Joey NMT: A Minimalist NMT Toolkit for Novices

Julia Kreutzer, Jasmijn Bastings, Stefan Riezler
7/29/2019
cs.CLcs.LG

Official code: joeynmt/joeynmt (712 stars, Apache-2.0, last commit 2024-01-29).

Abstract

We present Joey NMT, a minimalist neural machine translation toolkit based on PyTorch that is specifically designed for novices. Joey NMT provides many popular NMT features in a small and simple code base, so that novices can easily and quickly learn to use it and adapt it to their needs. Despite its focus on simplicity, Joey NMT supports classic architectures (RNNs, transformers), fast beam search, weight tying, and more, and achieves performance comparable to more complex toolkits on standard benchmarks. We evaluate the accessibility of our toolkit in a user study where novices with general knowledge about Pytorch and NMT and experts work through a self-contained Joey NMT tutorial, showing that novices perform almost as well as experts in a subsequent code quiz. Joey NMT is available at https://github.com/joeynmt/joeynmt .

View on arXivView PDF

Code Implementations(1)

joeynmt/joeynmtOfficial100%
Apache-2.0

Minimalist NMT for educational purposes

712223PythonOct 15, 20182 years agoApache-2.0
educationjoey-nmtmachine-translationneural-machine-translationnmt+8 more

Similar Papers

OpenNMT: Open-Source Toolkit for Neural Machine Translation

88% match1 repo2.4k2017

OpenNMT: Neural Machine Translation Toolkit

88% match1 repo8052018

The eBible Corpus: Data and Model Benchmarks for Bible Translation for Low-Resource Languages

81% match1 repo62023

UM4: Unified Multilingual Multiple Teacher-Student Model for Zero-Resource Neural Machine Translation

81% match2022

SOTASTREAM: A Streaming Approach to Machine Translation Training

80% match1 repo222023

Joint Dropout: Improving Generalizability in Low-Resource Neural Machine Translation through Phrase Pair Variables

80% match1 repo2023

Cite this paper

@article{kreutzer2019joey,
  title  = {Joey NMT: A Minimalist NMT Toolkit for Novices},
  author = {Julia Kreutzer and Jasmijn Bastings and Stefan Riezler},
  year   = {2019},
  eprint = {1907.12484},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1907.12484}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1907.12484.svg)](https://codeofpaper.com/paper/1907.12484)

Discussion