bert-finetuned-ner-conll
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: 0.0593
- Precision: 0.9152
- Recall: 0.9403
- F1: 0.9275
- Accuracy: 0.9845
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: 48
- eval_batch_size: 48
- 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 |
---|---|---|---|---|---|---|---|
No log | 1.0 | 293 | 0.0773 | 0.8819 | 0.9138 | 0.8976 | 0.9775 |
0.1657 | 2.0 | 586 | 0.0598 | 0.9101 | 0.9374 | 0.9236 | 0.9835 |
0.1657 | 3.0 | 879 | 0.0593 | 0.9152 | 0.9403 | 0.9275 | 0.9845 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.13.1
- Datasets 2.6.1
- Tokenizers 0.11.0
- Downloads last month
- 7
Dataset used to train theArif/bert-finetuned-ner-conll
Evaluation results
- Precision on conll2003self-reported0.915
- Recall on conll2003self-reported0.940
- F1 on conll2003self-reported0.928
- Accuracy on conll2003self-reported0.985