Joey NMT: a minimalist NMT toolkit for novices

Abstract: JWe 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....

Full description

Saved in:
Bibliographic Details
Main Authors: Kreutzer, Julia (Author) , Bastings, Joost (Author) , Riezler, Stefan (Author)
Format: Chapter/Article Conference Paper
Language:English
Published: 2019
In: The 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing - proceedings of system demonstrations
Year: 2019, Pages: 109-114
Online Access: Get full text
Author Notes:Julia Kreutzer, Joost Bastings, Stefan Riezler
Description
Summary:Abstract: JWe 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 toolk its 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.
Physical Description:Online Resource
ISBN:9781950737925