-
Notifications
You must be signed in to change notification settings - Fork 233
AmuNMT for Automatic Post Editing
Marcin Junczys-Dowmunt edited this page Aug 30, 2016
·
25 revisions
This page provides data and model files for our shared task winning APE system described in Log-linear Combinations of Monolingual and Bilingual Neural Machine Translation Models for Automatic Post-Editing. If you use any of the data or systems please cite:
@InProceedings{junczysdowmunt-grundkiewicz:2016:WMT,
author = {Junczys-Dowmunt, Marcin and Grundkiewicz, Roman},
title = {Log-linear Combinations of Monolingual and Bilingual Neural Machine Translation Models for Automatic Post-Editing},
booktitle = {Proceedings of the First Conference on Machine Translation},
month = {August},
year = {2016},
address = {Berlin, Germany},
publisher = {Association for Computational Linguistics},
pages = {751--758},
url = {http://www.aclweb.org/anthology/W16-2378}
}
http://odkrywka.wmi.amu.edu.pl/static/data/ape/data.tgz
data
├── 4M
│ ├── 4M.mt
│ ├── 4M.pe
│ └── 4M.src
└── 500K
├── 500K.mt
├── 500K.pe
└── 500K.src
http://odkrywka.wmi.amu.edu.pl/static/data/ape/system.tgz
system
├── data
│ ├── de.bpe
│ ├── en.bpe
│ ├── true.de
│ └── true.en
├── Makefile
├── models
│ ├── configs
│ │ ├── mt-pe.ensemble4.tuned.yml
│ │ ├── mtsrc-pe.ensemble.ape.tuned.yml
│ │ └── src-pe.ensemble4.yml
│ ├── mt-pe
│ │ ├── model.iter260000.npz
│ │ ├── model.iter270000.npz
│ │ ├── model.iter280000.npz
│ │ ├── model.iter290000.npz
│ │ ├── vocab.mt.json
│ │ └── vocab.pe.json
│ └── src-pe
│ ├── model.iter340000.npz
│ ├── model.iter350000.npz
│ ├── model.iter360000.npz
│ ├── model.iter370000.npz
│ ├── vocab.pe.json
│ └── vocab.src.json
├── scripts
│ ├── apply_bpe.py
│ ├── deescape-special-chars.perl
│ ├── detruecase.perl
│ ├── escape-special-chars.perl
│ ├── prepare_submission.py
│ ├── truecase.perl
│ └── unproc.sh
└── test
├── test.mt
└── test.src