bert-finetuned-ner
This model is a fine-tuned version of thanapolsjr/bert-finetuned-ner-accelerate on the conll2003 dataset. It achieves the following results on the evaluation set:
- Loss: 0.0730
- Precision: 0.9350
- Recall: 0.9519
- F1: 0.9434
- Accuracy: 0.9869
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: 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: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.0230 | 1.0 | 1756 | 0.0606 | 0.9317 | 0.9465 | 0.9391 | 0.9859 |
| 0.0130 | 2.0 | 3512 | 0.0692 | 0.9384 | 0.9490 | 0.9437 | 0.9867 |
| 0.0056 | 3.0 | 5268 | 0.0730 | 0.9350 | 0.9519 | 0.9434 | 0.9869 |
Framework versions
- Transformers 5.16.1
- Pytorch 2.11.0+cu128
- Datasets 3.6.0
- Tokenizers 0.23.1