ner-from-bert
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: 0.0615
- Precision: 0.9351
- Recall: 0.9504
- F1: 0.9427
- Accuracy: 0.9859
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 |
---|---|---|---|---|---|---|---|
0.0879 | 1.0 | 1756 | 0.0685 | 0.9170 | 0.9320 | 0.9245 | 0.9815 |
0.0328 | 2.0 | 3512 | 0.0625 | 0.9267 | 0.9495 | 0.9380 | 0.9853 |
0.0189 | 3.0 | 5268 | 0.0615 | 0.9351 | 0.9504 | 0.9427 | 0.9859 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.0+cu116
- Datasets 2.8.0
- Tokenizers 0.13.2
- Downloads last month
- 9
Dataset used to train dsoum/ner-from-bert
Evaluation results
- Precision on conll2003self-reported0.935
- Recall on conll2003self-reported0.950
- F1 on conll2003self-reported0.943
- Accuracy on conll2003self-reported0.986