randSVD: Singular value decomposition in sparse triangular matrix

View source: R/rppca.R

randSVDR Documentation

Singular value decomposition in sparse triangular matrix

Description

Uses randomised linear algebra, see Halko et al. (2010). Singular value decomposition (SVD) decomposes a matrix X=U\Sigma W^T

Usage

randSVD(L, rank, depth, numVectors, cent = FALSE)

Arguments

L

a pedigree's L inverse matrix in sparse 'spam' format

rank

An integer, how many principal components to return

depth

integer, the number of iterations for generating the range matrix

numVectors

An integer > rank to specify the oversampling for the

cent

logical, whether or not to (implicitly) centre the additive relationship matrix

Value

A list of three: u (=U), d (=Sigma), and v (=W^T)


randPedPCA documentation built on Aug. 8, 2025, 6:37 p.m.