multilingual-roberta-base-lumasaba-ner-v1
This model is a fine-tuned version of roberta-base on the Beijuka/Multilingual_PII_NER_dataset dataset. It achieves the following results on the evaluation set:
- Loss: 0.3357
- Precision: 0.9441
- Recall: 0.9357
- F1: 0.9399
- Accuracy: 0.9349
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 | 398 | 0.7383 | 0.7988 | 0.7533 | 0.7754 | 0.7568 |
| 1.1862 | 2.0 | 796 | 0.4723 | 0.8857 | 0.8457 | 0.8653 | 0.8432 |
| 0.4873 | 3.0 | 1194 | 0.4485 | 0.9198 | 0.8687 | 0.8935 | 0.8807 |
| 0.2817 | 4.0 | 1592 | 0.5033 | 0.8993 | 0.9187 | 0.9089 | 0.8989 |
| 0.2817 | 5.0 | 1990 | 0.3005 | 0.9416 | 0.9409 | 0.9413 | 0.9352 |
| 0.1806 | 6.0 | 2388 | 0.4968 | 0.9479 | 0.9097 | 0.9284 | 0.9220 |
| 0.1095 | 7.0 | 2786 | 0.5409 | 0.9118 | 0.9409 | 0.9261 | 0.9246 |
| 0.062 | 8.0 | 3184 | 0.5375 | 0.9282 | 0.9340 | 0.9311 | 0.9212 |
Framework versions
- Transformers 4.55.4
- Pytorch 2.8.0+cu126
- Datasets 4.0.0
- Tokenizers 0.21.4