Skip to content
/ DA-CRF Public
forked from DCdream/DA-CRF

This is a implementation of "动态注意力机制:一种面向属性抽取的CRF优化器分类方法"

Notifications You must be signed in to change notification settings

wq1315/DA-CRF

 
 

Repository files navigation

DA-CRF

  • Framework: Bi-lstm-Attention-CRF
    Gated dynamic Attention

installation

  • python >= 3.5.2
    tensorflow >= 1.2.0
    numpy >=1.11.1

Training, developmenting and testing simultaneously

  • python main.py

Other

  • If you want some other parameter settings, you can change them on main.py
    This project also implements a static attention, see model.py, if you wanna use it, please manually modify on model.py
    You can change the implemented CoupledInputForgetGateLSTMCell to tensorflow's LSTM_Cell manually in model.py
    This implementation also suports Chinese, you just feed data like the data on directery "data1"

About

This is a implementation of "动态注意力机制:一种面向属性抽取的CRF优化器分类方法"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 87.9%
  • Perl 12.1%