Modelos de NER-conll2002
Collection
En ésta colección se encuentran los últimos modelos de NER usando las últimas técnicas • 6 items • Updated
NER-finetunining-Bert-Large-cased
This model is a fine-tuned version of google-bert/bert-large-cased on the conll2002 dataset. It achieves the following results on the evaluation set:
- Loss: 0.1418
- Precision: 0.7853
- Recall: 0.8235
- F1: 0.8039
- Accuracy: 0.9718
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: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.0896 | 1.0 | 1041 | 0.1297 | 0.7488 | 0.7829 | 0.7654 | 0.9667 |
| 0.0525 | 2.0 | 2082 | 0.1303 | 0.7548 | 0.8134 | 0.7830 | 0.9691 |
| 0.0275 | 3.0 | 3123 | 0.1418 | 0.7853 | 0.8235 | 0.8039 | 0.9718 |
Framework versions
- Transformers 4.51.3
- Pytorch 2.6.0+cu124
- Datasets 3.5.1
- Tokenizers 0.21.1
- Downloads last month
- 3
Model tree for raulgdp/NER-finetunining-Bert-Large-cased
Base model
google-bert/bert-large-casedDataset used to train raulgdp/NER-finetunining-Bert-Large-cased
Collection including raulgdp/NER-finetunining-Bert-Large-cased
Evaluation results
- Precision on conll2002validation set self-reported0.785
- Recall on conll2002validation set self-reported0.824
- F1 on conll2002validation set self-reported0.804
- Accuracy on conll2002validation set self-reported0.972