bert-finetuned-ner
This model is a fine-tuned version of bert-base-cased on the conll2003 dataset. It achieves the following results on the evaluation set:
- Loss: 0.0646
- Precision: 0.9386
- Recall: 0.9542
- F1: 0.9463
- Accuracy: 0.9870
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: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.0344 | 1.0 | 1756 | 0.0655 | 0.9264 | 0.9463 | 0.9362 | 0.9851 |
| 0.0189 | 2.0 | 3512 | 0.0713 | 0.9345 | 0.9507 | 0.9425 | 0.9863 |
| 0.0094 | 3.0 | 5268 | 0.0646 | 0.9386 | 0.9542 | 0.9463 | 0.9870 |
Framework versions
- Transformers 4.41.1
- Pytorch 2.3.0
- Datasets 2.19.1
- Tokenizers 0.19.1