Google Git
Sign in
apache / mxnet-test / refs/heads/asf-site / . / versions / master / tutorials / r / CallbackFunctionTutorial.html
blob: b4a909d7db69cd958e62489e09dd391d22f3ff60 [file] [log] [blame]
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="utf-8"/>
<meta content="IE=edge" http-equiv="X-UA-Compatible"/>
<meta content="width=device-width, initial-scale=1" name="viewport"/>
<title>Callback Function — mxnet documentation</title>
<link crossorigin="anonymous" href="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.6/css/bootstrap.min.css" integrity="sha384-1q8mTJOASx8j1Au+a5WDVnPi2lkFfwwEAa8hDDdjZlpLegxhjVME1fgjWPGmkzs7" rel="stylesheet"/>
<link href="https://maxcdn.bootstrapcdn.com/font-awesome/4.5.0/css/font-awesome.min.css" rel="stylesheet"/>
<link href="../../_static/basic.css" rel="stylesheet" type="text/css">
<link href="../../_static/pygments.css" rel="stylesheet" type="text/css">
<link href="../../_static/mxnet.css" rel="stylesheet" type="text/css"/>
<script type="text/javascript">
var DOCUMENTATION_OPTIONS = {
URL_ROOT: '../../',
VERSION: '',
COLLAPSE_INDEX: false,
FILE_SUFFIX: '.html',
HAS_SOURCE: true,
SOURCELINK_SUFFIX: ''
};
</script>
<script src="../../_static/jquery-1.11.1.js" type="text/javascript"></script>
<script src="../../_static/underscore.js" type="text/javascript"></script>
<script src="../../_static/searchtools_custom.js" type="text/javascript"></script>
<script src="../../_static/doctools.js" type="text/javascript"></script>
<script src="../../_static/selectlang.js" type="text/javascript"></script>
<script src="https://cdn.mathjax.org/mathjax/latest/MathJax.js?config=TeX-AMS-MML_HTMLorMML" type="text/javascript"></script>
<script type="text/javascript"> jQuery(function() { Search.loadIndex("/searchindex.js"); Search.init();}); </script>
<script>
(function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){
(i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new
Date();a=s.createElement(o),
m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m)
})(window,document,'script','https://www.google-analytics.com/analytics.js','ga');
ga('create', 'UA-96378503-1', 'auto');
ga('send', 'pageview');
</script>
<!-- -->
<!-- <script type="text/javascript" src="../../_static/jquery.js"></script> -->
<!-- -->
<!-- <script type="text/javascript" src="../../_static/underscore.js"></script> -->
<!-- -->
<!-- <script type="text/javascript" src="../../_static/doctools.js"></script> -->
<!-- -->
<!-- <script type="text/javascript" src="https://cdn.mathjax.org/mathjax/latest/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script> -->
<!-- -->
<link href="https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/mxnet-icon.png" rel="icon" type="image/png"/>
</link></link></head>
<body role="document"><!-- Previous Navbar Layout
<div class="navbar navbar-default navbar-fixed-top">
<div class="container">
<div class="navbar-header">
<button type="button" class="navbar-toggle collapsed" data-toggle="collapse" data-target="#navbar" aria-expanded="false" aria-controls="navbar">
<span class="sr-only">Toggle navigation</span>
<span class="icon-bar"></span>
<span class="icon-bar"></span>
<span class="icon-bar"></span>
</button>
<a href="../../" class="navbar-brand">
<img src="http://data.mxnet.io/theme/mxnet.png">
</a>
</div>
<div id="navbar" class="navbar-collapse collapse">
<ul id="navbar" class="navbar navbar-left">
<li> <a href="../../get_started/index.html">Get Started</a> </li>
<li> <a href="../../tutorials/index.html">Tutorials</a> </li>
<li> <a href="../../how_to/index.html">How To</a> </li>
<li class="dropdown">
<a href="#" class="dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">Packages <span class="caret"></span></a>
<ul class="dropdown-menu">
<li><a href="../../packages/python/index.html">
Python
</a></li>
<li><a href="../../packages/r/index.html">
R
</a></li>
<li><a href="../../packages/julia/index.html">
Julia
</a></li>
<li><a href="../../packages/c++/index.html">
C++
</a></li>
<li><a href="../../packages/scala/index.html">
Scala
</a></li>
<li><a href="../../packages/perl/index.html">
Perl
</a></li>
</ul>
</li>
<li> <a href="../../system/index.html">System</a> </li>
<li>
<form class="" role="search" action="../../search.html" method="get" autocomplete="off">
<div class="form-group inner-addon left-addon">
<i class="glyphicon glyphicon-search"></i>
<input type="text" name="q" class="form-control" placeholder="Search">
</div>
<input type="hidden" name="check_keywords" value="yes" />
<input type="hidden" name="area" value="default" />
</form> </li>
</ul>
<ul id="navbar" class="navbar navbar-right">
<li> <a href="../../index.html"><span class="flag-icon flag-icon-us"></span></a> </li>
<li> <a href="../..//zh/index.html"><span class="flag-icon flag-icon-cn"></span></a> </li>
</ul>
</div>
</div>
</div>
Previous Navbar Layout End -->
<div class="navbar navbar-fixed-top">
<div class="container" id="navContainer">
<div class="innder" id="header-inner">
<h1 id="logo-wrap">
<a href="../../" id="logo"><img src="http://data.mxnet.io/theme/mxnet.png"/></a>
</h1>
<nav class="nav-bar" id="main-nav">
<a class="main-nav-link" href="../../get_started/install.html">Install</a>
<a class="main-nav-link" href="../../tutorials/index.html">Tutorials</a>
<a class="main-nav-link" href="../../how_to/index.html">How To</a>
<span id="dropdown-menu-position-anchor">
<a aria-expanded="true" aria-haspopup="true" class="main-nav-link dropdown-toggle" data-toggle="dropdown" href="#" role="button">API <span class="caret"></span></a>
<ul class="dropdown-menu" id="package-dropdown-menu">
<li><a class="main-nav-link" href="../../api/python/index.html">Python</a></li>
<li><a class="main-nav-link" href="../../api/scala/index.html">Scala</a></li>
<li><a class="main-nav-link" href="../../api/r/index.html">R</a></li>
<li><a class="main-nav-link" href="../../api/julia/index.html">Julia</a></li>
<li><a class="main-nav-link" href="../../api/c++/index.html">C++</a></li>
<li><a class="main-nav-link" href="../../api/perl/index.html">Perl</a></li>
</ul>
</span>
<a class="main-nav-link" href="../../architecture/index.html">Architecture</a>
<!-- <a class="main-nav-link" href="../../community/index.html">Community</a> -->
<a class="main-nav-link" href="https://github.com/dmlc/mxnet">Github</a>
<span id="dropdown-menu-position-anchor-version" style="position: relative"><a href="#" class="main-nav-link dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">Versions(master)<span class="caret"></span></a><ul id="package-dropdown-menu" class="dropdown-menu"><li><a class="main-nav-link" href=http://mxnet.incubator.apache.org/test/>v0.10.14</a></li><li><a class="main-nav-link" href=http://mxnet.incubator.apache.org/test/versions/0.10/index.html>0.10</a></li><li><a class="main-nav-link" href=http://mxnet.incubator.apache.org/test/versions/master/index.html>master</a></li></ul></span></nav>
<script> function getRootPath(){ return "../../" } </script>
<div class="burgerIcon dropdown">
<a class="dropdown-toggle" data-toggle="dropdown" href="#" role="button">☰</a>
<ul class="dropdown-menu dropdown-menu-right" id="burgerMenu">
<li><a href="../../get_started/install.html">Install</a></li>
<li><a href="../../tutorials/index.html">Tutorials</a></li>
<li><a href="../../how_to/index.html">How To</a></li>
<li class="dropdown-submenu">
<a href="#" tabindex="-1">API</a>
<ul class="dropdown-menu">
<li><a href="../../api/python/index.html" tabindex="-1">Python</a>
</li>
<li><a href="../../api/scala/index.html" tabindex="-1">Scala</a>
</li>
<li><a href="../../api/r/index.html" tabindex="-1">R</a>
</li>
<li><a href="../../api/julia/index.html" tabindex="-1">Julia</a>
</li>
<li><a href="../../api/c++/index.html" tabindex="-1">C++</a>
</li>
<li><a href="../../api/perl/index.html" tabindex="-1">Perl</a>
</li>
</ul>
</li>
<li><a href="../../architecture/index.html">Architecture</a></li>
<li><a class="main-nav-link" href="https://github.com/dmlc/mxnet">Github</a></li>
<li id="dropdown-menu-position-anchor-version-mobile" class="dropdown-submenu" style="position: relative"><a href="#" tabindex="-1">Versions(master)</a><ul class="dropdown-menu"><li><a tabindex="-1" href=http://mxnet.incubator.apache.org/test/>v0.10.14</a></li><li><a tabindex="-1" href=http://mxnet.incubator.apache.org/test/versions/0.10/index.html>0.10</a></li><li><a tabindex="-1" href=http://mxnet.incubator.apache.org/test/versions/master/index.html>master</a></li></ul></li></ul>
</div>
<div class="plusIcon dropdown">
<a class="dropdown-toggle" data-toggle="dropdown" href="#" role="button"><span aria-hidden="true" class="glyphicon glyphicon-plus"></span></a>
<ul class="dropdown-menu dropdown-menu-right" id="plusMenu"></ul>
</div>
<div id="search-input-wrap">
<form action="../../search.html" autocomplete="off" class="" method="get" role="search">
<div class="form-group inner-addon left-addon">
<i class="glyphicon glyphicon-search"></i>
<input class="form-control" name="q" placeholder="Search" type="text"/>
</div>
<input name="check_keywords" type="hidden" value="yes">
<input name="area" type="hidden" value="default"/>
</input></form>
<div id="search-preview"></div>
</div>
<div id="searchIcon">
<span aria-hidden="true" class="glyphicon glyphicon-search"></span>
</div>
<!-- <div id="lang-select-wrap"> -->
<!-- <label id="lang-select-label"> -->
<!-- <\!-- <i class="fa fa-globe"></i> -\-> -->
<!-- <span></span> -->
<!-- </label> -->
<!-- <select id="lang-select"> -->
<!-- <option value="en">Eng</option> -->
<!-- <option value="zh">中文</option> -->
<!-- </select> -->
<!-- </div> -->
<!-- <a id="mobile-nav-toggle">
<span class="mobile-nav-toggle-bar"></span>
<span class="mobile-nav-toggle-bar"></span>
<span class="mobile-nav-toggle-bar"></span>
</a> -->
</div>
</div>
</div>
<div class="container">
<div class="row">
<div aria-label="main navigation" class="sphinxsidebar leftsidebar" role="navigation">
<div class="sphinxsidebarwrapper">
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../api/python/index.html">Python Documents</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../api/r/index.html">R Documents</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../api/julia/index.html">Julia Documents</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../api/c++/index.html">C++ Documents</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../api/scala/index.html">Scala Documents</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../api/perl/index.html">Perl Documents</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../how_to/index.html">HowTo Documents</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../architecture/index.html">System Documents</a></li>
<li class="toctree-l1"><a class="reference internal" href="../index.html">Tutorials</a></li>
</ul>
</div>
</div>
<div class="content">
<div class="section" id="callback-function">
<span id="callback-function"></span><h1>Callback Function<a class="headerlink" href="#callback-function" title="Permalink to this headline"></a></h1>
<p>This tutorial provides guidelines for using and writing callback functions,
which can very useful in model training.</p>
<div class="section" id="model-training-example">
<span id="model-training-example"></span><h2>Model Training Example<a class="headerlink" href="#model-training-example" title="Permalink to this headline"></a></h2>
<p>Let’s begin with a small example. We can build and train a model with the following code:</p>
<div class="highlight-r"><div class="highlight"><pre><span></span> <span class="kn">library</span><span class="p">(</span>mxnet<span class="p">)</span>
data<span class="p">(</span>BostonHousing<span class="p">,</span> package<span class="o">=</span><span class="s">"mlbench"</span><span class="p">)</span>
train.ind <span class="o">=</span> <span class="kp">seq</span><span class="p">(</span><span class="m">1</span><span class="p">,</span> <span class="m">506</span><span class="p">,</span> <span class="m">3</span><span class="p">)</span>
train.x <span class="o">=</span> <span class="kp">data.matrix</span><span class="p">(</span>BostonHousing<span class="p">[</span>train.ind<span class="p">,</span> <span class="m">-14</span><span class="p">])</span>
train.y <span class="o">=</span> BostonHousing<span class="p">[</span>train.ind<span class="p">,</span> <span class="m">14</span><span class="p">]</span>
test.x <span class="o">=</span> <span class="kp">data.matrix</span><span class="p">(</span>BostonHousing<span class="p">[</span><span class="o">-</span>train.ind<span class="p">,</span> <span class="m">-14</span><span class="p">])</span>
test.y <span class="o">=</span> BostonHousing<span class="p">[</span><span class="o">-</span>train.ind<span class="p">,</span> <span class="m">14</span><span class="p">]</span>
data <span class="o"><-</span> mx.symbol.Variable<span class="p">(</span><span class="s">"data"</span><span class="p">)</span>
fc1 <span class="o"><-</span> mx.symbol.FullyConnected<span class="p">(</span>data<span class="p">,</span> num_hidden<span class="o">=</span><span class="m">1</span><span class="p">)</span>
lro <span class="o"><-</span> mx.symbol.LinearRegressionOutput<span class="p">(</span>fc1<span class="p">)</span>
mx.set.seed<span class="p">(</span><span class="m">0</span><span class="p">)</span>
model <span class="o"><-</span> mx.model.FeedForward.create<span class="p">(</span>
lro<span class="p">,</span> X<span class="o">=</span>train.x<span class="p">,</span> y<span class="o">=</span>train.y<span class="p">,</span>
eval.data<span class="o">=</span><span class="kt">list</span><span class="p">(</span>data<span class="o">=</span>test.x<span class="p">,</span> label<span class="o">=</span>test.y<span class="p">),</span>
ctx<span class="o">=</span>mx.cpu<span class="p">(),</span> num.round<span class="o">=</span><span class="m">10</span><span class="p">,</span> array.batch.size<span class="o">=</span><span class="m">20</span><span class="p">,</span>
learning.rate<span class="o">=</span><span class="m">2e-6</span><span class="p">,</span> momentum<span class="o">=</span><span class="m">0.9</span><span class="p">,</span> eval.metric<span class="o">=</span>mx.metric.rmse<span class="p">)</span>
</pre></div>
</div>
<div class="highlight-python"><div class="highlight"><pre><span></span> <span class="c1">## Auto detect layout of input matrix, use row major..</span>
<span class="c1">## Start training with 1 devices</span>
<span class="c1">## [1] Train-rmse=16.063282524034</span>
<span class="c1">## [1] Validation-rmse=10.1766446093622</span>
<span class="c1">## [2] Train-rmse=12.2792375712573</span>
<span class="c1">## [2] Validation-rmse=12.4331776190813</span>
<span class="c1">## [3] Train-rmse=11.1984634005885</span>
<span class="c1">## [3] Validation-rmse=10.3303041888193</span>
<span class="c1">## [4] Train-rmse=10.2645236892904</span>
<span class="c1">## [4] Validation-rmse=8.42760407903415</span>
<span class="c1">## [5] Train-rmse=9.49711005504284</span>
<span class="c1">## [5] Validation-rmse=8.44557808483234</span>
<span class="c1">## [6] Train-rmse=9.07733734175182</span>
<span class="c1">## [6] Validation-rmse=8.33225500266177</span>
<span class="c1">## [7] Train-rmse=9.07884450847991</span>
<span class="c1">## [7] Validation-rmse=8.38827833418459</span>
<span class="c1">## [8] Train-rmse=9.10463850277417</span>
<span class="c1">## [8] Validation-rmse=8.37394452365264</span>
<span class="c1">## [9] Train-rmse=9.03977049028532</span>
<span class="c1">## [9] Validation-rmse=8.25927979725672</span>
<span class="c1">## [10] Train-rmse=8.96870685004475</span>
<span class="c1">## [10] Validation-rmse=8.19509291481822</span>
</pre></div>
</div>
<p>We also provide two optional parameters, <code class="docutils literal"><span class="pre">batch.end.callback</span></code> and <code class="docutils literal"><span class="pre">epoch.end.callback</span></code>, which can provide great flexibility in model training.</p>
</div>
<div class="section" id="how-to-use-callback-functions">
<span id="how-to-use-callback-functions"></span><h2>How to Use Callback Functions<a class="headerlink" href="#how-to-use-callback-functions" title="Permalink to this headline"></a></h2>
<p>This package provides two callback functions:</p>
<ul class="simple">
<li><code class="docutils literal"><span class="pre">mx.callback.save.checkpoint</span></code> saves a checkpoint to files during each period iteration.</li>
</ul>
<div class="highlight-r"><div class="highlight"><pre><span></span> model <span class="o"><-</span> mx.model.FeedForward.create<span class="p">(</span>
lro<span class="p">,</span> X<span class="o">=</span>train.x<span class="p">,</span> y<span class="o">=</span>train.y<span class="p">,</span>
eval.data<span class="o">=</span><span class="kt">list</span><span class="p">(</span>data<span class="o">=</span>test.x<span class="p">,</span> label<span class="o">=</span>test.y<span class="p">),</span>
ctx<span class="o">=</span>mx.cpu<span class="p">(),</span> num.round<span class="o">=</span><span class="m">10</span><span class="p">,</span> array.batch.size<span class="o">=</span><span class="m">20</span><span class="p">,</span>
learning.rate<span class="o">=</span><span class="m">2e-6</span><span class="p">,</span> momentum<span class="o">=</span><span class="m">0.9</span><span class="p">,</span> eval.metric<span class="o">=</span>mx.metric.rmse<span class="p">,</span>
epoch.end.callback <span class="o">=</span> mx.callback.save.checkpoint<span class="p">(</span><span class="s">"boston"</span><span class="p">))</span>
</pre></div>
</div>
<div class="highlight-python"><div class="highlight"><pre><span></span> <span class="c1">## Auto detect layout of input matrix, use row major..</span>
<span class="c1">## Start training with 1 devices</span>
<span class="c1">## [1] Train-rmse=19.1621424021617</span>
<span class="c1">## [1] Validation-rmse=20.721515592165</span>
<span class="c1">## Model checkpoint saved to boston-0001.params</span>
<span class="c1">## [2] Train-rmse=13.5127391952367</span>
<span class="c1">## [2] Validation-rmse=14.1822123675007</span>
<span class="c1">## Model checkpoint saved to boston-0002.params</span>
</pre></div>
</div>
<ul class="simple">
<li><code class="docutils literal"><span class="pre">mx.callback.log.train.metric</span></code> logs a training metric each period. You can use it either as a <code class="docutils literal"><span class="pre">batch.end.callback</span></code> or an
<code class="docutils literal"><span class="pre">epoch.end.callback</span></code>.</li>
</ul>
<div class="highlight-r"><div class="highlight"><pre><span></span> model <span class="o"><-</span> mx.model.FeedForward.create<span class="p">(</span>
lro<span class="p">,</span> X<span class="o">=</span>train.x<span class="p">,</span> y<span class="o">=</span>train.y<span class="p">,</span>
eval.data<span class="o">=</span><span class="kt">list</span><span class="p">(</span>data<span class="o">=</span>test.x<span class="p">,</span> label<span class="o">=</span>test.y<span class="p">),</span>
ctx<span class="o">=</span>mx.cpu<span class="p">(),</span> num.round<span class="o">=</span><span class="m">10</span><span class="p">,</span> array.batch.size<span class="o">=</span><span class="m">20</span><span class="p">,</span>
learning.rate<span class="o">=</span><span class="m">2e-6</span><span class="p">,</span> momentum<span class="o">=</span><span class="m">0.9</span><span class="p">,</span> eval.metric<span class="o">=</span>mx.metric.rmse<span class="p">,</span>
batch.end.callback <span class="o">=</span> mx.callback.log.train.metric<span class="p">(</span><span class="m">5</span><span class="p">))</span>
</pre></div>
</div>
<div class="highlight-python"><div class="highlight"><pre><span></span> <span class="c1">## Auto detect layout of input matrix, use row major..</span>
<span class="c1">## Start training with 1 devices</span>
<span class="c1">## Batch [5] Train-rmse=17.6514558545416</span>
<span class="c1">## [1] Train-rmse=15.2879610219001</span>
<span class="c1">## [1] Validation-rmse=12.3332062820921</span>
<span class="c1">## Batch [5] Train-rmse=11.939392828565</span>
<span class="c1">## [2] Train-rmse=11.4382242547217</span>
<span class="c1">## [2] Validation-rmse=9.91176550103181</span>
<span class="o">............</span>
</pre></div>
</div>
<p>You also can save the training and evaluation errors for later use by passing a reference class:</p>
<div class="highlight-r"><div class="highlight"><pre><span></span> logger <span class="o"><-</span> mx.metric.logger<span class="o">$</span>new<span class="p">()</span>
model <span class="o"><-</span> mx.model.FeedForward.create<span class="p">(</span>
lro<span class="p">,</span> X<span class="o">=</span>train.x<span class="p">,</span> y<span class="o">=</span>train.y<span class="p">,</span>
eval.data<span class="o">=</span><span class="kt">list</span><span class="p">(</span>data<span class="o">=</span>test.x<span class="p">,</span> label<span class="o">=</span>test.y<span class="p">),</span>
ctx<span class="o">=</span>mx.cpu<span class="p">(),</span> num.round<span class="o">=</span><span class="m">10</span><span class="p">,</span> array.batch.size<span class="o">=</span><span class="m">20</span><span class="p">,</span>
learning.rate<span class="o">=</span><span class="m">2e-6</span><span class="p">,</span> momentum<span class="o">=</span><span class="m">0.9</span><span class="p">,</span> eval.metric<span class="o">=</span>mx.metric.rmse<span class="p">,</span>
epoch.end.callback <span class="o">=</span> mx.callback.log.train.metric<span class="p">(</span><span class="m">5</span><span class="p">,</span> logger<span class="p">))</span>
</pre></div>
</div>
<div class="highlight-python"><div class="highlight"><pre><span></span> <span class="c1">## Auto detect layout of input matrix, use row major..</span>
<span class="c1">## Start training with 1 devices</span>
<span class="c1">## [1] Train-rmse=19.1083228733256</span>
<span class="c1">## [1] Validation-rmse=12.7150687428974</span>
<span class="c1">## [2] Train-rmse=15.7684378116157</span>
<span class="c1">## [2] Validation-rmse=14.8105319420491</span>
<span class="o">............</span>
</pre></div>
</div>
<div class="highlight-r"><div class="highlight"><pre><span></span> <span class="kp">head</span><span class="p">(</span>logger<span class="o">$</span>train<span class="p">)</span>
</pre></div>
</div>
<div class="highlight-python"><div class="highlight"><pre><span></span> <span class="c1">## [1] 19.108323 15.768438 13.531470 11.386050 9.555477 9.351324</span>
</pre></div>
</div>
<div class="highlight-r"><div class="highlight"><pre><span></span> <span class="kp">head</span><span class="p">(</span>logger<span class="o">$</span><span class="kp">eval</span><span class="p">)</span>
</pre></div>
</div>
<div class="highlight-python"><div class="highlight"><pre><span></span> <span class="c1">## [1] 12.715069 14.810532 15.840361 10.898733 9.349706 9.363087</span>
</pre></div>
</div>
</div>
<div class="section" id="how-to-write-your-own-callback-functions">
<span id="how-to-write-your-own-callback-functions"></span><h2>How to Write Your Own Callback Functions<a class="headerlink" href="#how-to-write-your-own-callback-functions" title="Permalink to this headline"></a></h2>
<p>You can find the source code for the two callback functions on <a class="reference external" href="https://github.com/dmlc/mxnet/blob/master/R-package/R/callback.R">GitHub</a> and use it as a template:</p>
<p>Basically, all callback functions follow the following structure:</p>
<div class="highlight-r"><div class="highlight"><pre><span></span> mx.callback.fun <span class="o"><-</span> <span class="kr">function</span><span class="p">()</span> <span class="p">{</span>
<span class="kr">function</span><span class="p">(</span>iteration<span class="p">,</span> nbatch<span class="p">,</span> env<span class="p">)</span> <span class="p">{</span>
<span class="p">}</span>
<span class="p">}</span>
</pre></div>
</div>
<p>The following <code class="docutils literal"><span class="pre">mx.callback.save.checkpoint</span></code> function is stateless. It gets the model from the environment and saves it:.</p>
<div class="highlight-r"><div class="highlight"><pre><span></span> mx.callback.save.checkpoint <span class="o"><-</span> <span class="kr">function</span><span class="p">(</span>prefix<span class="p">,</span> period<span class="o">=</span><span class="m">1</span><span class="p">)</span> <span class="p">{</span>
<span class="kr">function</span><span class="p">(</span>iteration<span class="p">,</span> nbatch<span class="p">,</span> env<span class="p">)</span> <span class="p">{</span>
<span class="kr">if</span> <span class="p">(</span>iteration <span class="o">%%</span> period <span class="o">==</span> <span class="m">0</span><span class="p">)</span> <span class="p">{</span>
mx.model.save<span class="p">(</span>env<span class="o">$</span>model<span class="p">,</span> prefix<span class="p">,</span> iteration<span class="p">)</span>
<span class="kp">cat</span><span class="p">(</span><span class="kp">sprintf</span><span class="p">(</span><span class="s">"Model checkpoint saved to %s-%04d.params\n"</span><span class="p">,</span> prefix<span class="p">,</span> iteration<span class="p">))</span>
<span class="p">}</span>
<span class="kr">return</span><span class="p">(</span><span class="kc">TRUE</span><span class="p">)</span>
<span class="p">}</span>
<span class="p">}</span>
</pre></div>
</div>
<p>The <code class="docutils literal"><span class="pre">mx.callback.log.train.metric</span></code> is a little more complex. It holds a reference class and updates it during the training
process:</p>
<div class="highlight-r"><div class="highlight"><pre><span></span> mx.callback.log.train.metric <span class="o"><-</span> <span class="kr">function</span><span class="p">(</span>period<span class="p">,</span> logger<span class="o">=</span><span class="kc">NULL</span><span class="p">)</span> <span class="p">{</span>
<span class="kr">function</span><span class="p">(</span>iteration<span class="p">,</span> nbatch<span class="p">,</span> env<span class="p">)</span> <span class="p">{</span>
<span class="kr">if</span> <span class="p">(</span>nbatch <span class="o">%%</span> period <span class="o">==</span> <span class="m">0</span> <span class="o">&amp;&amp;</span> <span class="o">!</span><span class="kp">is.null</span><span class="p">(</span>env<span class="o">$</span>metric<span class="p">))</span> <span class="p">{</span>
result <span class="o"><-</span> env<span class="o">$</span>metric<span class="o">$</span><span class="kp">get</span><span class="p">(</span>env<span class="o">$</span>train.metric<span class="p">)</span>
<span class="kr">if</span> <span class="p">(</span>nbatch <span class="o">!=</span> <span class="m">0</span><span class="p">)</span>
<span class="kp">cat</span><span class="p">(</span><span class="kp">paste0</span><span class="p">(</span><span class="s">"Batch ["</span><span class="p">,</span> nbatch<span class="p">,</span> <span class="s">"] Train-"</span><span class="p">,</span> result<span class="o">$</span>name<span class="p">,</span> <span class="s">"="</span><span class="p">,</span> result<span class="o">$</span>value<span class="p">,</span> <span class="s">"\n"</span><span class="p">))</span>
<span class="kr">if</span> <span class="p">(</span><span class="o">!</span><span class="kp">is.null</span><span class="p">(</span>logger<span class="p">))</span> <span class="p">{</span>
<span class="kr">if</span> <span class="p">(</span><span class="kp">class</span><span class="p">(</span>logger<span class="p">)</span> <span class="o">!=</span> <span class="s">"mx.metric.logger"</span><span class="p">)</span> <span class="p">{</span>
<span class="kp">stop</span><span class="p">(</span><span class="s">"Invalid mx.metric.logger."</span><span class="p">)</span>
<span class="p">}</span>
logger<span class="o">$</span>train <span class="o"><-</span> <span class="kt">c</span><span class="p">(</span>logger<span class="o">$</span>train<span class="p">,</span> result<span class="o">$</span>value<span class="p">)</span>
<span class="kr">if</span> <span class="p">(</span><span class="o">!</span><span class="kp">is.null</span><span class="p">(</span>env<span class="o">$</span>eval.metric<span class="p">))</span> <span class="p">{</span>
result <span class="o"><-</span> env<span class="o">$</span>metric<span class="o">$</span><span class="kp">get</span><span class="p">(</span>env<span class="o">$</span>eval.metric<span class="p">)</span>
<span class="kr">if</span> <span class="p">(</span>nbatch <span class="o">!=</span> <span class="m">0</span><span class="p">)</span>
<span class="kp">cat</span><span class="p">(</span><span class="kp">paste0</span><span class="p">(</span><span class="s">"Batch ["</span><span class="p">,</span> nbatch<span class="p">,</span> <span class="s">"] Validation-"</span><span class="p">,</span> result<span class="o">$</span>name<span class="p">,</span> <span class="s">"="</span><span class="p">,</span> result<span class="o">$</span>value<span class="p">,</span> <span class="s">"\n"</span><span class="p">))</span>
logger<span class="o">$</span>eval <span class="o"><-</span> <span class="kt">c</span><span class="p">(</span>logger<span class="o">$</span><span class="kp">eval</span><span class="p">,</span> result<span class="o">$</span>value<span class="p">)</span>
<span class="p">}</span>
<span class="p">}</span>
<span class="p">}</span>
<span class="kr">return</span><span class="p">(</span><span class="kc">TRUE</span><span class="p">)</span>
<span class="p">}</span>
<span class="p">}</span>
</pre></div>
</div>
<p>Now you might be curious why both callback functions <code class="docutils literal"><span class="pre">return(TRUE)</span></code>.</p>
<p>Can we <code class="docutils literal"><span class="pre">return(FALSE)</span></code>?</p>
<p>Yes! You can stop the training early with <code class="docutils literal"><span class="pre">return(FALSE)</span></code>. See the following examples.</p>
<div class="highlight-r"><div class="highlight"><pre><span></span> mx.callback.early.stop <span class="o"><-</span> <span class="kr">function</span><span class="p">(</span>eval.metric<span class="p">)</span> <span class="p">{</span>
<span class="kr">function</span><span class="p">(</span>iteration<span class="p">,</span> nbatch<span class="p">,</span> env<span class="p">)</span> <span class="p">{</span>
<span class="kr">if</span> <span class="p">(</span><span class="o">!</span><span class="kp">is.null</span><span class="p">(</span>env<span class="o">$</span>metric<span class="p">))</span> <span class="p">{</span>
<span class="kr">if</span> <span class="p">(</span><span class="o">!</span><span class="kp">is.null</span><span class="p">(</span>eval.metric<span class="p">))</span> <span class="p">{</span>
result <span class="o"><-</span> env<span class="o">$</span>metric<span class="o">$</span><span class="kp">get</span><span class="p">(</span>env<span class="o">$</span>eval.metric<span class="p">)</span>
<span class="kr">if</span> <span class="p">(</span>result<span class="o">$</span>value <span class="o"><</span> eval.metric<span class="p">)</span> <span class="p">{</span>
<span class="kr">return</span><span class="p">(</span><span class="kc">FALSE</span><span class="p">)</span>
<span class="p">}</span>
<span class="p">}</span>
<span class="p">}</span>
<span class="kr">return</span><span class="p">(</span><span class="kc">TRUE</span><span class="p">)</span>
<span class="p">}</span>
<span class="p">}</span>
model <span class="o"><-</span> mx.model.FeedForward.create<span class="p">(</span>
lro<span class="p">,</span> X<span class="o">=</span>train.x<span class="p">,</span> y<span class="o">=</span>train.y<span class="p">,</span>
eval.data<span class="o">=</span><span class="kt">list</span><span class="p">(</span>data<span class="o">=</span>test.x<span class="p">,</span> label<span class="o">=</span>test.y<span class="p">),</span>
ctx<span class="o">=</span>mx.cpu<span class="p">(),</span> num.round<span class="o">=</span><span class="m">10</span><span class="p">,</span> array.batch.size<span class="o">=</span><span class="m">20</span><span class="p">,</span>
learning.rate<span class="o">=</span><span class="m">2e-6</span><span class="p">,</span> momentum<span class="o">=</span><span class="m">0.9</span><span class="p">,</span> eval.metric<span class="o">=</span>mx.metric.rmse<span class="p">,</span>
epoch.end.callback <span class="o">=</span> mx.callback.early.stop<span class="p">(</span><span class="m">10</span><span class="p">))</span>
</pre></div>
</div>
<div class="highlight-python"><div class="highlight"><pre><span></span> <span class="c1">## Auto detect layout of input matrix, use row major..</span>
<span class="c1">## Start training with 1 devices</span>
<span class="c1">## [1] Train-rmse=18.5897984387033</span>
<span class="c1">## [1] Validation-rmse=13.5555213820571</span>
<span class="c1">## [2] Train-rmse=12.5867564040256</span>
<span class="c1">## [2] Validation-rmse=9.76304967080928</span>
</pre></div>
</div>
<p>When the validation metric dips below the threshold we set, the training process stops.</p>
</div>
<div class="section" id="next-steps">
<span id="next-steps"></span><h2>Next Steps<a class="headerlink" href="#next-steps" title="Permalink to this headline"></a></h2>
<div class="toctree-wrapper compound">
<ul>
<li class="toctree-l1"><a class="reference external" href="http://mxnet.io/tutorials/r/fiveMinutesNeuralNetwork.html">Neural Networks with MXNet in Five Minutes</a></li>
<li class="toctree-l1"><a class="reference external" href="http://mxnet.io/tutorials/r/classifyRealImageWithPretrainedModel.html">Classify Real-World Images with a Pretrained Model</a></li>
<li class="toctree-l1"><a class="reference external" href="http://mxnet.io/tutorials/r/mnistCompetition.html">Handwritten Digits Classification Competition</a></li>
<li class="toctree-l1"><a class="reference external" href="http://mxnet.io/tutorials/r/charRnnModel.html">Character Language Model Using RNN</a></li>
</ul>
</div>
</div>
</div>
<div class="container">
<div class="footer">
<p> © 2015-2017 DMLC. All rights reserved. </p>
</div>
</div>
</div>
<div aria-label="main navigation" class="sphinxsidebar rightsidebar" role="navigation">
<div class="sphinxsidebarwrapper">
<h3><a href="../../index.html">Table Of Contents</a></h3>
<ul>
<li><a class="reference internal" href="#">Callback Function</a><ul>
<li><a class="reference internal" href="#model-training-example">Model Training Example</a></li>
<li><a class="reference internal" href="#how-to-use-callback-functions">How to Use Callback Functions</a></li>
<li><a class="reference internal" href="#how-to-write-your-own-callback-functions">How to Write Your Own Callback Functions</a></li>
<li><a class="reference internal" href="#next-steps">Next Steps</a></li>
</ul>
</li>
</ul>
</div>
</div>
</div> <!-- pagename != index -->
<script crossorigin="anonymous" integrity="sha384-0mSbJDEHialfmuBBQP6A4Qrprq5OVfW37PRR3j5ELqxss1yVqOtnepnHVP9aJ7xS" src="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.6/js/bootstrap.min.js"></script>
<script src="../../_static/js/sidebar.js" type="text/javascript"></script>
<script src="../../_static/js/search.js" type="text/javascript"></script>
<script src="../../_static/js/navbar.js" type="text/javascript"></script>
<script src="../../_static/js/clipboard.min.js" type="text/javascript"></script>
<script src="../../_static/js/copycode.js" type="text/javascript"></script>
<script type="text/javascript">
$('body').ready(function () {
$('body').css('visibility', 'visible');
});
</script>
</div></body>
</html>