https://github.com/kyzhouhzau/BERT-NER

Use Google's BERT for named entity recognition （CoNLL-2003 as the dataset）.
https://github.com/kyzhouhzau/BERT-NER

bert conll-2003 google-bert ner tensorflow

Last synced: about 1 month ago
JSON representation

Use Google's BERT for named entity recognition （CoNLL-2003 as the dataset）.

• Host: GitHub
• URL: https://github.com/kyzhouhzau/BERT-NER
• Owner: kyzhouhzau
• License: mit
• Created: 2018-11-07T06:55:44.000Z (over 5 years ago)
• Default Branch: master
• Last Pushed: 2022-05-19T05:06:32.000Z (about 2 years ago)
• Last Synced: 2024-02-23T00:32:12.205Z (5 months ago)
• Topics: bert, conll-2003, google-bert, ner, tensorflow
• Language: Python
• Homepage:
• Size: 2.18 MB
• Stars: 1,218
• Watchers: 37
• Forks: 337
• Open Issues: 78
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Lists

• awesome-bert - kyzhouhzau/BERT-NER - 2003 NER ! , (BERT NER task:)
• my-awesome-stars - kyzhouhzau/BERT-NER - Use Google's BERT for named entity recognition （CoNLL-2003 as the dataset）. (Python)
• awesome-transformer-nlp - kyzhouhzau/BERT-NER - Use google BERT to do CoNLL-2003 NER. (Tasks / Named-Entity Recognition (NER))

README

        
## For better performance, you can try NLPGNN, see [NLPGNN](https://github.com/kyzhouhzau/NLPGNN) for more details.

# BERT-NER Version 2

Use Google's BERT for named entity recognition （CoNLL-2003 as the dataset）. 

The original version （see old_version for more detail） contains some hard codes and lacks corresponding annotations,which is inconvenient to understand. So in this updated version,there are some new ideas and tricks （On data Preprocessing and layer design） that can help you quickly implement the fine-tuning model (you just need to try to modify crf_layer or softmax_layer).

### Folder Description:

```

BERT-NER

|____ bert                          # need git from [here](https://github.com/google-research/bert)

|____ cased_L-12_H-768_A-12	    # need download from [here](https://storage.googleapis.com/bert_models/2018_10_18/cased_L-12_H-768_A-12.zip)

|____ data		            # train data

|____ middle_data	            # middle data (label id map)

|____ output			    # output (final model, predict results)

|____ BERT_NER.py		    # mian code

|____ conlleval.pl		    # eval code

|____ run_ner.sh    		    # run model and eval result

```

### Usage:

```

bash run_ner.sh

```

### What's in run_ner.sh:

```

python BERT_NER.py\

    --task_name="NER"  \

    --do_lower_case=False \

    --crf=False \

    --do_train=True   \

    --do_eval=True   \

    --do_predict=True \

    --data_dir=data   \

    --vocab_file=cased_L-12_H-768_A-12/vocab.txt  \

    --bert_config_file=cased_L-12_H-768_A-12/bert_config.json \

    --init_checkpoint=cased_L-12_H-768_A-12/bert_model.ckpt   \

    --max_seq_length=128   \

    --train_batch_size=32   \

    --learning_rate=2e-5   \

    --num_train_epochs=3.0   \

    --output_dir=./output/result_dir

perl conlleval.pl -d '\t' < ./output/result_dir/label_test.txt

```

**Notice:** cased model was recommened, according to [this](https://arxiv.org/abs/1810.04805) paper. CoNLL-2003 dataset and perl Script comes from [here](https://www.clips.uantwerpen.be/conll2003/ner/)

### RESULTS:(On test set)

#### Parameter setting:

* do_lower_case=False 

* num_train_epochs=4.0

* crf=False

  

```

accuracy:  98.15%; precision:  90.61%; recall:  88.85%; FB1:  89.72

              LOC: precision:  91.93%; recall:  91.79%; FB1:  91.86  1387

             MISC: precision:  83.83%; recall:  78.43%; FB1:  81.04  668

              ORG: precision:  87.83%; recall:  85.18%; FB1:  86.48  1191

              PER: precision:  95.19%; recall:  94.83%; FB1:  95.01  1311

```

### Result description:

Here i just use the default paramaters, but as Google's paper says a 0.2% error is reasonable(reported 92.4%).

Maybe some tricks need to be added to the above model. 

### reference:

[1] https://arxiv.org/abs/1810.04805

[2] https://github.com/google-research/bert