# Joey NMT: A Minimalist NMT Toolkit for Novices

## TL;DR

Joey NMT is a simple, beginner-friendly neural machine translation toolkit built on PyTorch, supporting key architectures and features, enabling novices to learn and adapt NMT models effectively while achieving competitive performance.

## Contribution

It introduces a minimalist NMT toolkit designed for novices, combining simplicity with essential features and competitive results, facilitating easier learning and customization.

## Key findings

- Novices perform nearly as well as experts after using Joey NMT tutorial.
- Joey NMT achieves performance comparable to complex toolkits on standard benchmarks.
- The toolkit is easy to learn and adapt for beginners with basic PyTorch and NMT knowledge.

## 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 .