SVM工具箱快速入手简易教程

SVM工具箱快速入手简易教程(by faruto)

一. matlab 自带的函数(matlab帮助文件里的例子)[只有较新版本的matlab中有这两个SVM的函数]

=====

svmtrain svmclassify

=====简要语法规则====

svmtrain

Train support vector machine classifier

Syntax

SVMStruct = svmtrain(Training, Group)

SVMStruct = svmtrain(..., 'Kernel_Function', Kernel_FunctionValue, ...)

SVMStruct = svmtrain(..., 'RBF_Sigma', RBFSigmaValue, ...)

SVMStruct = svmtrain(..., 'Polyorder', PolyorderValue, ...)

SVMStruct = svmtrain(..., 'Mlp_Params', Mlp_ParamsValue, ...)

SVMStruct = svmtrain(..., 'Method', MethodValue, ...)

SVMStruct = svmtrain(..., 'QuadProg_Opts', QuadProg_OptsValue, ...)

SVMStruct = svmtrain(..., 'SMO_Opts', SMO_OptsValue, ...)

SVMStruct = svmtrain(..., 'BoxConstraint', BoxConstraintValue, ...)

SVMStruct = svmtrain(..., 'Autoscale', AutoscaleValue, ...)

SVMStruct = svmtrain(..., 'Showplot', ShowplotValue, ...)

---------------------

svmclassify

Classify data using support vector machine

Syntax

Group = svmclassify(SVMStruct, Sample)

Group = svmclassify(SVMStruct, Sample, 'Showplot', ShowplotValue)

============================实例研究====================

load fisheriris

%载入matlab自带的数据[有关数据的信息可以自己到UCI查找,这是UCI的经典数据之一],得到的数据如下图:

tu1

1.jpg (7.94 KB)
2009-5-12 19:50

其中meas是150*4的矩阵代表着有150个样本每个样本有4个属性描述,species代表着这150个样本的分类.

data = [meas(:,1), meas(:,2)];

%在这里只取meas的第一列和第二列,即只选取前两个属性.

groups = ismember(species,'setosa');

%由于species分类中是有三个分类:setosa,versicolor,virginica,为了使问题简单,我们将其变为二分类问题:Setosa and non-Setosa.

[train, test] = crossvalind('holdOut',groups);

cp = classperf(groups);

%随机选择训练集合测试集[有关crossvalind的使用请自己help一下]

其中cp作用是后来用来评价分类器的.

svmStruct = svmtrain(data(train,:),groups(train),'showplot',true);

%使用svmtrain进行训练,得到训练后的结构svmStruct,在预测时使用.

训练结果如图:

tu2

2.jpg (26.86 KB)
2009-5-12 19:50

classes = svmclassify(svmStruct,data(test,:),'showplot',true);

%对于未知的测试集进行分类预测,结果如图:

tu3

3.jpg (37.34 KB)
2009-5-12 19:50

classperf(cp,classes,test);

cp.CorrectRate

ans =

0.9867

%分类器效果测评,就是看测试集分类的准确率的高低.

二.台湾林智仁的libsvm工具箱

该工具箱下载[libsvm-mat-2.86-1]:
libsvm-mat-2.86-1.rar(73.75 KB)

libsvm-mat-2.86-1.rar(73.75 KB)

下载次数: 373
2009-5-12 20:02

安装方法也很简单,解压文件,把当前工作目录调整到libsvm所在的文件夹下,再在set path里将libsvm所在的文件夹加到里面.然后

在命令行里输入

mex -setup %选择一下编译器

make

这样就可以了.

建议大家使用libsvm工具箱,这个更好用一些.可以进行分类[多类别],预测....

=========

svmtrain

svmpredict

================

简要语法:

Usage

=====

matlab> model = svmtrain(training_label_vector, training_instance_matrix [,'libsvm_options']);

-training_label_vector:

An m by 1 vector oftraining labels (type must be double).

-training_instance_matrix:

An m by n matrix of mtraining instances with n features.

It can be dense or sparse(type must be double).

-libsvm_options:

A string of trainingoptions in the same format as that of LIBSVM.

matlab> [predicted_label, accuracy, decision_values/prob_estimates] =svmpredict(testing_label_vector, testing_instance_matrix, model [,'libsvm_options']);

-testing_label_vector:

An m by 1 vector ofprediction labels. If labels of test

data are unknown, simplyuse any random values. (type must be double)

-testing_instance_matrix:

An m by n matrix of mtesting instances with n features.

It can be dense or sparse.(type must be double)

-model:

The output of svmtrain.

-libsvm_options:

A string of testing optionsin the same format as that of LIBSVM.

Returned Model Structure

========================

实例研究:

load heart_scale.mat

%工具箱里自带的数据

如图:

tu4

4.jpg(9.36 KB)
2009-5-12 20:08

其中 heart_scale_inst是样本,heart_scale_label是样本标签

model = svmtrain(heart_scale_label, heart_scale_inst, '-c 1 -g 0.07');

%训练样本,具体参数的调整请看帮助文件

[predict_label, accuracy, dec_values] = svmpredict(heart_scale_label,heart_scale_inst, model);

%分类预测,这里把训练集当作测试集,验证效果如下:

>> [predict_label, accuracy, dec_values] = svmpredict(heart_scale_label,heart_scale_inst, model); % test the training data

Accuracy = 86.6667% (234/270) (classification)

==============

这回把SVM这点入门的东西都说完了,大家可以参照着上手了,有关SVM的原理我下面有个简易的PPT,是以前做项目时我做的[当时我负责有关SVM这一块代码实现讲解什么的],感兴趣的你可以看看,都是上手较快的东西,想要深入学习SVM,你的学习统计学习理论什么的....挺多的呢..

SVM.ppt(391 KB)

SVM.ppt(391 KB)

下载次数: 429
2009-5-12 20:18

-----------有关SVM和libsvm的非常好的资料,想要详细研究SVM看这个------

libsvm_guide.pdf(194.53 KB)

libsvm_guide.pdf(194.53 KB)

下载次数:186
2009-8-19 14:58

libsvm_library.pdf(316.82 KB)

libsvm_library.pdf(316.82 KB)

下载次数: 137
2009-8-19 14:58

OptimizationSupportVectorMachinesandMachine