The R Journal: article published in 2016, volume 8:2

nmfgpu4R: GPU-Accelerated Computation of the Non-Negative Matrix Factorization (NMF) Using CUDA Capable Hardware
Sven Koitka and Christoph M. Friedrich , The R Journal (2016) 8:2, pages 382-392.

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 NMF gets more attractive for real-world sized problems because the time to compute a factorization is reduced by an order of magnitude.

Received: 2016-04-30; online 2016-11-21
CRAN packages: NMF, NMFN, nmfgpu4R, Matrix, SparseM
CRAN Task Views implied by cited CRAN packages: Econometrics, Multivariate, NumericalMathematics

This article is licensed under a Creative Commons Attribution 3.0 Unported license .

@article{RJ-2016-053,
  author = {Sven Koitka and Christoph M. Friedrich},
  title = {{nmfgpu4R: GPU-Accelerated Computation of the Non-Negative
          Matrix Factorization (NMF) Using CUDA Capable Hardware}},
  year = {2016},
  journal = {{The R Journal}},
  doi = {10.32614/RJ-2016-053},
  url = {https://doi.org/10.32614/RJ-2016-053},
  pages = {382--392},
  volume = {8},
  number = {2}
}