Counting 2,285 Big Data & Machine Learning Frameworks, Toolsets, and Examples...
Suggestion? Feedback? Tweet @stkim1

Author
Last Commit
Jan. 22, 2018
Created
Sep. 17, 2017

CNN-DCNN text autoencoder

Implementations of the models in the paper "Deconvolutional Paragraph Representation Learning" by Yizhe Zhang, Dinghan Shen, Guoyin Wang, Zhe Gan, Ricardo Henao and Lawrence Carin, NIPS 2017

Prerequisite:

  • Tensorflow (version >1.0)
  • CUDA, cudnn

Run

  • Run: python demo.py for reconstruction task
  • Run: python char_correction.py for character-level correction task
  • Options: options can be made by changing option class in the demo.py code.
  • opt.n_hidden: number of hidden units.
  • opt.layer: number of CNN/DCNN layer [2,3,4].
  • opt.lr: learning rate.
  • opt.batch_size: number of batchsize.
  • Training roughly takes 6-7 hours (around 10 epochs) (for recontruction task) to converge on a K80 GPU machine.
  • See output.txt for a sample of screen output.

Data:

For any question or suggestions, feel free to contact [email protected]

Citation

@inproceedings{zhang2017deconvolutional,
  title={Deconvolutional Paragraph Representation Learning},
  author={Zhang, Yizhe and Shen, Dinghan and Wang, Guoyin and Gan, Zhe and Henao, Ricardo and Carin, Lawrence},
  Booktitle={NIPS},
  year={2017}
}