Skip to content
NNHetSeq is a neural network based package for heterogeneous sequence labling tasks. The system can be used for POS tagging, Named Entity Recognition, and other sequence labeling tasks.
C++ Other
  1. C++ 99.4%
  2. Other 0.6%
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
basic
example
CMakeLists.txt
COUPLESTACK_LSTMCRFMMLabeler.cpp
COUPLESTACK_LSTMCRFMMLabeler.h
COUPLE_LSTMCRFMMLabeler.cpp
COUPLE_LSTMCRFMMLabeler.h
LSTMCRFMMLabeler.cpp
LSTMCRFMMLabeler.h
Options.h
README.md
STACK_LSTMCRFMMLabeler.cpp
STACK_LSTMCRFMMLabeler.h

README.md

NNHetSeq

NNHetSeq is a Bi-LSTM CRF based package for heterogeneous sequence labelling tasks. The system can be used for POS tagging, Named Entity Recognition, and other sequence labelling tasks.

Performance

POS-tagging accuracies of CTB5 boosts from 94.24 to 95.53 aided by PD corpus.

Prerequisition

LibN3L

Compile

cmake .
make

Usage

cd example/

  1. run stacking model a. run base model and preserve the parameters using corpus A as command in run_base.sh. b. then, run stacking model as command in run_stack.sh using corpus B, and initialize the stacking model with the base model parameters.
  2. run multi-view coupling model as command in run_couple.sh directly set two resources of corpora (corpus A and corpus B) and run it
  3. run integrated stacking and multi-view coupling model a. run base model and preserve the parameters using corpus A as command in run_base.sh.. b. run integrated model using corpus A and corpus B as command in run_couplestack.sh, and initialize the integrated model with the base model parameters.

Notes:

  1. NN stacking model (and the integrated model) convergent faster than multi-view coupling model based on our observations.
You can’t perform that action at this time.