multilingual-xlm-roberta-base-kanuri-ner-v1
This model is a fine-tuned version of xlm-roberta-base on the Beijuka/Multilingual_PII_NER_dataset dataset. It achieves the following results on the evaluation set:
- Loss: 0.0719
- Precision: 0.9064
- Recall: 0.9416
- F1: 0.9237
- Accuracy: 0.9822
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 | 0.1158 | 0.9102 | 0.8262 | 0.8662 | 0.9666 |
| 0.229 | 2.0 | 602 | 0.0918 | 0.8883 | 0.9287 | 0.9080 | 0.9736 |
| 0.229 | 3.0 | 903 | 0.0924 | 0.8654 | 0.9401 | 0.9012 | 0.9751 |
| 0.0702 | 4.0 | 1204 | 0.1025 | 0.8772 | 0.9461 | 0.9103 | 0.9750 |
| 0.0514 | 5.0 | 1505 | 0.1446 | 0.8542 | 0.8670 | 0.8605 | 0.9671 |
| 0.0514 | 6.0 | 1806 | 0.1227 | 0.8946 | 0.9203 | 0.9073 | 0.9732 |
| 0.034 | 7.0 | 2107 | 0.1240 | 0.8949 | 0.9233 | 0.9089 | 0.9747 |
Framework versions
- Transformers 4.55.4
- Pytorch 2.8.0+cu126
- Datasets 4.0.0
- Tokenizers 0.21.4