mirror of
https://github.com/fofolee/uTools-Manuals.git
synced 2025-06-20 07:06:50 +08:00
119 lines
5.0 KiB
HTML
119 lines
5.0 KiB
HTML
<!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>在整个数据集上训练，使用一段时间的 Cascade2 训练算法。</title>
|
||
</head>
|
||
<body class="docs"><div id="layout">
|
||
<div id="layout-content"><div id="function.fann-cascadetrain-on-data" class="refentry">
|
||
<div class="refnamediv">
|
||
<h1 class="refname">fann_cascadetrain_on_data</h1>
|
||
<p class="verinfo">(PECL fann >= 1.0.0)</p><p class="refpurpose"><span class="refname">fann_cascadetrain_on_data</span> — <span class="dc-title">在整个数据集上训练，使用一段时间的 Cascade2 训练算法。</span></p>
|
||
|
||
</div>
|
||
|
||
<div class="refsect1 description" id="refsect1-function.fann-cascadetrain-on-data-description">
|
||
<h3 class="title">说明</h3>
|
||
<div class="methodsynopsis dc-description">
|
||
<span class="methodname"><strong>fann_cascadetrain_on_data</strong></span>
|
||
( <span class="methodparam"><span class="type">resource</span> <code class="parameter">$ann</code></span>
|
||
, <span class="methodparam"><span class="type">resource</span> <code class="parameter">$data</code></span>
|
||
, <span class="methodparam"><span class="type">int</span> <code class="parameter">$max_neurons</code></span>
|
||
, <span class="methodparam"><span class="type">int</span> <code class="parameter">$neurons_between_reports</code></span>
|
||
, <span class="methodparam"><span class="type">float</span> <code class="parameter">$desired_error</code></span>
|
||
) : <span class="type">bool</span></div>
|
||
|
||
<p class="para rdfs-comment">
|
||
级联输出改变小数是一个0到1之间的数字，表示在输出连接的训练中，为了使训练不停滞的情况下，经过 <span class="function"><a href="fann_get_cascade_output_stagnation_epochs.html" class="function">fann_get_cascade_output_stagnation_epochs()</a></span> 次迭代的后，<span class="function"><a href="fann_get_mse.html" class="function">fann_get_MSE()</a></span> 将会改变多大。如果训练停滞了，训练的输出连接将会结束，新的候选神经元将会准备好。
|
||
</p>
|
||
<p class="para">
|
||
该训练使用由 fann_set_cascade_ 前缀设置的参数，但它也采用了另一种训练算法，即内部训练算法。该训练算法要么是 <span class="function"><a href="fann_set_training_algorithm.html" class="function">fann_set_training_algorithm()</a></span> 设置的 <strong><code>FANN_TRAIN_RPROP</code></strong> 算法，要么是 <strong><code>FANN_TRAIN_QUICKPROP</code></strong>，这些算法设置的参数同样也会影响到级联训练。
|
||
</p>
|
||
</div>
|
||
|
||
|
||
<div class="refsect1 parameters" id="refsect1-function.fann-cascadetrain-on-data-parameters">
|
||
<h3 class="title">参数</h3>
|
||
<dl>
|
||
|
||
|
||
<dt>
|
||
<code class="parameter">ann</code></dt>
|
||
|
||
<dd>
|
||
|
||
<p class="para">Neural network <span class="type"><a href="language.types.resource.html" class="type resource">resource</a></span>.</p>
|
||
</dd>
|
||
|
||
|
||
|
||
<dt>
|
||
<code class="parameter">data</code></dt>
|
||
|
||
<dd>
|
||
|
||
<p class="para">Neural network training data <span class="type"><a href="language.types.resource.html" class="type resource">resource</a></span>.</p>
|
||
</dd>
|
||
|
||
|
||
|
||
<dt>
|
||
<code class="parameter">max_neurons</code></dt>
|
||
|
||
<dd>
|
||
|
||
<p class="para">
|
||
被添加入神经网络中最大的神经元数。
|
||
</p>
|
||
</dd>
|
||
|
||
|
||
|
||
<dt>
|
||
<code class="parameter">neurons_between_reports</code></dt>
|
||
|
||
<dd>
|
||
|
||
<p class="para">
|
||
打印状态报告之间的神经元数。0表示没有报告会被打印。
|
||
</p>
|
||
</dd>
|
||
|
||
|
||
|
||
<dt>
|
||
<code class="parameter">desired_error</code></dt>
|
||
|
||
<dd>
|
||
|
||
<p class="para">
|
||
预期的 <span class="function"><a href="fann_get_mse.html" class="function">fann_get_MSE()</a></span> 或 <span class="function"><a href="fann_get_bit_fail.html" class="function">fann_get_bit_fail()</a></span>,
|
||
取决于 <span class="function"><a href="fann_set_train_stop_function.html" class="function">fann_set_train_stop_function()</a></span> 选择的停止函数
|
||
</p>
|
||
</dd>
|
||
|
||
|
||
</dl>
|
||
|
||
</div>
|
||
|
||
|
||
<div class="refsect1 returnvalues" id="refsect1-function.fann-cascadetrain-on-data-returnvalues">
|
||
<h3 class="title">返回值</h3>
|
||
<p class="para">Returns <strong><code>TRUE</code></strong> on success, or <strong><code>FALSE</code></strong> otherwise.</p>
|
||
</div>
|
||
|
||
|
||
<div class="refsect1 seealso" id="refsect1-function.fann-cascadetrain-on-data-seealso">
|
||
<h3 class="title">参见</h3>
|
||
<p class="para">
|
||
<ul class="simplelist">
|
||
<li class="member"><span class="function"><a href="fann_train_on_data.html" class="function" rel="rdfs-seeAlso">fann_train_on_data()</a> - 在整个数据集上训练一段时间。</span></li>
|
||
<li class="member"><span class="function"><a href="fann_cascadetrain_on_file.html" class="function" rel="rdfs-seeAlso">fann_cascadetrain_on_file()</a> - 读取文件并在整个数据集上训练，使用 Cascade2 训练算法训练一段时间。</span></li>
|
||
</ul>
|
||
</p>
|
||
</div>
|
||
|
||
|
||
|
||
</div></div></div></body></html> |