📄 matlab support vector machine toolbox.htm
字号:
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<!-- saved from url=(0046)http://theoval.sys.uea.ac.uk/~gcc/svm/toolbox/ -->
<HTML><HEAD><TITLE>MATLAB Support Vector Machine Toolbox</TITLE>
<META http-equiv=Content-Type content="text/html; charset=gb2312">
<META content="MSHTML 6.00.2900.2722" name=GENERATOR></HEAD>
<BODY text=#000000 vLink=#ff0000 link=#0000ff bgColor=#ffffff><BR><IMG
src="MATLAB Support Vector Machine Toolbox.files/uea_logo.gif" align=right>
<H2>MATLAB Support Vector Machine Toolbox</H2>
<H3><B>Dr Gavin Cawley</B></H3>
<H4>School of Information Systems,<BR>University of East Anglia</H4><BR>
<HR>
<P>
<H4>Introduction</H4>This is a (very) beta version of a MATLAB toolbox
implementing Vapnik's support vector machine, as described in [1]. Training is
performed using the SMO algorithm, due to Platt [2], implemented as a mex file
(for speed). Before you use the toolbox you need to run the compilemex script to
recompile them (if there are problems running this script, make sure you have
the mex compiler set up correctly - you may need to see your sys-admin to do
this). At the moment this is the only documentation for the toolbox but the file
demo.m provides a simple demonstration that ought to be enough to get started.
For a good introduction to support vector machines, see the excellent <A
href="http://support-vector.net/">book</A> by Cristianini and Shawe-Taylor [3].
Key features of this toolbox:
<UL>
<LI>C++ MEX implementation of the SMO training algorithm, with caching of
kernel evaluations for efficiency.
<LI>Support for multi-class support vector classification using max wins,
pairwise [4] and DAG-SVM [5] algorithms.
<LI>A model selection criterion (the xi-alpha bound [6,7] on the leave-one-out
cross-validation error).
<LI>Object oriented design, currently this just means that you can supply
bespoke kernel functions for particular applications, but will in future
releases also support a range of training algorithms, model selection criteria
etc. </LI></UL>
<H4>Licensing Arrangements</H4>The toolbox is provided free for non-commercial
use under the terms of the GNU <A
href="http://www.gnu.org/copyleft/gpl.html">General Public License</A> (<A
href="http://theoval.sys.uea.ac.uk/~gcc/svm/toolbox/license.txt">license.txt</A>),
however, I would be grateful if:
<UL>
<LI>you let me know about any bugs you find,
<LI>you send suggestions of ideas to improve the toolbox (e.g. references to
other training algorithms),
<LI>reference the toolbox web page in any publication describing research
performed using the toolbox, or software derived from the toolbox. A suitable
BibTeX entry would look something like this: </LI></UL><PRE>@misc{Cawley2000,
author = "Cawley, G. C.",
title = "{MATLAB} Support Vector Machine Toolbox (v0.55$\beta$) $[$
\texttt{http://theoval.sys.uea.ac.uk/\~{}gcc/svm/toolbox}$]$",
howpublished = "University of East Anglia, School of Information Systems,
Norwich, Norfolk, U.K. NR4 7TJ",
year = 2000
}
</PRE>
<H4>Download</H4>
<H5>Current Version</H5>
<UL>
<LI>v0.55beta ( <A
href="http://theoval.sys.uea.ac.uk/~gcc/svm/toolbox/svm_v0.55beta.tar.gz">svm_v0.55beta.tar.gz</A>,
<A
href="http://theoval.sys.uea.ac.uk/~gcc/svm/toolbox/svm_v0.55beta.zip">svm_v0.55beta.zip</A>)<BR>New
in this version:
<UL>
<LI>compilemex script to recompile all mex files implemented in the toolbox.
</LI></UL></LI></UL>
<H5>Previous Releases</H5>
<UL>
<LI>v0.54beta (<A
href="http://theoval.sys.uea.ac.uk/~gcc/svm/toolbox/svm_v0.54beta.tar.gz">svm_v0.54beta.tar.gz</A>)<BR>New
in this version:
<UL>
<LI>.mexsol binaries for Solaris </LI></UL>
<LI>v0.53beta (<A
href="http://theoval.sys.uea.ac.uk/~gcc/svm/toolbox/svm_v0.53beta.tar.gz">svm_v0.53beta.tar.gz</A>)<BR>New
in this version:
<UL>
<LI>nothing! (minor bug-fix in <CODE>@dagsvm/train.m</CODE>) </LI></UL>
<LI>v0.52beta (<A
href="http://theoval.sys.uea.ac.uk/~gcc/svm/toolbox/svm_v0.52beta.tar.gz">svm_v0.52beta.tar.gz</A>)<BR>New
in this version:
<UL>
<LI>A simple demo of the DAG-SVM algorithm for multi-class pattern
recognition (<CODE>dagsvmdemo.m</CODE>). </LI></UL>
<LI>v0.51beta (<A
href="http://theoval.sys.uea.ac.uk/~gcc/svm/toolbox/svm_v0.51beta.tar.gz">svm_v0.51beta.tar.gz</A>)<BR>New
in this version:
<UL>
<LI>Compiled mex files for Windows systems
<UL>
<LI>Uses a relatively small cache size for SMO training (32Mb), this will
be adjustable in a future release.
<LI>Caveat emptor: runs demo.m, but not thoroughly tested.
</LI></UL></LI></UL>
<LI>v0.50beta (<A
href="http://theoval.sys.uea.ac.uk/~gcc/svm/toolbox/svm_v0.50beta.tar.gz">svm_v0.50beta.tar.gz</A>)<BR>New
in this version:
<UL>
<LI>Support for multi-class pattern recognition using maxwins, pairwise [4]
and DAG-SVM [5] algorithms.
<LI>A model selection criterion (the xi-alpha bound [6,7] on the
leave-one-out cross-validation error). </LI></UL>
<LI>v0.01beta (<A
href="http://theoval.sys.uea.ac.uk/~gcc/svm/toolbox/svm_v0.01beta.tar.gz">svm_v0.01beta.tar.gz</A>)
</LI></UL>
<H4>"To Do" List</H4>
<OL>
<LI>Find time to write a proper list of things to do!
<LI>Documentation.
<LI>Support Vector Regression.
<LI>Automated model selection.
<LI>Sparse matrix support </LI></OL>
<H4>References</H4><PRE>[1] V.N. Vapnik,
"The Nature of Statistical Learning Theory",
Springer-Verlag, New York, ISBN 0-387-94559-8,
1995.
[2] J. C. Platt,
"Fast training of support vector machines using sequential minimal
optimization", in Advances in Kernel Methods - Support Vector Learning,
(Eds) B. Scholkopf, C. Burges, and A. J. Smola, MIT Press, Cambridge,
Massachusetts, chapter 12, pp 185-208, 1999.
[3] N. Cristianini and J. Shawe-Taylor,
"Support Vector Machines and other kernel-based learning methods",
Cambridge University Press, ISBN 0-521-78019-5,
2000. (<A href="http://support-vector.net/">Web site</A>)
[4] U. Kressel,
"Pairwise Classification and Support Vector Machines",
in Advances in Kernel Methods - Support Vector Learning,
(Eds) B. Scholkopf, C. Burges, and A. J. Smola, MIT Press, Cambridge,
Massachusetts, chapter 15, 1999.
[5] J. Platt, N. Cristianini, J. Shawe-Taylor,
"Large Margin DAGs for Multiclass Classification",
in Advances in Neural Information Processing Systems 12, pp. 547-553,
MIT Press, 2000.
(<A href="http://www.cs.cmu.edu/Web/Groups/NIPS/NIPS99/99papers-pub-on-web/Named-gz/PlattCristianiniShawe-Taylor.ps.gz">PlattCristianiniShawe-Taylor.ps.gz</A>).
[6] T. Joachims,
"Estimating the Generalization Performance of a SVM Efficiently",
25, Universit鋞 Dortmund, LS VIII, 1999.
[7] T. Joachims,
"Estimating the Generalization Performance of a SVM Efficiently",
in Proceedings of the International Conference on Machine Learning,
Morgan Kaufman, 2000.
</PRE>
<HR>
<CENTER>
<ADDRESS>Gavin Cawley / <A href="mailto:gcc@sys.uea.ac.uk">gcc@sys.uea.ac.uk</A>
</ADDRESS></CENTER></BODY></HTML>
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -