English

The Sockeye 2 Neural Machine Translation Toolkit at AMTA 2020

Computation and Language 2020-08-12 v1

Abstract

We present Sockeye 2, a modernized and streamlined version of the Sockeye neural machine translation (NMT) toolkit. New features include a simplified code base through the use of MXNet's Gluon API, a focus on state of the art model architectures, distributed mixed precision training, and efficient CPU decoding with 8-bit quantization. These improvements result in faster training and inference, higher automatic metric scores, and a shorter path from research to production.

Keywords

Cite

@article{arxiv.2008.04885,
  title  = {The Sockeye 2 Neural Machine Translation Toolkit at AMTA 2020},
  author = {Tobias Domhan and Michael Denkowski and David Vilar and Xing Niu and Felix Hieber and Kenneth Heafield},
  journal= {arXiv preprint arXiv:2008.04885},
  year   = {2020}
}