Contextual Encoding for Translation Quality Estimation

Implemented by Junjie Hu

Contact: [email protected]

If you use the codes in this repo, please cite our WMT18 paper.

@InProceedings{hu-EtAl:2018:WMT,
  author    = {Hu, Junjie  and  Chang, Wei-Cheng  and  Wu, Yuexin  and  Neubig, Graham},
  title     = {Contextual Encoding for Translation Quality Estimation},
  booktitle = {Proceedings of the Third Conference on Machine Translation: Shared Task Papers},
  month     = {October},
  year      = {2018},
  address   = {Belgium, Brussels},
  publisher = {Association for Computational Linguistics},
  pages     = {788--793},
  url       = {http://www.aclweb.org/anthology/W18-6462}
  }

Install

sklearn 
pytorch >= 0.4.0
scipy
scipy.stats
conda env create --file wmt-qe-py3.5-pt4.0.txt

Download Data

cd data/WMT2018QE/
bash download_data.sh

Train Model

bash scripts/WMT2018QE/de-en/train-word-ClBasic-final.sh [GPU ID]