/rnmt

A C++/CUDA toolkit for neural machine translation (RNN-Based NMT) across multiple GPUs

Primary LanguageC++

RNMT

A C++/CUDA toolkit for neural machine translation (RNN-Based NMT) across multiple GPUs

Requirements

  1. CUDA >= 7.5
  2. CuDNN >= 4
  3. Gcc >= 4.8
  4. Boost = 1.58

Usage

  1. Compiling, cd ./script and run ./compile_train.sh and ./compile_test.sh
  2. Training, cd ./work and run ./train_nmt.sh.
  3. Infenrence, cd ./work and run ./test_nmt.sh.

Development Team

Project leaders: Jiajun Zhang

Project members: Long Zhou, Yang Zhao, and Yining Wang

Contact

If you have questions, suggestions and bug reports, please email wszlong@gmail.com or long.zhou@nlpr.ia.ac.cn.