文件列表:

incomplete_Cho_solution\blas_headers.h (2906, 2007-07-11)
incomplete_Cho_solution\chol_incomplete.c (10423, 2008-04-29)
incomplete_Cho_solution\chol_incomplete.m (2005, 2007-07-11)
incomplete_Cho_solution\Makefile (31, 2007-07-11)
incomplete_Cho_solution\mex_helper.h (6685, 2007-07-11)
incomplete_Cho_solution\testprog0.m (1162, 2007-07-11)
incomplete_Cho_solution\testprog1.m (1207, 2007-07-11)
incomplete_Cho_solution (0, 2014-06-27)

INCOMPLETE CHOLESKY DECOMPOSITION FOR DENSE MATRICES ---------------------------------------------------- Copyright (C) 2005 Matthias Seeger 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., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. Matthias Seeger Max Planck Institute for Biological Cybernetics P.O. Box: 21 69 72012 Tuebingen, Germany seeger@tuebingen.mpg.de www.kyb.tuebingen.mpg.de/bs/people/seeger/ 1) Installation - Untar chol_incomplete.tar.gz: tar xzf chol_incomplete.tar.gz Go to subdirectory chol_incomplete. - Run: make This should produce a file chol_incomplete., where is "mexglx" in Linux, but maybe something else in other systems. If this does not work, check your Matlab docs for how to compile MEX files and edit the Makefile. - Copy the files chol_incomplete. chol_incomplete.m somewhere into your Matlab path 2) How to use it The Matlab help for CHOL_INCOMPLETE should be clear enough, and the code is well-documented. The incomplete Cholesky decomposition works like a normal CD, but you stop after less than n iterations. Furthermore, you reorder the rows/columns as you go, s.t. you always choose the next column to have the largest possible pivot element (diagonal of L). If you want to know more, you might want to check @article{Fine:01, author = {Fine, S. and Scheinberg, K.}, title = {Efficient {SVM} Training using Low-Rank Kernel Representations}, journal = JMLR, volume = {2}, pages = {243--2***}, year = {2001} } Once L is computed, you can use L*L' as approximation to P'*A*P, where A is the matrix to be approximated (usually a kernel matrix), and P is a permutation matrix.

近期下载者：

相关文件：

评论：[我要评论] [举报此文件]

收藏者：