NMT-Keras: a Very Flexible Toolkit with a Focus on Interactive NMT and Online Learning
Authors:
Preprint
Abstract
We present NMT-Keras, a flexible toolkit for training deep learning models, which puts a particular emphasis on the development of advanced applications of neural machine translation systems, such as interactive-predictive translation protocols and long-term adaptation of the translation system via continuous learning. NMT-Keras is based on an extended version of the popular Keras library, and it runs on Theano and Tensorflow. State-of-the-art neural machine translation models are deployed and used following the high-level framework provided by Keras. Given its high modularity and flexibility, it also has been extended to tackle different problems, such as image and video captioning, sentence classification and visual question answering.Keywords
All Related Versions
This publication has 18 references indexed in Scilit:
- Online learning for effort reduction in interactive neural machine translationComputer Speech & Language, 2019
- Egocentric video description based on temporally-linked sequencesJournal of Visual Communication and Image Representation, 2018
- Food Ingredients Recognition Through Multi-label Learning2017
- Interactive neural machine translationComputer Speech & Language, 2017
- VIBIKNet: Visual Bidirectional Kernelized Network for Visual Question AnsweringLecture Notes in Computer Science, 2017
- Simultaneous food localization and recognition2016
- Video Description Using Bidirectional Recurrent Neural NetworksLecture Notes in Computer Science, 2016
- Learning Deep Features for Discriminative Localization2016
- Going deeper with convolutions2015
- Bidirectional recurrent neural networksIEEE Transactions on Signal Processing, 1997