ner-jopara-mmaguero-multilingual-bert-gn
This model is a fine-tuned version of mmaguero/multilingual-bert-gn-base-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4157
- Precision: 0.8498
- Recall: 0.8714
- F1: 0.8604
- Accuracy: 0.9350
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: 16
- eval_batch_size: 16
- 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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.3407 | 1.0 | 1764 | 0.2853 | 0.7565 | 0.7949 | 0.7752 | 0.9040 |
| 0.2297 | 2.0 | 3528 | 0.2458 | 0.8116 | 0.8390 | 0.8251 | 0.9236 |
| 0.1595 | 3.0 | 5292 | 0.2321 | 0.8263 | 0.8554 | 0.8406 | 0.9296 |
| 0.111 | 4.0 | 7056 | 0.2705 | 0.8196 | 0.8638 | 0.8411 | 0.9287 |
| 0.0808 | 5.0 | 8820 | 0.3160 | 0.8376 | 0.8671 | 0.8521 | 0.9315 |
| 0.0667 | 6.0 | 10584 | 0.3319 | 0.8526 | 0.8728 | 0.8626 | 0.9340 |
| 0.038 | 7.0 | 12348 | 0.3622 | 0.8467 | 0.8695 | 0.8579 | 0.9344 |
| 0.0332 | 8.0 | 14112 | 0.3844 | 0.8496 | 0.8712 | 0.8603 | 0.9352 |
| 0.0213 | 9.0 | 15876 | 0.4027 | 0.8522 | 0.8717 | 0.8618 | 0.9343 |
| 0.0183 | 10.0 | 17640 | 0.4108 | 0.8529 | 0.8752 | 0.8639 | 0.9344 |
Framework versions
- Transformers 4.57.1
- Pytorch 2.8.0+cu126
- Datasets 4.0.0
- Tokenizers 0.22.1