Marian: Fast Neural Machine Translation in C++
Marian: Fast Neural Machine Translation in C++
Marcin Junczys-Dowmunt,Roman Grundkiewicz,9 Authors,Alexandra Birch
2018 · DOI: 10.18653/v1/P18-4020
Annual Meeting of the Association for Computational Linguistics · 811 Citations
TLDR
Marian is an efficient and self-contained Neural Machine Translation framework with an integrated automatic differentiation engine based on dynamic computation graphs that can achieve high training and translation speed.
Abstract
We present Marian, an efficient and self-contained Neural Machine Translation framework with an integrated automatic differentiation engine based on dynamic computation graphs. Marian is written entirely in C++. We describe the design of the encoder-decoder framework and demonstrate that a research-friendly toolkit can achieve high training and translation speed.