nmfgpu4R: GPU-Accelerated Computation of the Non-Negative Matrix Factorization (NMF) Using CUDA Capable Hardware
Department of Computer Science, University of Applied Sciences and Arts Dortmund (FHDO), Emil-Figge-Strasse 42, 44227 Dortmund, Germany
The R Journal, 2016
@article{koitka2016nmfgpu4r,
title={nmfgpu4R: GPU-Accelerated Computation of the Non-Negative Matrix Factorization (NMF) Using CUDA Capable Hardware},
author={Koitka, Sven and Friedrich, Christoph M},
year={2016}
}
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 NMF gets more attractive for real-world sized problems because the time to compute a factorization is reduced by an order of magnitude.
December 14, 2016 by hgpu
