nmfgpu4R: GPU-Accelerated Computation of the Non-Negative Matrix Factorization (NMF) Using CUDA Capable Hardware

Authors

• Sven Koitka, University of Applied Sciences and Arts Dortmund
• Christoph M. Friedrich, University of Applied Sciences and Arts Dortmund

Date of this Version

12-2016

Document Type

Article

Citation

The R Journal (December 2016) 8(2); Editor: Michael Lawrence

Comments

Copyright 2016, The R Foundation. Open access material. License: CC BY 3.0 Unported

Abstract

In this work, a novel package called nmfgpu4R is presented, which offers the computation of Non-negative Matrix Factorization (NMF) on Compute Unified Device Architecture (CUDA) platforms within the R environment. Benchmarks show a remarkable speed-up in terms of time per iteration by utilizing the parallelization capabilities of modern graphics cards. Therefore the application of NMFgets more attractive for real-world sized problems because the time to compute a factorization is reduced by an order of magnitude.