unpack: Representation of Packed and Unpacked (Dense) Matrices

In bedatadriven/renjin-matrix: Sparse and Dense Matrix Classes and Methods

Description

“Packed” matrix storage here applies to dense matrices (denseMatrix) only, and there is available only for symmetric (symmetricMatrix) or triangular (triangularMatrix) matrices, where only one triangle of the matrix needs to be stored.

unpack() unpacks “packed” matrices, where
pack() produces “packed” matrices.

Usage

pack(x, ...)
## S4 method for signature 'matrix'
pack(x, symmetric = NA, upperTri = NA, ...)

unpack(x, ...)

Arguments

x	for unpack(): a matrix stored in packed form, e.g., of class "d?pMatrix" where "?" is "t" for triangular or "s" for symmetric. for pack(): a (symmetric or triangular) matrix stored in full storage.
symmetric	logical (including NA) for optionally specifying if x is symmetric (or rather triangular).
upperTri	(for the triangular case only) logical (incl. NA) indicating if x is upper (or lower) triangular.
...	further arguments passed to or from other methods.

Details

These are generic functions with special methods for different types of packed (or non-packed) symmetric or triangular dense matrices. Use showMethods("unpack") to list the methods for unpack(), and similarly for pack().

Value

for unpack():

A Matrix object containing the full-storage representation of x.

for pack():

A packed Matrix (i.e. of class "..pMatrix") representation of x.

Examples

showMethods("unpack")
(cp4 <- chol(Hilbert(4))) # is triangular
tp4 <- as(cp4,"dtpMatrix")# [t]riangular [p]acked
str(tp4)
(unpack(tp4))
stopifnot(identical(tp4, pack(unpack(tp4))))

(s <- crossprod(matrix(sample(15), 5,3))) # traditional symmetric matrix
(sp <- pack(s))
mt <- as.matrix(tt <- tril(s))
(pt <- pack(mt))
stopifnot(identical(pt, pack(tt)),
	  dim(s ) == dim(sp), all(s  == sp),
	  dim(mt) == dim(pt), all(mt == pt), all(mt == tt))
showMethods("pack")

bedatadriven/renjin-matrix documentation built on May 12, 2019, 10:05 a.m.