<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<title>Test training params on subsets of the training data.</title>
</head>
<body><div class="manualnavbar" style="text-align: center;">
<div class="prev" style="text-align: left; float: left;"><a href="svm.construct.html">SVM::__construct</a></div>
<div class="next" style="text-align: right; float: right;"><a href="svm.getoptions.html">SVM::getOptions</a></div>
<div class="up"><a href="class.svm.html">SVM</a></div>
<div class="home"><a href="index.html">PHP Manual</a></div>
</div><hr /><div id="svm.crossvalidate" class="refentry">
<div class="refnamediv">
<h1 class="refname">SVM::crossvalidate</h1>
<p class="verinfo">(PECL svm >= 0.1.0)</p><p class="refpurpose"><span class="refname">SVM::crossvalidate</span> — <span class="dc-title">Test training params on subsets of the training data.</span></p>
</div>
<div class="refsect1 description" id="refsect1-svm.crossvalidate-description">
<h3 class="title">Description</h3>
<div class="methodsynopsis dc-description">
<span class="modifier">public</span> <span class="type">float</span> <span class="methodname"><strong>svm::crossvalidate</strong></span>
( <span class="methodparam"><span class="type">array</span> <code class="parameter">$problem</code></span>
, <span class="methodparam"><span class="type">int</span> <code class="parameter">$number_of_folds</code></span>
)</div>
<p class="para rdfs-comment">
Crossvalidate can be used to test the effectiveness of the current
parameter set on a subset of the training data. Given a problem set
and a n "folds", it separates the problem set into n subsets, and the
repeatedly trains on one subset and tests on another. While the accuracy
will generally be lower than a SVM trained on the enter data set, the
accuracy score returned should be relatively useful, so it can be used to
test different training parameters.
</p>
</div>
<div class="refsect1 parameters" id="refsect1-svm.crossvalidate-parameters">
<h3 class="title">Parameters</h3>
<p class="para">
<dl>
<dt>
<span class="term"><em><code class="parameter">problem</code></em></span>
<dd>
<p class="para">
The problem data. This can either be in the form of an array, the URL
of an SVMLight formatted file, or a stream to an opened SVMLight
formatted datasource.
</p>
</dd>
</dt>
<dt>
<span class="term"><em><code class="parameter">number_of_folds</code></em></span>
<dd>
<p class="para">
The number of sets the data should be divided into and cross tested.
A higher number means smaller training sets and less reliability. 5 is
a good number to start with.
</p>
</dd>
</dt>
</dl>
</p>
</div>
<div class="refsect1 returnvalues" id="refsect1-svm.crossvalidate-returnvalues">
<h3 class="title">Return Values</h3>
<p class="para">
The correct percentage, expressed as a floating point number from 0-1.
In the case of NU_SVC or EPSILON_SVR kernels the mean squared error will
returned instead.
</p>
</div>
<div class="refsect1 seealso" id="refsect1-svm.crossvalidate-seealso">
<h3 class="title">See Also</h3>
<p class="para">
<ul class="simplelist">
<li class="member"> <span class="methodname"><a href="svm.train.html" class="methodname" rel="rdfs-seeAlso">SVM::train()</a> - Create a SVMModel based on training data</span></li>
</ul>
</p>
</div>
</div><hr /><div class="manualnavbar" style="text-align: center;">
<div class="prev" style="text-align: left; float: left;"><a href="svm.construct.html">SVM::__construct</a></div>
<div class="next" style="text-align: right; float: right;"><a href="svm.getoptions.html">SVM::getOptions</a></div>
<div class="up"><a href="class.svm.html">SVM</a></div>
<div class="home"><a href="index.html">PHP Manual</a></div>
</div></body></html>
