Skip to content

Latest commit

 

History

History
31 lines (20 loc) · 984 Bytes

README.md

File metadata and controls

31 lines (20 loc) · 984 Bytes

RNN-Stega

This code belongs to RNN-Stega: Linguistic Steganography Based on Recurrent Neural Networks.

Requirements

  • python 2.7
  • tensorflow = 1.0.0

Prepare for Generating

  • Download our corpus file movie or tweet and put it in ./data/
  • Download pretrained models movie model or tweet model and put it in ./models/movie/ or ./models/tweet/
  • Modify ./Config_movie.py or ./Config_tweet.py to adjust the hyperparameters

Generate

  • movie dataset
python huffman_movie_v2.py 1 1
  • tweet dataset
python huffman_tweet_v2.py 1 1

The args stand for bit and index.