crs2mat: Convert a CRS-formatted matrix to standard square format

In kexpmv: Matrix Exponential using Krylov Subspace Routines

Description

The EXPOKIT's DMEPXV and DGEXPV functions both deal with sparse matrices. These matrices have a lot of zeros, and can therefore be stored more efficiently by converting the matrix into CRS (Compressed Row Storage) format.

Usage

crs2mat(mat, n)

Arguments

mat	a 3 column list including the vectors; ia, ja and a.
n	the dimension of the square (nxn) matrix.

Details

In EXPOKIT and its wrapper functions, a CRS-formatted matrix is input as 3 vectors:

ia = row pointer. This vector stores the location in the ‘a’ vector that is the first non-zero element in a row.
ja = column indices for non-zero elements.
a = non-zero elements of the matrix.

This function takes a 3-column list CRS matrix and converts it back to standard square format.

Author(s)

Meabh G. McCurdy [e-mail]

Examples

# Create a CRS format matrix
ia = c(1, 2, 4, 6)
ja = c(1, 1, 2, 1, 2)
a  = c(-0.071207,  0.065573, -0.041206,  0.005634,  0.041206)
crsmat=list(ia, ja, a)

# Convert CRS format matrix to square format
n = 3
mat = crs2mat(crsmat, n)
print(mat)

kexpmv documentation built on May 1, 2019, 7:56 p.m.