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: 0.0641
- Precision: 0.9290
- Recall: 0.9475
- F1: 0.9382
- Accuracy: 0.9858
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 |
|---|---|---|---|---|---|---|---|
| 0.0867 | 1.0 | 1756 | 0.0716 | 0.9102 | 0.9297 | 0.9198 | 0.9820 |
| 0.0345 | 2.0 | 3512 | 0.0680 | 0.9290 | 0.9465 | 0.9376 | 0.9854 |
| 0.0191 | 3.0 | 5268 | 0.0641 | 0.9290 | 0.9475 | 0.9382 | 0.9858 |
Framework versions
- Transformers 4.21.1
- Pytorch 1.12.1+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1