========================================= README for Math-SparseMatrix version 0.01 ========================================= By Ted Pedersen, University of Minnesota, Duluth Mahesh Joshi University of Minnesota, Duluth 06/18/2006 ============ INTRODUCTION ============ Math::SparseMatrix is a CPAN module that implements operations on sparse matrices. This module uses Math::SparseVector as its building block. ============ INSTALLATION ============ DEPENDENCIES This module requires these other modules and libraries: Math::SparseVector To install this module type the following: perl Makefile.PL make make test make install The exact location of where Math::SparseMatrix will be installed depends on your system configuration. If you do not have authority to write into system directories, you can install Math::SparseMatrix in a local directory that you own and have permissions to read and write into as follows: perl Makefile.PL PREFIX=/YOUR/DIR make make test make install This will install the module into /YOUR/DIR/lib/perl5/site_perl/Math/SparseMatrix.pm If you install Math::SparseMatrix in a local directory, you will have to explicitly set your PERL5LIB environment variable to include: /YOUR/DIR/lib/perl5/site_perl if this directory was not already included. If you have any troubles during the installation, please contact the authors at tpederse@d.umn.edu or joshi031@d.umn.edu =============== GETTING STARTED =============== 1. To begin with, Math::SparseMatrix should be included in your Perl program as follows: # include this module for use in your program use Math::SparseMatrix; 2. To create an empty sparse matrix object with the required dimensions, use the following constructor: # create a new sparse matrix with 10 rows and 15 columns my $spmatrix = Math::SparseMatrix->new(10, 15); 3. To update the values in the sparse matrix, use the "set" function as follows: # set the value at row 5, column 3 to 10 $spmatrix->set(5, 3, 10); 4. To retrieve a stored value, use the "get" function as follows: # get the value at row 6, column 5 if present, or zero $val = $spmatrix->get(6, 5); 5. A sparse matrix can be written out to a file in the supported format (explained below) as follows: # write out the sparse matrix to the file "matrix.txt" $spmatrix->writeToFile("matrix.txt"); 6. A new sparse matrix object can be created from a file in the supported format as follows: # create a matrix object by reading the file "matrix.txt" my $spmatrix = Math::SparseMatrix->createFromFile("matrix.txt"); 7. A new sparse matrix that is the transpose of the matrix stored in the given input file can be created as follows: # create the transpose of the matrix stored in "matrix.txt" my $spmatrix = Math::SparseMatrix->createTransposeFromFile("matrix.txt"); 8. Finally, to generate the transpose of a matrix stored in a file, read the transpose as in #7 above and write out the read transpose to a new file as in #5 above. # create the transpose of the matrix stored in "matrix.txt" my $spmatrix = Math::SparseMatrix->createTransposeFromFile("matrix.txt"); # write out the created transpose to another file "transpose.txt" $spmatrix->writeToFile("transpose.txt"); SPARSE FILE FORMAT The sparse matrix file format that Math::SparseMatrix expects is described below with an example. The first line (or the header line) of the file should contain 3 number separated by a single space. The first number is the number of rows in the sparse matrix, the second number is the number of columns and the third number is the number of non-zero elements present in the stored matrix. Each subsequent line represents one row of the sparse matrix, therefore there should be as many number of lines after the header line as the number of rows mentioned in the header line. In every line representing a row, there should be as many pairs of numbers as the number of non-zero elements in that row. The first number in the pair represents the column number of the non-zero element (column numbers start with 1). The row number is implicitly provided by the line number in the file. The second number in the pair is the actual non-zero matrix element. Numbers in a pair and multiple pairs should all be separated by single spaces. If a row does not contain any non-zero element, then an empty line should be present in the file. NOTE: There should be no empty lines except those representing empty rows, neither should there be any comment lines. Commenting is not supported. Consider the sparse matrix of 5 rows and 4 columns below: 10 0 0 0 0 0 6 8 0 0 0 0 0 21 0 0 7 0 0 9 The sparse file representation for the same is: 5 4 6 1 10 3 6 4 8 2 21 1 7 4 9 Notice the empty line in between for the third row. ========= COPYRIGHT ========= Copyright (c) 2006, Ted Pedersen, University of Minnesota, Duluth. tpederse@d.umn.edu Mahesh Joshi, University of Minnesota, Duluth. joshi031@d.umn.edu This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with this program; if not, write to The Free Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. =============================================================================