DeepNER

An Easy-to-use, Modular and Prolongable package of deep-learning based Named Entity Recognition Models.

This repository contains complex Deep Learning models for named entity recognition.

Requirements

• Tensorflow-gpu=1.14.0
• jieba=0.37

Train Models

• Train Transformer with CRF

python test_Transformer_CRF.py --num_blocks 2

• Train Bilstm with CRF

python test_BiLSTM_CRF.py

• fintuning bert with freezing bert Variables

python test_BiLSTM_CRF.py --freeze_bert True

• fintuning bert Variables simultaneously

python test_BiLSTM_CRF.py --freeze_bert False

Performances Comparison

TODO

• Lexicon enhance
• Label Attention Network for fine-gained NER
• Nested NER

References

• Bert
• Chinese_NER
• Transformer