Bert-NER
This model is a fine-tuned version of bert-base-uncased on the ner dataset. It achieves the following results on the evaluation set:
- Loss: 0.1205
- Precision: 0.9752
- Recall: 0.9924
- F1: 0.9837
- Accuracy: 0.9730
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: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.0825 | 1.0 | 501 | 0.1031 | 0.9600 | 0.9917 | 0.9756 | 0.9770 |
| 0.0337 | 2.0 | 1002 | 0.1491 | 0.9615 | 0.9942 | 0.9776 | 0.9648 |
| 0.0285 | 3.0 | 1503 | 0.1169 | 0.9754 | 0.9913 | 0.9833 | 0.9723 |
| 0.0249 | 4.0 | 2004 | 0.1054 | 0.9724 | 0.9921 | 0.9821 | 0.9783 |
| 0.0232 | 5.0 | 2505 | 0.1205 | 0.9752 | 0.9924 | 0.9837 | 0.9730 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1