/TextGAN

Project of course "Structured Data: Learning, Prediction Dependency, Testing" of Master Data Sciences (École Polytechnique).

Primary LanguagePython

TextGAN

This is a project of course "Structured Data: Learning, Prediction Dependency, Testing" of Master Data Sciences at École Polytechnique.

Prerequisite

PyTorch v1.0.1

Data Link

Three_corpus

Citation

This project is an implementation of the framework TextGAN proposed in the following papers:

  • Adversarial Feature Matching for Text Generation, Yizhe Zhang, Zhe Gan, Kai Fan, Zhi Chen, Ricardo Henao, Lawrence Carin. ICML, 2017.
  • Generating Text via Adversarial Training. Yizhe Zhang, Zhe Gan, Lawrence Carin.  Workshop on Adversarial Training, NIPS, 2016.

Structure

I implemented mainly the basic architecture of TextGAN. The pre-training and approximation techniques used by authors are not yet finished.

textGAN.py gives the architecture

train.py can be run to train a TextGAN model (TO FINISH)

utils.py gives some helper functions