libsvm_matlab
所属分类:matlab编程
开发工具:matlab
文件大小:29KB
下载次数:393
上传日期:2010-05-04 10:25:10
上 传 者:
xmfch
说明: 一个MATLAB的支持向量机程序,并包含几个使用实例
(A MATLAB support vector machine procedures, and contains several use cases)
文件列表:
libsvm_matlab\contents.m (489, 2005-09-12)
libsvm_matlab\demo1.m (2587, 2005-09-09)
libsvm_matlab\demo2.m (2418, 2005-09-12)
libsvm_matlab\demo3.m (1924, 2005-09-13)
libsvm_matlab\demo4.m (1960, 2005-09-05)
libsvm_matlab\demo5.m (2036, 2005-09-12)
libsvm_matlab\libsvmopt.m (4447, 2005-09-12)
libsvm_matlab\libsvmsim.m (1490, 2005-09-12)
libsvm_matlab\mexlibsvm.m (1286, 2005-09-13)
libsvm_matlab\source\libsvmaux.c (27405, 2005-09-13)
libsvm_matlab\source\libsvmaux.h (3687, 2005-09-09)
libsvm_matlab\source\libsvmopt.c (8664, 2005-09-05)
libsvm_matlab\source\libsvmsim.c (5979, 2005-09-13)
libsvm_matlab\libsvm\libsvm.txt (119, 2005-09-12)
libsvm_matlab\docs\changes.txt (1354, 2005-09-13)
libsvm_matlab\docs\license.txt (17995, 2005-08-23)
libsvm_matlab\source (0, 2009-07-18)
libsvm_matlab\libsvm (0, 2009-07-18)
libsvm_matlab\docs (0, 2009-07-18)
libsvm_matlab (0, 2009-07-18)
MATLAB Interface for LIBSVM, Version 1.2
========================================
Copyright (C) 2004-2005 Michael Vogt
Written by Michael Vogt, Atanas Ayarov and Bennet Gedan
This software package provides a MATLAB interface for LIBSVM,
a library for the optimization of support vector machines.
DESCRIPTION
The interface offers two commands LIBSVMOPT and LIBSVMSIM for the
optimization and the simulation of support vector machines. The usage
is explained in the MATLAB help files. Because both commands are
implemented as C MEX files, they need to be compiled from the sources
before. Several demos show the capabilites of LIBSVM.
The package is distributed as source code under the GNU General Public
License, see "license.txt" for details. Since LIBSVM uses a different
type of license, it is not included. To build the interface, you will
need LIBSVM 2.8 (or maybe a later version) available at
http://www.csie.ntu.edu.tw/~cjlin/libsvm
Most features of LIBSVM are accessible. Currently, two things are
missing: Multiclass SVMs and support for sparse MATLAB matrices.
SYSTEM REQUIREMENTS
1. MATLAB 6.0 or higher.
2. A C++ compiler. On Microsoft Windows systems, Microsoft Visual
C/C++ 6.0 and Borland Free Commandline Tools 5.5 have been testet.
On Linux it has been compiled with GCC 3.3.
INSTALLATION
1. Configure the C++ compiler by typing "mex -setup" in the MATLAB
command window and following the instructions.
2. Extract all files of the interface in a directory of your choice,
and LIBSVM ("svm.cpp" and "svm.h") in the subdirectory "libsvm".
3. Add the main directory to the MATLAB search path.
4. Call "mexlibsvm" in order to compile the MEX functions.
近期下载者:
相关文件:
收藏者: