Design of an Open-Source Architecture for Neural Machine Translation

arXiv:2403.03582v1 Announce Type: new
Abstract: adaptNMT is an open-source application that offers a streamlined approach to the development and deployment of Recurrent Neural Networks and Transformer models. This application is built upon the widely-adopted OpenNMT ecosystem, and is particularly useful for new entrants to the field, as it simplifies the setup of the development environment and creation of train, validation, and test splits. The application offers a graphing feature that illustrates the progress of model training, and employs SentencePiece for …

abstract application architecture arxiv cs.ai cs.cl deployment design development ecosystem machine machine translation networks neural machine translation neural networks recurrent neural networks setup transformer transformer models translation type