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: 0.0823
- Precision: 0.9365
- Recall: 0.9525
- F1: 0.9444
- Accuracy: 0.9868
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 |
---|---|---|---|---|---|---|---|
0.0198 | 1.0 | 1756 | 0.0917 | 0.9145 | 0.9360 | 0.9251 | 0.9823 |
0.0111 | 2.0 | 3512 | 0.0783 | 0.9340 | 0.9507 | 0.9423 | 0.9866 |
0.007 | 3.0 | 5268 | 0.0823 | 0.9365 | 0.9525 | 0.9444 | 0.9868 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.14.5
- Tokenizers 0.15.0
- Downloads last month
- 12
Finetuned from
Dataset used to train scotssman/bert-finetuned-ner
Evaluation results
- Precision on conll2003validation set self-reported0.936
- Recall on conll2003validation set self-reported0.953
- F1 on conll2003validation set self-reported0.944
- Accuracy on conll2003validation set self-reported0.987