mirror of
https://github.com/SheffieldML/GPy.git
synced 2026-05-05 09:42:39 +02:00
456 lines
No EOL
38 KiB
HTML
456 lines
No EOL
38 KiB
HTML
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
|
|
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
|
|
|
|
|
|
<html xmlns="http://www.w3.org/1999/xhtml">
|
|
<head>
|
|
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
|
|
|
|
<title>GPy.inference.optimization package — GPy documentation</title>
|
|
|
|
<link rel="stylesheet" href="_static//default.css" type="text/css" />
|
|
<link rel="stylesheet" href="_static/pygments.css" type="text/css" />
|
|
|
|
<script type="text/javascript">
|
|
var DOCUMENTATION_OPTIONS = {
|
|
URL_ROOT: './',
|
|
VERSION: '',
|
|
COLLAPSE_INDEX: false,
|
|
FILE_SUFFIX: '.html',
|
|
HAS_SOURCE: true
|
|
};
|
|
</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>
|
|
<link rel="top" title="GPy documentation" href="index.html" />
|
|
<link rel="up" title="GPy.inference package" href="GPy.inference.html" />
|
|
<link rel="next" title="GPy.kern package" href="GPy.kern.html" />
|
|
<link rel="prev" title="GPy.inference.mcmc package" href="GPy.inference.mcmc.html" />
|
|
</head>
|
|
<body role="document">
|
|
<div class="related" role="navigation" aria-label="related navigation">
|
|
<h3>Navigation</h3>
|
|
<ul>
|
|
<li class="right" style="margin-right: 10px">
|
|
<a href="genindex.html" title="General Index"
|
|
accesskey="I">index</a></li>
|
|
<li class="right" >
|
|
<a href="py-modindex.html" title="Python Module Index"
|
|
>modules</a> |</li>
|
|
<li class="right" >
|
|
<a href="GPy.kern.html" title="GPy.kern package"
|
|
accesskey="N">next</a> |</li>
|
|
<li class="right" >
|
|
<a href="GPy.inference.mcmc.html" title="GPy.inference.mcmc package"
|
|
accesskey="P">previous</a> |</li>
|
|
<li class="nav-item nav-item-0"><a href="index.html">GPy documentation</a> »</li>
|
|
<li class="nav-item nav-item-1"><a href="GPy.html" >GPy package</a> »</li>
|
|
<li class="nav-item nav-item-2"><a href="GPy.inference.html" accesskey="U">GPy.inference package</a> »</li>
|
|
</ul>
|
|
</div>
|
|
|
|
<div class="document">
|
|
<div class="documentwrapper">
|
|
<div class="bodywrapper">
|
|
<div class="body" role="main">
|
|
|
|
<div class="section" id="gpy-inference-optimization-package">
|
|
<h1>GPy.inference.optimization package<a class="headerlink" href="#gpy-inference-optimization-package" title="Permalink to this headline">¶</a></h1>
|
|
<div class="section" id="submodules">
|
|
<h2>Submodules<a class="headerlink" href="#submodules" title="Permalink to this headline">¶</a></h2>
|
|
</div>
|
|
<div class="section" id="module-GPy.inference.optimization.conjugate_gradient_descent">
|
|
<span id="gpy-inference-optimization-conjugate-gradient-descent-module"></span><h2>GPy.inference.optimization.conjugate_gradient_descent module<a class="headerlink" href="#module-GPy.inference.optimization.conjugate_gradient_descent" title="Permalink to this headline">¶</a></h2>
|
|
<dl class="class">
|
|
<dt id="GPy.inference.optimization.conjugate_gradient_descent.Async_Optimize">
|
|
<em class="property">class </em><code class="descclassname">GPy.inference.optimization.conjugate_gradient_descent.</code><code class="descname">Async_Optimize</code><a class="reference internal" href="_modules/GPy/inference/optimization/conjugate_gradient_descent.html#Async_Optimize"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.conjugate_gradient_descent.Async_Optimize" title="Permalink to this definition">¶</a></dt>
|
|
<dd><p>Bases: <code class="xref py py-class docutils literal"><span class="pre">object</span></code></p>
|
|
<dl class="attribute">
|
|
<dt id="GPy.inference.optimization.conjugate_gradient_descent.Async_Optimize.SENTINEL">
|
|
<code class="descname">SENTINEL</code><em class="property"> = 'SENTINEL'</em><a class="headerlink" href="#GPy.inference.optimization.conjugate_gradient_descent.Async_Optimize.SENTINEL" title="Permalink to this definition">¶</a></dt>
|
|
<dd></dd></dl>
|
|
|
|
<dl class="method">
|
|
<dt id="GPy.inference.optimization.conjugate_gradient_descent.Async_Optimize.async_callback_collect">
|
|
<code class="descname">async_callback_collect</code><span class="sig-paren">(</span><em>q</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/conjugate_gradient_descent.html#Async_Optimize.async_callback_collect"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.conjugate_gradient_descent.Async_Optimize.async_callback_collect" title="Permalink to this definition">¶</a></dt>
|
|
<dd></dd></dl>
|
|
|
|
<dl class="method">
|
|
<dt id="GPy.inference.optimization.conjugate_gradient_descent.Async_Optimize.callback">
|
|
<code class="descname">callback</code><span class="sig-paren">(</span><em>*x</em><span class="sig-paren">)</span><a class="headerlink" href="#GPy.inference.optimization.conjugate_gradient_descent.Async_Optimize.callback" title="Permalink to this definition">¶</a></dt>
|
|
<dd></dd></dl>
|
|
|
|
<dl class="method">
|
|
<dt id="GPy.inference.optimization.conjugate_gradient_descent.Async_Optimize.opt">
|
|
<code class="descname">opt</code><span class="sig-paren">(</span><em>f</em>, <em>df</em>, <em>x0</em>, <em>callback=None</em>, <em>update_rule=<class GPy.inference.optimization.gradient_descent_update_rules.FletcherReeves></em>, <em>messages=0</em>, <em>maxiter=5000.0</em>, <em>max_f_eval=15000.0</em>, <em>gtol=1e-06</em>, <em>report_every=10</em>, <em>*args</em>, <em>**kwargs</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/conjugate_gradient_descent.html#Async_Optimize.opt"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.conjugate_gradient_descent.Async_Optimize.opt" title="Permalink to this definition">¶</a></dt>
|
|
<dd></dd></dl>
|
|
|
|
<dl class="method">
|
|
<dt id="GPy.inference.optimization.conjugate_gradient_descent.Async_Optimize.opt_async">
|
|
<code class="descname">opt_async</code><span class="sig-paren">(</span><em>f</em>, <em>df</em>, <em>x0</em>, <em>callback</em>, <em>update_rule=<class GPy.inference.optimization.gradient_descent_update_rules.PolakRibiere></em>, <em>messages=0</em>, <em>maxiter=5000.0</em>, <em>max_f_eval=15000.0</em>, <em>gtol=1e-06</em>, <em>report_every=10</em>, <em>*args</em>, <em>**kwargs</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/conjugate_gradient_descent.html#Async_Optimize.opt_async"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.conjugate_gradient_descent.Async_Optimize.opt_async" title="Permalink to this definition">¶</a></dt>
|
|
<dd></dd></dl>
|
|
|
|
<dl class="attribute">
|
|
<dt id="GPy.inference.optimization.conjugate_gradient_descent.Async_Optimize.runsignal">
|
|
<code class="descname">runsignal</code><em class="property"> = <multiprocessing.synchronize.Event object></em><a class="headerlink" href="#GPy.inference.optimization.conjugate_gradient_descent.Async_Optimize.runsignal" title="Permalink to this definition">¶</a></dt>
|
|
<dd></dd></dl>
|
|
|
|
</dd></dl>
|
|
|
|
<dl class="class">
|
|
<dt id="GPy.inference.optimization.conjugate_gradient_descent.CGD">
|
|
<em class="property">class </em><code class="descclassname">GPy.inference.optimization.conjugate_gradient_descent.</code><code class="descname">CGD</code><a class="reference internal" href="_modules/GPy/inference/optimization/conjugate_gradient_descent.html#CGD"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.conjugate_gradient_descent.CGD" title="Permalink to this definition">¶</a></dt>
|
|
<dd><p>Bases: <a class="reference internal" href="#GPy.inference.optimization.conjugate_gradient_descent.Async_Optimize" title="GPy.inference.optimization.conjugate_gradient_descent.Async_Optimize"><code class="xref py py-class docutils literal"><span class="pre">GPy.inference.optimization.conjugate_gradient_descent.Async_Optimize</span></code></a></p>
|
|
<p>Conjugate gradient descent algorithm to minimize
|
|
function f with gradients df, starting at x0
|
|
with update rule update_rule</p>
|
|
<p>if df returns tuple (grad, natgrad) it will optimize according
|
|
to natural gradient rules</p>
|
|
<dl class="method">
|
|
<dt id="GPy.inference.optimization.conjugate_gradient_descent.CGD.opt">
|
|
<code class="descname">opt</code><span class="sig-paren">(</span><em>*a</em>, <em>**kw</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/conjugate_gradient_descent.html#CGD.opt"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.conjugate_gradient_descent.CGD.opt" title="Permalink to this definition">¶</a></dt>
|
|
<dd><dl class="docutils">
|
|
<dt>opt(self, f, df, x0, callback=None, update_rule=FletcherReeves,</dt>
|
|
<dd>messages=0, maxiter=5e3, max_f_eval=15e3, gtol=1e-6,
|
|
report_every=10, *args, **kwargs)</dd>
|
|
</dl>
|
|
<p>Minimize f, calling callback every <cite>report_every</cite> iterations with following syntax:</p>
|
|
<blockquote>
|
|
<div>callback(xi, fi, gi, iteration, function_calls, gradient_calls, status_message)</div></blockquote>
|
|
<p>if df returns tuple (grad, natgrad) it will optimize according
|
|
to natural gradient rules</p>
|
|
<p>f, and df will be called with</p>
|
|
<blockquote>
|
|
<div>f(xi, *args, **kwargs)
|
|
df(xi, *args, **kwargs)</div></blockquote>
|
|
<p><strong>returns</strong></p>
|
|
<blockquote>
|
|
<div>x_opt, f_opt, g_opt, iteration, function_calls, gradient_calls, status_message</div></blockquote>
|
|
<p>at end of optimization</p>
|
|
</dd></dl>
|
|
|
|
<dl class="method">
|
|
<dt id="GPy.inference.optimization.conjugate_gradient_descent.CGD.opt_async">
|
|
<code class="descname">opt_async</code><span class="sig-paren">(</span><em>*a</em>, <em>**kw</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/conjugate_gradient_descent.html#CGD.opt_async"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.conjugate_gradient_descent.CGD.opt_async" title="Permalink to this definition">¶</a></dt>
|
|
<dd><dl class="docutils">
|
|
<dt>opt_async(self, f, df, x0, callback, update_rule=FletcherReeves,</dt>
|
|
<dd>messages=0, maxiter=5e3, max_f_eval=15e3, gtol=1e-6,
|
|
report_every=10, *args, **kwargs)</dd>
|
|
</dl>
|
|
<p>callback gets called every <cite>report_every</cite> iterations</p>
|
|
<blockquote>
|
|
<div>callback(xi, fi, gi, iteration, function_calls, gradient_calls, status_message)</div></blockquote>
|
|
<p>if df returns tuple (grad, natgrad) it will optimize according
|
|
to natural gradient rules</p>
|
|
<p>f, and df will be called with</p>
|
|
<blockquote>
|
|
<div>f(xi, *args, **kwargs)
|
|
df(xi, *args, **kwargs)</div></blockquote>
|
|
<p><strong>Returns:</strong></p>
|
|
<blockquote>
|
|
<div>Started <cite>Process</cite> object, optimizing asynchronously</div></blockquote>
|
|
<p><strong>Calls:</strong></p>
|
|
<blockquote>
|
|
<div>callback(x_opt, f_opt, g_opt, iteration, function_calls, gradient_calls, status_message)</div></blockquote>
|
|
<p>at end of optimization!</p>
|
|
</dd></dl>
|
|
|
|
<dl class="attribute">
|
|
<dt id="GPy.inference.optimization.conjugate_gradient_descent.CGD.opt_name">
|
|
<code class="descname">opt_name</code><em class="property"> = 'Conjugate Gradient Descent'</em><a class="headerlink" href="#GPy.inference.optimization.conjugate_gradient_descent.CGD.opt_name" title="Permalink to this definition">¶</a></dt>
|
|
<dd></dd></dl>
|
|
|
|
</dd></dl>
|
|
|
|
</div>
|
|
<div class="section" id="module-GPy.inference.optimization.gradient_descent_update_rules">
|
|
<span id="gpy-inference-optimization-gradient-descent-update-rules-module"></span><h2>GPy.inference.optimization.gradient_descent_update_rules module<a class="headerlink" href="#module-GPy.inference.optimization.gradient_descent_update_rules" title="Permalink to this headline">¶</a></h2>
|
|
<dl class="class">
|
|
<dt id="GPy.inference.optimization.gradient_descent_update_rules.FletcherReeves">
|
|
<em class="property">class </em><code class="descclassname">GPy.inference.optimization.gradient_descent_update_rules.</code><code class="descname">FletcherReeves</code><span class="sig-paren">(</span><em>initgrad</em>, <em>initgradnat=None</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/gradient_descent_update_rules.html#FletcherReeves"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.gradient_descent_update_rules.FletcherReeves" title="Permalink to this definition">¶</a></dt>
|
|
<dd><p>Bases: <a class="reference internal" href="#GPy.inference.optimization.gradient_descent_update_rules.GDUpdateRule" title="GPy.inference.optimization.gradient_descent_update_rules.GDUpdateRule"><code class="xref py py-class docutils literal"><span class="pre">GPy.inference.optimization.gradient_descent_update_rules.GDUpdateRule</span></code></a></p>
|
|
<p>Fletcher Reeves update rule for gamma</p>
|
|
</dd></dl>
|
|
|
|
<dl class="class">
|
|
<dt id="GPy.inference.optimization.gradient_descent_update_rules.GDUpdateRule">
|
|
<em class="property">class </em><code class="descclassname">GPy.inference.optimization.gradient_descent_update_rules.</code><code class="descname">GDUpdateRule</code><span class="sig-paren">(</span><em>initgrad</em>, <em>initgradnat=None</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/gradient_descent_update_rules.html#GDUpdateRule"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.gradient_descent_update_rules.GDUpdateRule" title="Permalink to this definition">¶</a></dt>
|
|
<dd></dd></dl>
|
|
|
|
<dl class="class">
|
|
<dt id="GPy.inference.optimization.gradient_descent_update_rules.PolakRibiere">
|
|
<em class="property">class </em><code class="descclassname">GPy.inference.optimization.gradient_descent_update_rules.</code><code class="descname">PolakRibiere</code><span class="sig-paren">(</span><em>initgrad</em>, <em>initgradnat=None</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/gradient_descent_update_rules.html#PolakRibiere"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.gradient_descent_update_rules.PolakRibiere" title="Permalink to this definition">¶</a></dt>
|
|
<dd><p>Bases: <a class="reference internal" href="#GPy.inference.optimization.gradient_descent_update_rules.GDUpdateRule" title="GPy.inference.optimization.gradient_descent_update_rules.GDUpdateRule"><code class="xref py py-class docutils literal"><span class="pre">GPy.inference.optimization.gradient_descent_update_rules.GDUpdateRule</span></code></a></p>
|
|
<p>Fletcher Reeves update rule for gamma</p>
|
|
</dd></dl>
|
|
|
|
</div>
|
|
<div class="section" id="module-GPy.inference.optimization.optimization">
|
|
<span id="gpy-inference-optimization-optimization-module"></span><h2>GPy.inference.optimization.optimization module<a class="headerlink" href="#module-GPy.inference.optimization.optimization" title="Permalink to this headline">¶</a></h2>
|
|
<dl class="class">
|
|
<dt id="GPy.inference.optimization.optimization.Optimizer">
|
|
<em class="property">class </em><code class="descclassname">GPy.inference.optimization.optimization.</code><code class="descname">Optimizer</code><span class="sig-paren">(</span><em>x_init</em>, <em>messages=False</em>, <em>model=None</em>, <em>max_f_eval=10000.0</em>, <em>max_iters=1000.0</em>, <em>ftol=None</em>, <em>gtol=None</em>, <em>xtol=None</em>, <em>bfgs_factor=None</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/optimization.html#Optimizer"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.optimization.Optimizer" title="Permalink to this definition">¶</a></dt>
|
|
<dd><p>Superclass for all the optimizers.</p>
|
|
<table class="docutils field-list" frame="void" rules="none">
|
|
<col class="field-name" />
|
|
<col class="field-body" />
|
|
<tbody valign="top">
|
|
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
|
|
<li><strong>x_init</strong> – initial set of parameters</li>
|
|
<li><strong>f_fp</strong> – function that returns the function AND the gradients at the same time</li>
|
|
<li><strong>f</strong> – function to optimize</li>
|
|
<li><strong>fp</strong> – gradients</li>
|
|
<li><strong>messages</strong> (<em>(True | False)</em>) – print messages from the optimizer?</li>
|
|
<li><strong>max_f_eval</strong> – maximum number of function evaluations</li>
|
|
</ul>
|
|
</td>
|
|
</tr>
|
|
<tr class="field-even field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">optimizer object.</p>
|
|
</td>
|
|
</tr>
|
|
</tbody>
|
|
</table>
|
|
<dl class="method">
|
|
<dt id="GPy.inference.optimization.optimization.Optimizer.opt">
|
|
<code class="descname">opt</code><span class="sig-paren">(</span><em>f_fp=None</em>, <em>f=None</em>, <em>fp=None</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/optimization.html#Optimizer.opt"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.optimization.Optimizer.opt" title="Permalink to this definition">¶</a></dt>
|
|
<dd></dd></dl>
|
|
|
|
<dl class="method">
|
|
<dt id="GPy.inference.optimization.optimization.Optimizer.plot">
|
|
<code class="descname">plot</code><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/optimization.html#Optimizer.plot"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.optimization.Optimizer.plot" title="Permalink to this definition">¶</a></dt>
|
|
<dd><p>See GPy.plotting.matplot_dep.inference_plots</p>
|
|
</dd></dl>
|
|
|
|
<dl class="method">
|
|
<dt id="GPy.inference.optimization.optimization.Optimizer.run">
|
|
<code class="descname">run</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/optimization.html#Optimizer.run"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.optimization.Optimizer.run" title="Permalink to this definition">¶</a></dt>
|
|
<dd></dd></dl>
|
|
|
|
</dd></dl>
|
|
|
|
<dl class="function">
|
|
<dt id="GPy.inference.optimization.optimization.get_optimizer">
|
|
<code class="descclassname">GPy.inference.optimization.optimization.</code><code class="descname">get_optimizer</code><span class="sig-paren">(</span><em>f_min</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/optimization.html#get_optimizer"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.optimization.get_optimizer" title="Permalink to this definition">¶</a></dt>
|
|
<dd></dd></dl>
|
|
|
|
<dl class="class">
|
|
<dt id="GPy.inference.optimization.optimization.opt_SCG">
|
|
<em class="property">class </em><code class="descclassname">GPy.inference.optimization.optimization.</code><code class="descname">opt_SCG</code><span class="sig-paren">(</span><em>*args</em>, <em>**kwargs</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/optimization.html#opt_SCG"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.optimization.opt_SCG" title="Permalink to this definition">¶</a></dt>
|
|
<dd><p>Bases: <a class="reference internal" href="#GPy.inference.optimization.optimization.Optimizer" title="GPy.inference.optimization.optimization.Optimizer"><code class="xref py py-class docutils literal"><span class="pre">GPy.inference.optimization.optimization.Optimizer</span></code></a></p>
|
|
<dl class="method">
|
|
<dt id="GPy.inference.optimization.optimization.opt_SCG.opt">
|
|
<code class="descname">opt</code><span class="sig-paren">(</span><em>f_fp=None</em>, <em>f=None</em>, <em>fp=None</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/optimization.html#opt_SCG.opt"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.optimization.opt_SCG.opt" title="Permalink to this definition">¶</a></dt>
|
|
<dd></dd></dl>
|
|
|
|
</dd></dl>
|
|
|
|
<dl class="class">
|
|
<dt id="GPy.inference.optimization.optimization.opt_lbfgsb">
|
|
<em class="property">class </em><code class="descclassname">GPy.inference.optimization.optimization.</code><code class="descname">opt_lbfgsb</code><span class="sig-paren">(</span><em>*args</em>, <em>**kwargs</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/optimization.html#opt_lbfgsb"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.optimization.opt_lbfgsb" title="Permalink to this definition">¶</a></dt>
|
|
<dd><p>Bases: <a class="reference internal" href="#GPy.inference.optimization.optimization.Optimizer" title="GPy.inference.optimization.optimization.Optimizer"><code class="xref py py-class docutils literal"><span class="pre">GPy.inference.optimization.optimization.Optimizer</span></code></a></p>
|
|
<dl class="method">
|
|
<dt id="GPy.inference.optimization.optimization.opt_lbfgsb.opt">
|
|
<code class="descname">opt</code><span class="sig-paren">(</span><em>f_fp=None</em>, <em>f=None</em>, <em>fp=None</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/optimization.html#opt_lbfgsb.opt"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.optimization.opt_lbfgsb.opt" title="Permalink to this definition">¶</a></dt>
|
|
<dd><p>Run the optimizer</p>
|
|
</dd></dl>
|
|
|
|
</dd></dl>
|
|
|
|
<dl class="class">
|
|
<dt id="GPy.inference.optimization.optimization.opt_rasm">
|
|
<em class="property">class </em><code class="descclassname">GPy.inference.optimization.optimization.</code><code class="descname">opt_rasm</code><span class="sig-paren">(</span><em>*args</em>, <em>**kwargs</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/optimization.html#opt_rasm"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.optimization.opt_rasm" title="Permalink to this definition">¶</a></dt>
|
|
<dd><p>Bases: <a class="reference internal" href="#GPy.inference.optimization.optimization.Optimizer" title="GPy.inference.optimization.optimization.Optimizer"><code class="xref py py-class docutils literal"><span class="pre">GPy.inference.optimization.optimization.Optimizer</span></code></a></p>
|
|
<dl class="method">
|
|
<dt id="GPy.inference.optimization.optimization.opt_rasm.opt">
|
|
<code class="descname">opt</code><span class="sig-paren">(</span><em>f_fp=None</em>, <em>f=None</em>, <em>fp=None</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/optimization.html#opt_rasm.opt"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.optimization.opt_rasm.opt" title="Permalink to this definition">¶</a></dt>
|
|
<dd><p>Run Rasmussen’s Conjugate Gradient optimizer</p>
|
|
</dd></dl>
|
|
|
|
</dd></dl>
|
|
|
|
<dl class="class">
|
|
<dt id="GPy.inference.optimization.optimization.opt_simplex">
|
|
<em class="property">class </em><code class="descclassname">GPy.inference.optimization.optimization.</code><code class="descname">opt_simplex</code><span class="sig-paren">(</span><em>*args</em>, <em>**kwargs</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/optimization.html#opt_simplex"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.optimization.opt_simplex" title="Permalink to this definition">¶</a></dt>
|
|
<dd><p>Bases: <a class="reference internal" href="#GPy.inference.optimization.optimization.Optimizer" title="GPy.inference.optimization.optimization.Optimizer"><code class="xref py py-class docutils literal"><span class="pre">GPy.inference.optimization.optimization.Optimizer</span></code></a></p>
|
|
<dl class="method">
|
|
<dt id="GPy.inference.optimization.optimization.opt_simplex.opt">
|
|
<code class="descname">opt</code><span class="sig-paren">(</span><em>f_fp=None</em>, <em>f=None</em>, <em>fp=None</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/optimization.html#opt_simplex.opt"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.optimization.opt_simplex.opt" title="Permalink to this definition">¶</a></dt>
|
|
<dd><p>The simplex optimizer does not require gradients.</p>
|
|
</dd></dl>
|
|
|
|
</dd></dl>
|
|
|
|
<dl class="class">
|
|
<dt id="GPy.inference.optimization.optimization.opt_tnc">
|
|
<em class="property">class </em><code class="descclassname">GPy.inference.optimization.optimization.</code><code class="descname">opt_tnc</code><span class="sig-paren">(</span><em>*args</em>, <em>**kwargs</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/optimization.html#opt_tnc"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.optimization.opt_tnc" title="Permalink to this definition">¶</a></dt>
|
|
<dd><p>Bases: <a class="reference internal" href="#GPy.inference.optimization.optimization.Optimizer" title="GPy.inference.optimization.optimization.Optimizer"><code class="xref py py-class docutils literal"><span class="pre">GPy.inference.optimization.optimization.Optimizer</span></code></a></p>
|
|
<dl class="method">
|
|
<dt id="GPy.inference.optimization.optimization.opt_tnc.opt">
|
|
<code class="descname">opt</code><span class="sig-paren">(</span><em>f_fp=None</em>, <em>f=None</em>, <em>fp=None</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/optimization.html#opt_tnc.opt"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.optimization.opt_tnc.opt" title="Permalink to this definition">¶</a></dt>
|
|
<dd><p>Run the TNC optimizer</p>
|
|
</dd></dl>
|
|
|
|
</dd></dl>
|
|
|
|
</div>
|
|
<div class="section" id="module-GPy.inference.optimization.scg">
|
|
<span id="gpy-inference-optimization-scg-module"></span><h2>GPy.inference.optimization.scg module<a class="headerlink" href="#module-GPy.inference.optimization.scg" title="Permalink to this headline">¶</a></h2>
|
|
<dl class="function">
|
|
<dt id="GPy.inference.optimization.scg.SCG">
|
|
<code class="descclassname">GPy.inference.optimization.scg.</code><code class="descname">SCG</code><span class="sig-paren">(</span><em>f</em>, <em>gradf</em>, <em>x</em>, <em>optargs=()</em>, <em>maxiters=500</em>, <em>max_f_eval=inf</em>, <em>display=True</em>, <em>xtol=None</em>, <em>ftol=None</em>, <em>gtol=None</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/scg.html#SCG"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.scg.SCG" title="Permalink to this definition">¶</a></dt>
|
|
<dd><p>Optimisation through Scaled Conjugate Gradients (SCG)</p>
|
|
<p>f: the objective function
|
|
gradf : the gradient function (should return a 1D np.ndarray)
|
|
x : the initial condition</p>
|
|
<p>Returns
|
|
x the optimal value for x
|
|
flog : a list of all the objective values
|
|
function_eval number of fn evaluations
|
|
status: string describing convergence status</p>
|
|
</dd></dl>
|
|
|
|
<dl class="function">
|
|
<dt id="GPy.inference.optimization.scg.exponents">
|
|
<code class="descclassname">GPy.inference.optimization.scg.</code><code class="descname">exponents</code><span class="sig-paren">(</span><em>fnow</em>, <em>current_grad</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/scg.html#exponents"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.scg.exponents" title="Permalink to this definition">¶</a></dt>
|
|
<dd></dd></dl>
|
|
|
|
<dl class="function">
|
|
<dt id="GPy.inference.optimization.scg.print_out">
|
|
<code class="descclassname">GPy.inference.optimization.scg.</code><code class="descname">print_out</code><span class="sig-paren">(</span><em>len_maxiters</em>, <em>fnow</em>, <em>current_grad</em>, <em>beta</em>, <em>iteration</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/scg.html#print_out"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.scg.print_out" title="Permalink to this definition">¶</a></dt>
|
|
<dd></dd></dl>
|
|
|
|
</div>
|
|
<div class="section" id="gpy-inference-optimization-sgd-module">
|
|
<h2>GPy.inference.optimization.sgd module<a class="headerlink" href="#gpy-inference-optimization-sgd-module" title="Permalink to this headline">¶</a></h2>
|
|
</div>
|
|
<div class="section" id="module-GPy.inference.optimization.stochastics">
|
|
<span id="gpy-inference-optimization-stochastics-module"></span><h2>GPy.inference.optimization.stochastics module<a class="headerlink" href="#module-GPy.inference.optimization.stochastics" title="Permalink to this headline">¶</a></h2>
|
|
<dl class="class">
|
|
<dt id="GPy.inference.optimization.stochastics.SparseGPMissing">
|
|
<em class="property">class </em><code class="descclassname">GPy.inference.optimization.stochastics.</code><code class="descname">SparseGPMissing</code><span class="sig-paren">(</span><em>model</em>, <em>batchsize=1</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/stochastics.html#SparseGPMissing"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.stochastics.SparseGPMissing" title="Permalink to this definition">¶</a></dt>
|
|
<dd><p>Bases: <a class="reference internal" href="#GPy.inference.optimization.stochastics.StochasticStorage" title="GPy.inference.optimization.stochastics.StochasticStorage"><code class="xref py py-class docutils literal"><span class="pre">GPy.inference.optimization.stochastics.StochasticStorage</span></code></a></p>
|
|
</dd></dl>
|
|
|
|
<dl class="class">
|
|
<dt id="GPy.inference.optimization.stochastics.SparseGPStochastics">
|
|
<em class="property">class </em><code class="descclassname">GPy.inference.optimization.stochastics.</code><code class="descname">SparseGPStochastics</code><span class="sig-paren">(</span><em>model</em>, <em>batchsize=1</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/stochastics.html#SparseGPStochastics"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.stochastics.SparseGPStochastics" title="Permalink to this definition">¶</a></dt>
|
|
<dd><p>Bases: <a class="reference internal" href="#GPy.inference.optimization.stochastics.StochasticStorage" title="GPy.inference.optimization.stochastics.StochasticStorage"><code class="xref py py-class docutils literal"><span class="pre">GPy.inference.optimization.stochastics.StochasticStorage</span></code></a></p>
|
|
<p>For the sparse gp we need to store the dimension we are in,
|
|
and the indices corresponding to those</p>
|
|
<dl class="method">
|
|
<dt id="GPy.inference.optimization.stochastics.SparseGPStochastics.do_stochastics">
|
|
<code class="descname">do_stochastics</code><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/stochastics.html#SparseGPStochastics.do_stochastics"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.stochastics.SparseGPStochastics.do_stochastics" title="Permalink to this definition">¶</a></dt>
|
|
<dd></dd></dl>
|
|
|
|
<dl class="method">
|
|
<dt id="GPy.inference.optimization.stochastics.SparseGPStochastics.reset">
|
|
<code class="descname">reset</code><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/stochastics.html#SparseGPStochastics.reset"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.stochastics.SparseGPStochastics.reset" title="Permalink to this definition">¶</a></dt>
|
|
<dd></dd></dl>
|
|
|
|
</dd></dl>
|
|
|
|
<dl class="class">
|
|
<dt id="GPy.inference.optimization.stochastics.StochasticStorage">
|
|
<em class="property">class </em><code class="descclassname">GPy.inference.optimization.stochastics.</code><code class="descname">StochasticStorage</code><span class="sig-paren">(</span><em>model</em><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/stochastics.html#StochasticStorage"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.stochastics.StochasticStorage" title="Permalink to this definition">¶</a></dt>
|
|
<dd><p>Bases: <code class="xref py py-class docutils literal"><span class="pre">object</span></code></p>
|
|
<p>This is a container for holding the stochastic parameters,
|
|
such as subset indices or step length and so on.</p>
|
|
<dl class="method">
|
|
<dt id="GPy.inference.optimization.stochastics.StochasticStorage.do_stochastics">
|
|
<code class="descname">do_stochastics</code><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/stochastics.html#StochasticStorage.do_stochastics"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.stochastics.StochasticStorage.do_stochastics" title="Permalink to this definition">¶</a></dt>
|
|
<dd><p>Update the internal state to the next batch of the stochastic
|
|
descent algorithm.</p>
|
|
</dd></dl>
|
|
|
|
<dl class="method">
|
|
<dt id="GPy.inference.optimization.stochastics.StochasticStorage.reset">
|
|
<code class="descname">reset</code><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="_modules/GPy/inference/optimization/stochastics.html#StochasticStorage.reset"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#GPy.inference.optimization.stochastics.StochasticStorage.reset" title="Permalink to this definition">¶</a></dt>
|
|
<dd><p>Reset the state of this stochastics generator.</p>
|
|
</dd></dl>
|
|
|
|
</dd></dl>
|
|
|
|
</div>
|
|
<div class="section" id="module-GPy.inference.optimization">
|
|
<span id="module-contents"></span><h2>Module contents<a class="headerlink" href="#module-GPy.inference.optimization" title="Permalink to this headline">¶</a></h2>
|
|
</div>
|
|
</div>
|
|
|
|
|
|
</div>
|
|
</div>
|
|
</div>
|
|
<div class="sphinxsidebar" role="navigation" aria-label="main navigation">
|
|
<div class="sphinxsidebarwrapper">
|
|
<h3><a href="index.html">Table Of Contents</a></h3>
|
|
<ul>
|
|
<li><a class="reference internal" href="#">GPy.inference.optimization package</a><ul>
|
|
<li><a class="reference internal" href="#submodules">Submodules</a></li>
|
|
<li><a class="reference internal" href="#module-GPy.inference.optimization.conjugate_gradient_descent">GPy.inference.optimization.conjugate_gradient_descent module</a></li>
|
|
<li><a class="reference internal" href="#module-GPy.inference.optimization.gradient_descent_update_rules">GPy.inference.optimization.gradient_descent_update_rules module</a></li>
|
|
<li><a class="reference internal" href="#module-GPy.inference.optimization.optimization">GPy.inference.optimization.optimization module</a></li>
|
|
<li><a class="reference internal" href="#module-GPy.inference.optimization.scg">GPy.inference.optimization.scg module</a></li>
|
|
<li><a class="reference internal" href="#gpy-inference-optimization-sgd-module">GPy.inference.optimization.sgd module</a></li>
|
|
<li><a class="reference internal" href="#module-GPy.inference.optimization.stochastics">GPy.inference.optimization.stochastics module</a></li>
|
|
<li><a class="reference internal" href="#module-GPy.inference.optimization">Module contents</a></li>
|
|
</ul>
|
|
</li>
|
|
</ul>
|
|
|
|
<h4>Previous topic</h4>
|
|
<p class="topless"><a href="GPy.inference.mcmc.html"
|
|
title="previous chapter">GPy.inference.mcmc package</a></p>
|
|
<h4>Next topic</h4>
|
|
<p class="topless"><a href="GPy.kern.html"
|
|
title="next chapter">GPy.kern package</a></p>
|
|
<div role="note" aria-label="source link">
|
|
<h3>This Page</h3>
|
|
<ul class="this-page-menu">
|
|
<li><a href="_sources/GPy.inference.optimization.txt"
|
|
rel="nofollow">Show Source</a></li>
|
|
</ul>
|
|
</div>
|
|
<div id="searchbox" style="display: none" role="search">
|
|
<h3>Quick search</h3>
|
|
<form class="search" action="search.html" method="get">
|
|
<input type="text" name="q" />
|
|
<input type="submit" value="Go" />
|
|
<input type="hidden" name="check_keywords" value="yes" />
|
|
<input type="hidden" name="area" value="default" />
|
|
</form>
|
|
<p class="searchtip" style="font-size: 90%">
|
|
Enter search terms or a module, class or function name.
|
|
</p>
|
|
</div>
|
|
<script type="text/javascript">$('#searchbox').show(0);</script>
|
|
</div>
|
|
</div>
|
|
<div class="clearer"></div>
|
|
</div>
|
|
<div class="related" role="navigation" aria-label="related navigation">
|
|
<h3>Navigation</h3>
|
|
<ul>
|
|
<li class="right" style="margin-right: 10px">
|
|
<a href="genindex.html" title="General Index"
|
|
>index</a></li>
|
|
<li class="right" >
|
|
<a href="py-modindex.html" title="Python Module Index"
|
|
>modules</a> |</li>
|
|
<li class="right" >
|
|
<a href="GPy.kern.html" title="GPy.kern package"
|
|
>next</a> |</li>
|
|
<li class="right" >
|
|
<a href="GPy.inference.mcmc.html" title="GPy.inference.mcmc package"
|
|
>previous</a> |</li>
|
|
<li class="nav-item nav-item-0"><a href="index.html">GPy documentation</a> »</li>
|
|
<li class="nav-item nav-item-1"><a href="GPy.html" >GPy package</a> »</li>
|
|
<li class="nav-item nav-item-2"><a href="GPy.inference.html" >GPy.inference package</a> »</li>
|
|
</ul>
|
|
</div>
|
|
<div class="footer" role="contentinfo">
|
|
© Copyright 2013, Author.
|
|
Created using <a href="http://sphinx-doc.org/">Sphinx</a> 1.3.1.
|
|
</div>
|
|
</body>
|
|
</html> |