opnmfRcpp: Orthogonal Projective Non-Negative Factorization utilising R...

In kaurao/opnmfR: Orthogonal Projective Non-negative Matrix Factorization

Description

Orthogonal Projective Non-Negative Factorization utilising R C++ integration for improved computational speed. Implementation is partly based on brainparts: https://github.com/asotiras/brainparts

Usage

opnmfRcpp(
  X,
  r,
  W0 = NULL,
  max.iter = 50000,
  tol = 1e-05,
  memsave = TRUE,
  eps = 1e-16,
  ...
)

Arguments

X	A matrix, rows are the number of features and columns are number of samples
r	A number, rank used for factorization.
W0	A string or matrix for initialization (default NULL)
max.iter	A number, number of iterations before stopping (default 50000)
tol	A number, convergence tolerance (default 1e-5)
memsave	A logical, if TRUE update rule is modified to better deal with high dimensional data (default TRUE)
eps	A number, (default 1e-16)

Value

A list containing the approximation matrices W and H, of input matrix X; the interation number at which a solution was found; the reconstruction error

Examples

 result <- opnmfRcpp()

kaurao/opnmfR documentation built on March 12, 2021, 4:15 a.m.