textGAN (Adversarial Feature Matching for Text Generation)

Theano and tensorflow implementation for

• 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.

Run

• (Theano) Run:

THEANO_FLAGS=mode=FAST_RUN,device=gpu,floatX=float32 python textGAN.py

• (Tensorflow) Run:

python textGAN.py

Prerequisite

• Theano:

  • Theono >= 0.8
  • cPickle
  • cuda version 8.0
  • foxhound

• Tensorflow:

  • Tensorflow version == 1.2
  • cPickle

Data link

Three_corpus

Arxiv only

Pretrained parameter

http://people.duke.edu/~yz196/zip/param.zip

Evaluation

• (Theano only) python eval_kde.py

Citation

Please cite our paper if it helps with your research.

• Arxiv link: https://arxiv.org/abs/1706.03850

@inproceedings{zhang2017adversarial,
  title={Adversarial Feature Matching for Text Generation},
  author={Zhang, Yizhe and Gan, Zhe and Fan, Kai and Chen, Zhi and Henao, Ricardo and Shen, Dinghan and Carin, Lawrence},
  booktitle={ICML},
  year={2017}
}