NMTPY: A Flexible Toolkit for Advanced Neural Machine Translation Systems

In this paper, we present nmtpy, a flexible Python toolkit based on Theano for training Neu-ral Machine Translation and other neural sequence-to-sequence architectures. nmtpy decouples the specification of a network from the training and inference utilities to simplify the addition of a new architecture and reduce the amount of boilerplate code to be written. nmtpy has been used for LIUM's top-ranked submissions to WMT Multimodal Machine Translation and News Translation tasks in 2016 and 2017.

Domaines

Traitement du texte et du document

Fichier principal

[The Prague Bulletin of Mathematical Linguistics] NMTPY A Flexible Toolkit for Advanced Neural Machine Translation Systems.pdf (270.45 Ko)

Origine	Fichiers éditeurs autorisés sur une archive ouverte

Ozan Caglayan : Connectez-vous pour contacter le contributeur

https://univ-lemans.hal.science/hal-01647873

Soumis le : jeudi 11 janvier 2018-14:29:46

Dernière modification le : mercredi 30 octobre 2024-21:45:40

Archivage à long terme le : samedi 5 mai 2018-09:36:34

Dates et versions

hal-01647873 , version 1 (11-01-2018)

Identifiants

HAL Id : hal-01647873 , version 1
DOI : 10.1515/pralin-2017-0035

Citer

Ozan Caglayan, Mercedes Garcia-Martinez, Adrien Bardet, Walid Aransa, Fethi Bougares, et al.. NMTPY: A Flexible Toolkit for Advanced Neural Machine Translation Systems. The Prague Bulletin of Mathematical Linguistics, 2017, 109 (1), ⟨10.1515/pralin-2017-0035⟩. ⟨hal-01647873⟩

Exporter

BibTeX XML-TEI Dublin Core DC Terms EndNote DataCite

Collections

UNIV-AVIGNON UNIV-LEMANS LIUM LIUM-LST LIA ANR

271 Consultations

484 Téléchargements