DeconvDec

Code for the model proposed in our paper Deconvolution-Based Global Decoding for Neural Machine Translation, http://aclweb.org/anthology/C18-1276.

Requirements

• Ubuntu 16.0.4
• Python 3.5
• Pytorch 0.4.1

Preprocessing

python3 preprocess.py -load_data path_to_data -save_data path_to_store_data 

Remember to put the data into a folder and name them train.src, train.tgt, valid.src, valid.tgt, test.src and test.tgt, and make a new folder inside called data. For more detailed setting, check the options in the file.

Training

python3 train.py -log log_name -config config_yaml -gpus id

Create your own yaml file for hyperparameter setting.

Evaluation

python3 train.py -log log_name -config config_yaml -gpus id -restore checkpoint -mode eval

Citation

If you use this code for your research, please kindly cite our paper:.

@inproceedings{DeconvDec,
  author    = {Junyang Lin and
               Xu Sun and
               Xuancheng Ren and
               Shuming Ma and
               Jinsong Su and
               Qi Su},
  title     = {Deconvolution-Based Global Decoding for Neural Machine Translation},
  booktitle = {Proceedings of the 27th International Conference on Computational
               Linguistics, {COLING} 2018, Santa Fe, New Mexico, USA, August 20-26,
               2018},
  pages     = {3260--3271},
  year      = {2018}
}