Gmm++ User Documentation

• Introduction
• How to install and use GMM++
• Matrix and Vector type provided by GMM++
• • dense vectors
  • sparse vectors
  • skyline vectors
  • generic row and column matrices
  • dense matrices
  • sparse matrices
• Input and output with Harwell-Boeing and Matrix Market formats
• sub-vectors and sub-matrices
• • row and column of a matrix
• miscellaneous methods
• Basic linear algebra operations
• • scale and scaled
  • transposition
  • imaginary and real part
  • conjugate
  • add
  • mult
  • norms
  • trace
  • scalar product
• Solving triangular systems
• Dense LU decomposition
• Dense QR factorisation, eigenvalues and eigenvectors
• Condition number estimation
• Iterative solvers
• • iterations
  • various solvers
  • various preconditioners
  • Additive Schwarz method
• Catch errors
• Interface with BLAS, LAPACK or ATLAS
• Interface with SuperLU
• How to use GMM++ with QD type (double-double and quad-double)
• First steps with GMM++
• • How can I invert a matrix ?
  • How can I solve a linear system ?
  • How can I transform a vector into a matrix and reshape it ?
  • What is the better way to resize a matrix ?
• Deeper inside GMM++
• • The linalg_traits structure
  • How to iterate on the components of a vector
  • How to iterate on a matrix
  • How to make your algorithm working on all type of matrices
• How to disable verifications
• References
• Index

Introduction

GMM++ provides some basic types of sparse and dense matrices and vectors. It provides some generic operations on them (copy, addition, multiplication, sub-vector and sub-matrices, solvers ... ). The syntax of GMM++ is very close to MTL and ITL (see http://www.osl.iu.edu/research/mtl/). Especially, the code for most of the iterative solvers has been imported from ITL. The performance of GMM++ is also close to the one of MTL, sometimes better. The difference is that basically GMM++ has been written to be able to interface other libraries and gives an access to sub matrices and sub vectors in all cases. Also some optimizations has been made for matrix-matrix multiplication, usage of reference has been somewhat cleared, const qualifier usage is clarified, and we hope it is somewhat easier to use.




Copyright (C) 2000-2008 Yves Renard, Julien Pommier.
The program GETFEM++ is free software; you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation; either version 2.1 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 Lesser General Public License for more details. You should have received a copy of the GNU Lesser General Public License along with this program; if not, write to the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA

---