Data and re-implemented scripts for 2017 AAAI paper, "Robsut Wrod Reocginiton via semi-Character Recurrent Neural Network"

Last updated: Feb 10th , 2017

---

This repository contains the re-implementation on the following paper:

@InProceedings{sakaguchi-duh-post-vandurme:2017:AAAI2017,
  author = {{Sakaguchi}, Keisuke and {Duh}, Kevin and {Post}, Matt and {Van Durme}, Ben},
  title = "{Robsut Wrod Reocginiton via semi-Character Recurrent Neural Network}",
  booktitle = {Proceedings of the Thirty-First {AAAI} Conference on Artificial Intelligence},
  month     = {February},
  year      = {2017},
  address   = {San Francisco, California},
  publisher = {{AAAI} Press},
  pages     = {x--x},
  url       = {https://}
}

Data

.
├── README.md  # this file
├── train.py # script for training
├── predict.py # predict correct word
├── models # store model files
└── binarize.py # utility function

Basic Usage

(training) THEANO_FLAGS=device=gpu1,floatX=float32 python train.py
(predicting) THEANO_FLAGS=device=gpu1,floatX=float32 python predict.py -m models/train_j-INT_n-JUMBLE_u-650_batch-20_ep-10_model.h5

Questions?

• Please e-mail to Keisuke Sakaguchi (keisuke[at]cs.jhu.edu).
• The original code base (written in Chainer) will also be released later on. (now refactoring for the latest version of Chainer)