RETURNN: The RWTH Extensible Training framework for Universal Recurrent Neural Networks

In this work we release our extensible and easily configurable neural network training software. It provides a rich set of functional layers with a particular focus on efficient training of recurrent neural network topologies on multiple GPUs. The source of the software package is public and freely...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2017-01
Main Authors: Doetsch, Patrick, Zeyer, Albert, Voigtlaender, Paul, Kulikov, Ilya, Schlüter, Ralf, Ney, Hermann
Format: Article
Language:English
Subjects:
Extensibility
Handwriting
Network topologies
Neural networks
Recurrent neural networks
Software
Systems analysis
Training
Two dimensional models
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this work we release our extensible and easily configurable neural network training software. It provides a rich set of functional layers with a particular focus on efficient training of recurrent neural network topologies on multiple GPUs. The source of the software package is public and freely available for academic research purposes and can be used as a framework or as a standalone tool which supports a flexible configuration. The software allows to train state-of-the-art deep bidirectional long short-term memory (LSTM) models on both one dimensional data like speech or two dimensional data like handwritten text and was used to develop successful submission systems in several evaluation campaigns.
ISSN:2331-8422