Recall: 0.9365
- F1: 0.9425
- Accuracy: 0.9859
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: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- 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: linear
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 301 | 0.1221 | 0.9236 | 0.7388 | 0.8209 | 0.9584 |
| 0.1846 | 2.0 | 602 | 0.1144 | 0.8662 | 0.9155 | 0.8902 | 0.9703 |
| 0.1846 | 3.0 | 903 | 0.1049 | 0.8875 | 0.9173 | 0.9022 | 0.9735 |
| 0.075 | 4.0 | 1204 | 0.1182 | 0.8879 | 0.9257 | 0.9064 | 0.9760 |
| 0.0549 | 5.0 | 1505 | 0.0941 | 0.9009 | 0.9317 | 0.9161 | 0.9738 |
| 0.0549 | 6.0 | 1806 | 0.0916 | 0.9224 | 0.9395 | 0.9308 | 0.9808 |
| 0.0363 | 7.0 | 2107 | 0.1270 | 0.8856 | 0.9365 | 0.9103 | 0.9758 |
| 0.0363 | 8.0 | 2408 | 0.1093 | 0.8942 | 0.9521 | 0.9222 | 0.9778 |
| 0.0265 | 9.0 | 2709 | 0.1083 | 0.9191 | 0.9389 | 0.9289 | 0.9808 |
Framework versions
- Transformers 4.56.0
- Pytorch 2.8.0.dev20250319+cu128
- Datasets 4.0.0
- Tokenizers 0.22.0