Support Vector Machine toolbox for Matlab
Version 2.51, January 2002
Contents.m contains a brief description of all parts of this toolbox.
Main features are:
- Except for the QP solver, all parts are written in plain Matlab. This
guarantees for easy modification. Special kinds of kernels that require
much computation (such as the Fisher kernel, which is based on a model of
the data) can easily be incorporated.
- Extension to multi-class problems via error correcting output codes is
included.
- Unless many other SVM toolboxes, this one can handle SVMs with 1norm
or 2norm of the slack variables.
- For both cases, a decomposition algorithm is implemented for the training
routine, together with efficient working set selection strategies.
The training algorithm uses many of the ideas proposed by Thorsten
Joachims for his SVMlight. It thus should exhibit a scaling behaviour that
is comparable to SVMlight.
This toolbox optionally makes use of a Matlab wrapper for an interior point
code in LOQO style (Matlab wrapper by Steve Gunn, LOQO code by Alex Smola).
To compile the wrapper, run
mex loqo.c pr_loqo.c
Make sure you have turned on the compiler optimizations in mexopts.sh
The LOQO code can be retrieved from
http://www.kernel-machines.org/code/prloqo.tar.gz
The wrapper comes directly from Steve Gunn.
Copyright (c) Anton Schwaighofer (2001)
mailto:anton.schwaighofer@gmx.net
This program is released unter the GNU General Public License.
See License.txt for details.
Changes in version 2.51:
- fixed bug in SVMTRAIN that prevented correct initialisation with
NET.recompute==Inf
Changes in version 2.5:
- Handling of multi-class problems with ECOC
- NET.recompute is set to Inf by default, thus all training is done
incrementally by default.
- Handling the case of all training examples being -1 or +1 correctly
Changes in version 2.4:
- Better selection of the initial working set
- Added workaround for a (rare) Matlab quadprog bug with badly conditioned
matrices
- There is now a new kernel function 'rbffull' where a full matrix
("covariance matrix") C may be put into an RBF kernel:
K(X1,X2) = exp(-(X1-X2)'*C*(X1-X2))
Changes in version 2.3:
- slightly more compact debug output
Changes in version 2.2:
- New default values for parameter qpsize that make the whole toolbox
*much* faster
- Workaround for a Matlab bug with sparse matrices
- Changed the definition of the RBF-Kernel: from |x-y|^2/(2*nin*param^2)
to |x-y|^2/(nin*param). This means that all parameter settings for old
versions need to be updated!
- A few minor things I can't remember
Changes in version 2.1:
Fixed a nasty bug at the KKT check
Changes in version 2.0:
All relevant routines have been updated to allow the use of a SVM with
2norm of the slack variables (NET.use2norm==1).
没有合适的资源?快使用搜索试试~ 我知道了~
演示程序和一些基本的函数svm_toolbox.rar
共25个文件
m:16个
txt:3个
c:2个
4星 · 超过85%的资源 需积分: 5 19 下载量 54 浏览量
2009-11-22
10:12:36
上传
评论 2
收藏 244KB RAR 举报
温馨提示
演示程序和一些基本的函数svm_toolbox.rar演示程序和一些基本的函数svm_toolbox.rar演示程序和一些基本的函数svm_toolbox.rar
资源推荐
资源详情
资源评论
收起资源包目录
演示程序和一些基本的函数svm_toolbox.rar (25个子文件)
演示程序和一些基本的函数svm_toolbox
svm
loqo.c 7KB
ecoctrain.m 3KB
pr_loqo.c 16KB
code14-11 319B
demsvm3.m 1KB
Contents.m 2KB
svmkernel.m 3KB
svmfwd.m 2KB
demecoc1.m 1KB
License.txt 18KB
demsvm1.m 8KB
demecoc1.mat 86KB
ecoc.m 2KB
svmstat.m 4KB
pr_loqo.h 2KB
Readme.txt 3KB
consist.m 3KB
ecoc-codes.tar.gz 309KB
ecocload.m 3KB
svm.m 5KB
svmcv.m 8KB
demsvm2.m 11KB
svmtrain.m 21KB
ecocfwd.m 2KB
www.pudn.com.txt 218B
共 25 条
- 1
资源评论
- xyhit2013-01-10比较详细的程序,简单易懂
- ly42322013-01-09对学习SVM很有帮助!
wq33213
- 粉丝: 0
- 资源: 11
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功