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: 0.0284
- Precision: 0.9316
- Recall: 0.9505
- F1: 0.9409
- 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 |
---|---|---|---|---|---|---|---|
0.0349 | 1.0 | 1756 | 0.0347 | 0.9085 | 0.9362 | 0.9222 | 0.9810 |
0.0184 | 2.0 | 3512 | 0.0269 | 0.9281 | 0.9495 | 0.9387 | 0.9869 |
0.009 | 3.0 | 5268 | 0.0284 | 0.9316 | 0.9505 | 0.9409 | 0.9870 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1
- Datasets 2.14.5
- Tokenizers 0.13.3
- Downloads last month
- 8
Finetuned from
Dataset used to train michaelsungboklee/bert-finetuned-ner
Evaluation results
- Precision on conll2003validation set self-reported0.932
- Recall on conll2003validation set self-reported0.951
- F1 on conll2003validation set self-reported0.941
- Accuracy on conll2003validation set self-reported0.987