Skip to content

frankxu2004/feedforward-RE

 
 

Repository files navigation

feedforward-RE

Feedforward NeuralNet for Relation Extraction.

Data Preparation

  • Place data in CoType format obtained in data/source/$dataname.
  • Using KBP as the exmaple dataset, run ./brown_clustering.sh KBP to generate browncluster file (may take a while, don't repeat if nothing in data changed)

Feature Extraction

First, run the server using all jars in the CoreNLP home directory (in another terminal): java -mx4g -cp "*" edu.stanford.nlp.pipeline.StanfordCoreNLPServer -port 9000 -timeout 15000 Then: ./feature_extraction.sh KBP

Train FF-NN model

python3 FFNN/run.py KBP

About

Feedforward NeuralNet for Relation Extraction

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 75.3%
  • C++ 22.4%
  • C 1.5%
  • Shell 0.4%
  • HTML 0.2%
  • Makefile 0.1%
  • CSS 0.1%