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: 0.0604
- Precision: 0.9277
- Recall: 0.9483
- F1: 0.9379
- Accuracy: 0.9865
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 |
---|---|---|---|---|---|---|---|
0.0885 | 1.0 | 1756 | 0.0702 | 0.9124 | 0.9290 | 0.9206 | 0.9810 |
0.036 | 2.0 | 3512 | 0.0600 | 0.9303 | 0.9502 | 0.9401 | 0.9865 |
0.0192 | 3.0 | 5268 | 0.0604 | 0.9277 | 0.9483 | 0.9379 | 0.9865 |
Framework versions
- Transformers 4.21.1
- Pytorch 1.12.1+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1
- Downloads last month
- 1
Dataset used to train zeptrus/bert-finetuned-ner
Evaluation results
- Precision on conll2003self-reported0.928
- Recall on conll2003self-reported0.948
- F1 on conll2003self-reported0.938
- Accuracy on conll2003self-reported0.986