mirror of
https://github.com/gryf/coach.git
synced 2025-12-18 03:30:19 +01:00
update of api docstrings across coach and tutorials [WIP] (#91)
* updating the documentation website * adding the built docs * update of api docstrings across coach and tutorials 0-2 * added some missing api documentation * New Sphinx based documentation
This commit is contained in:
Itai Caspi
330
docs/_modules/rl_coach/exploration_policies/additive_noise.html
Normal file
330
docs/_modules/rl_coach/exploration_policies/additive_noise.html
Normal file
@@ -0,0 +1,330 @@
|
||||
|
||||
|
||||
<!DOCTYPE html>
|
||||
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
|
||||
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
|
||||
<head>
|
||||
<meta charset="utf-8">
|
||||
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||||
|
||||
<title>rl_coach.exploration_policies.additive_noise — Reinforcement Learning Coach 0.11.0 documentation</title>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/css/custom.css" type="text/css" />
|
||||
<link rel="index" title="Index" href="../../../genindex.html" />
|
||||
<link rel="search" title="Search" href="../../../search.html" />
|
||||
<link href="../../../_static/css/custom.css" rel="stylesheet" type="text/css">
|
||||
|
||||
|
||||
|
||||
<script src="../../../_static/js/modernizr.min.js"></script>
|
||||
|
||||
</head>
|
||||
|
||||
<body class="wy-body-for-nav">
|
||||
|
||||
|
||||
<div class="wy-grid-for-nav">
|
||||
|
||||
|
||||
<nav data-toggle="wy-nav-shift" class="wy-nav-side">
|
||||
<div class="wy-side-scroll">
|
||||
<div class="wy-side-nav-search">
|
||||
|
||||
|
||||
|
||||
<a href="../../../index.html" class="icon icon-home"> Reinforcement Learning Coach
|
||||
|
||||
|
||||
|
||||
|
||||
<img src="../../../_static/dark_logo.png" class="logo" alt="Logo"/>
|
||||
|
||||
</a>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="search">
|
||||
<form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
|
||||
<input type="text" name="q" placeholder="Search docs" />
|
||||
<input type="hidden" name="check_keywords" value="yes" />
|
||||
<input type="hidden" name="area" value="default" />
|
||||
</form>
|
||||
</div>
|
||||
|
||||
|
||||
</div>
|
||||
|
||||
<div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<p class="caption"><span class="caption-text">Intro</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../usage.html">Usage</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../features/index.html">Features</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../selecting_an_algorithm.html">Selecting an Algorithm</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../dashboard.html">Coach Dashboard</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Design</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/control_flow.html">Control Flow</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/network.html">Network Design</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Contributing</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_agent.html">Adding a New Agent</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_env.html">Adding a New Environment</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Components</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/agents/index.html">Agents</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/architectures/index.html">Architectures</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/environments/index.html">Environments</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/exploration_policies/index.html">Exploration Policies</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/filters/index.html">Filters</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/memories/index.html">Memories</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/core_types.html">Core Types</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/spaces.html">Spaces</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/additional_parameters.html">Additional Parameters</a></li>
|
||||
</ul>
|
||||
|
||||
|
||||
|
||||
</div>
|
||||
</div>
|
||||
</nav>
|
||||
|
||||
<section data-toggle="wy-nav-shift" class="wy-nav-content-wrap">
|
||||
|
||||
|
||||
<nav class="wy-nav-top" aria-label="top navigation">
|
||||
|
||||
<i data-toggle="wy-nav-top" class="fa fa-bars"></i>
|
||||
<a href="../../../index.html">Reinforcement Learning Coach</a>
|
||||
|
||||
</nav>
|
||||
|
||||
|
||||
<div class="wy-nav-content">
|
||||
|
||||
<div class="rst-content">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="navigation" aria-label="breadcrumbs navigation">
|
||||
|
||||
<ul class="wy-breadcrumbs">
|
||||
|
||||
<li><a href="../../../index.html">Docs</a> »</li>
|
||||
|
||||
<li><a href="../../index.html">Module code</a> »</li>
|
||||
|
||||
<li>rl_coach.exploration_policies.additive_noise</li>
|
||||
|
||||
|
||||
<li class="wy-breadcrumbs-aside">
|
||||
|
||||
</li>
|
||||
|
||||
</ul>
|
||||
|
||||
|
||||
<hr/>
|
||||
</div>
|
||||
<div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
|
||||
<div itemprop="articleBody">
|
||||
|
||||
<h1>Source code for rl_coach.exploration_policies.additive_noise</h1><div class="highlight"><pre>
|
||||
<span></span><span class="c1">#</span>
|
||||
<span class="c1"># Copyright (c) 2017 Intel Corporation </span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Licensed under the Apache License, Version 2.0 (the "License");</span>
|
||||
<span class="c1"># you may not use this file except in compliance with the License.</span>
|
||||
<span class="c1"># You may obtain a copy of the License at</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># http://www.apache.org/licenses/LICENSE-2.0</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Unless required by applicable law or agreed to in writing, software</span>
|
||||
<span class="c1"># distributed under the License is distributed on an "AS IS" BASIS,</span>
|
||||
<span class="c1"># WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.</span>
|
||||
<span class="c1"># See the License for the specific language governing permissions and</span>
|
||||
<span class="c1"># limitations under the License.</span>
|
||||
<span class="c1">#</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">typing</span> <span class="k">import</span> <span class="n">List</span>
|
||||
|
||||
<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.core_types</span> <span class="k">import</span> <span class="n">RunPhase</span><span class="p">,</span> <span class="n">ActionType</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.exploration_policies.exploration_policy</span> <span class="k">import</span> <span class="n">ExplorationPolicy</span><span class="p">,</span> <span class="n">ExplorationParameters</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.schedules</span> <span class="k">import</span> <span class="n">Schedule</span><span class="p">,</span> <span class="n">LinearSchedule</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.spaces</span> <span class="k">import</span> <span class="n">ActionSpace</span><span class="p">,</span> <span class="n">BoxActionSpace</span>
|
||||
|
||||
|
||||
<span class="c1"># TODO: consider renaming to gaussian sampling</span>
|
||||
<span class="k">class</span> <span class="nc">AdditiveNoiseParameters</span><span class="p">(</span><span class="n">ExplorationParameters</span><span class="p">):</span>
|
||||
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">noise_percentage_schedule</span> <span class="o">=</span> <span class="n">LinearSchedule</span><span class="p">(</span><span class="mf">0.1</span><span class="p">,</span> <span class="mf">0.1</span><span class="p">,</span> <span class="mi">50000</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">evaluation_noise_percentage</span> <span class="o">=</span> <span class="mf">0.05</span>
|
||||
|
||||
<span class="nd">@property</span>
|
||||
<span class="k">def</span> <span class="nf">path</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="s1">'rl_coach.exploration_policies.additive_noise:AdditiveNoise'</span>
|
||||
|
||||
|
||||
<div class="viewcode-block" id="AdditiveNoise"><a class="viewcode-back" href="../../../components/exploration_policies/index.html#rl_coach.exploration_policies.AdditiveNoise">[docs]</a><span class="k">class</span> <span class="nc">AdditiveNoise</span><span class="p">(</span><span class="n">ExplorationPolicy</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> AdditiveNoise is an exploration policy intended for continuous action spaces. It takes the action from the agent</span>
|
||||
<span class="sd"> and adds a Gaussian distributed noise to it. The amount of noise added to the action follows the noise amount that</span>
|
||||
<span class="sd"> can be given in two different ways:</span>
|
||||
<span class="sd"> 1. Specified by the user as a noise schedule which is taken in percentiles out of the action space size</span>
|
||||
<span class="sd"> 2. Specified by the agents action. In case the agents action is a list with 2 values, the 1st one is assumed to</span>
|
||||
<span class="sd"> be the mean of the action, and 2nd is assumed to be its standard deviation.</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">action_space</span><span class="p">:</span> <span class="n">ActionSpace</span><span class="p">,</span> <span class="n">noise_percentage_schedule</span><span class="p">:</span> <span class="n">Schedule</span><span class="p">,</span>
|
||||
<span class="n">evaluation_noise_percentage</span><span class="p">:</span> <span class="nb">float</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> :param action_space: the action space used by the environment</span>
|
||||
<span class="sd"> :param noise_percentage_schedule: the schedule for the noise variance percentage relative to the absolute range</span>
|
||||
<span class="sd"> of the action space</span>
|
||||
<span class="sd"> :param evaluation_noise_percentage: the noise variance percentage that will be used during evaluation phases</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">action_space</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">noise_percentage_schedule</span> <span class="o">=</span> <span class="n">noise_percentage_schedule</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">evaluation_noise_percentage</span> <span class="o">=</span> <span class="n">evaluation_noise_percentage</span>
|
||||
|
||||
<span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">action_space</span><span class="p">,</span> <span class="n">BoxActionSpace</span><span class="p">):</span>
|
||||
<span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"Additive noise exploration works only for continuous controls."</span>
|
||||
<span class="s2">"The given action space is of type: </span><span class="si">{}</span><span class="s2">"</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">action_space</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="p">))</span>
|
||||
|
||||
<span class="k">if</span> <span class="ow">not</span> <span class="n">np</span><span class="o">.</span><span class="n">all</span><span class="p">(</span><span class="o">-</span><span class="n">np</span><span class="o">.</span><span class="n">inf</span> <span class="o"><</span> <span class="n">action_space</span><span class="o">.</span><span class="n">high</span><span class="p">)</span> <span class="ow">or</span> <span class="ow">not</span> <span class="n">np</span><span class="o">.</span><span class="n">all</span><span class="p">(</span><span class="n">action_space</span><span class="o">.</span><span class="n">high</span> <span class="o"><</span> <span class="n">np</span><span class="o">.</span><span class="n">inf</span><span class="p">)</span>\
|
||||
<span class="ow">or</span> <span class="ow">not</span> <span class="n">np</span><span class="o">.</span><span class="n">all</span><span class="p">(</span><span class="o">-</span><span class="n">np</span><span class="o">.</span><span class="n">inf</span> <span class="o"><</span> <span class="n">action_space</span><span class="o">.</span><span class="n">low</span><span class="p">)</span> <span class="ow">or</span> <span class="ow">not</span> <span class="n">np</span><span class="o">.</span><span class="n">all</span><span class="p">(</span><span class="n">action_space</span><span class="o">.</span><span class="n">low</span> <span class="o"><</span> <span class="n">np</span><span class="o">.</span><span class="n">inf</span><span class="p">):</span>
|
||||
<span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"Additive noise exploration requires bounded actions"</span><span class="p">)</span>
|
||||
|
||||
<span class="c1"># TODO: allow working with unbounded actions by defining the noise in terms of range and not percentage</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">get_action</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">action_values</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">ActionType</span><span class="p">])</span> <span class="o">-></span> <span class="n">ActionType</span><span class="p">:</span>
|
||||
<span class="c1"># TODO-potential-bug consider separating internally defined stdev and externally defined stdev into 2 policies</span>
|
||||
|
||||
<span class="c1"># set the current noise percentage</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">phase</span> <span class="o">==</span> <span class="n">RunPhase</span><span class="o">.</span><span class="n">TEST</span><span class="p">:</span>
|
||||
<span class="n">current_noise_precentage</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">evaluation_noise_percentage</span>
|
||||
<span class="k">else</span><span class="p">:</span>
|
||||
<span class="n">current_noise_precentage</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">noise_percentage_schedule</span><span class="o">.</span><span class="n">current_value</span>
|
||||
|
||||
<span class="c1"># scale the noise to the action space range</span>
|
||||
<span class="n">action_values_std</span> <span class="o">=</span> <span class="n">current_noise_precentage</span> <span class="o">*</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="o">.</span><span class="n">high</span> <span class="o">-</span> <span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="o">.</span><span class="n">low</span><span class="p">)</span>
|
||||
|
||||
<span class="c1"># extract the mean values</span>
|
||||
<span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">action_values</span><span class="p">,</span> <span class="nb">list</span><span class="p">):</span>
|
||||
<span class="c1"># the action values are expected to be a list with the action mean and optionally the action stdev</span>
|
||||
<span class="n">action_values_mean</span> <span class="o">=</span> <span class="n">action_values</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">squeeze</span><span class="p">()</span>
|
||||
<span class="k">else</span><span class="p">:</span>
|
||||
<span class="c1"># the action values are expected to be a numpy array representing the action mean</span>
|
||||
<span class="n">action_values_mean</span> <span class="o">=</span> <span class="n">action_values</span><span class="o">.</span><span class="n">squeeze</span><span class="p">()</span>
|
||||
|
||||
<span class="c1"># step the noise schedule</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">phase</span> <span class="o">==</span> <span class="n">RunPhase</span><span class="o">.</span><span class="n">TRAIN</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">noise_percentage_schedule</span><span class="o">.</span><span class="n">step</span><span class="p">()</span>
|
||||
<span class="c1"># the second element of the list is assumed to be the standard deviation</span>
|
||||
<span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">action_values</span><span class="p">,</span> <span class="nb">list</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">action_values</span><span class="p">)</span> <span class="o">></span> <span class="mi">1</span><span class="p">:</span>
|
||||
<span class="n">action_values_std</span> <span class="o">=</span> <span class="n">action_values</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">squeeze</span><span class="p">()</span>
|
||||
|
||||
<span class="c1"># add noise to the action means</span>
|
||||
<span class="n">action</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">normal</span><span class="p">(</span><span class="n">action_values_mean</span><span class="p">,</span> <span class="n">action_values_std</span><span class="p">)</span>
|
||||
|
||||
<span class="k">return</span> <span class="n">action</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">get_control_param</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">ones</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span><span class="o">*</span><span class="bp">self</span><span class="o">.</span><span class="n">noise_percentage_schedule</span><span class="o">.</span><span class="n">current_value</span></div>
|
||||
</pre></div>
|
||||
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<footer>
|
||||
|
||||
|
||||
<hr/>
|
||||
|
||||
<div role="contentinfo">
|
||||
<p>
|
||||
© Copyright 2018, Intel AI Lab
|
||||
|
||||
</p>
|
||||
</div>
|
||||
Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
|
||||
|
||||
</footer>
|
||||
|
||||
</div>
|
||||
</div>
|
||||
|
||||
</section>
|
||||
|
||||
</div>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></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 async="async" type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.1/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script>
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" src="../../../_static/js/theme.js"></script>
|
||||
|
||||
<script type="text/javascript">
|
||||
jQuery(function () {
|
||||
SphinxRtdTheme.Navigation.enable(true);
|
||||
});
|
||||
</script>
|
||||
|
||||
</body>
|
||||
</html>
|
||||
292
docs/_modules/rl_coach/exploration_policies/boltzmann.html
Normal file
292
docs/_modules/rl_coach/exploration_policies/boltzmann.html
Normal file
@@ -0,0 +1,292 @@
|
||||
|
||||
|
||||
<!DOCTYPE html>
|
||||
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
|
||||
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
|
||||
<head>
|
||||
<meta charset="utf-8">
|
||||
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||||
|
||||
<title>rl_coach.exploration_policies.boltzmann — Reinforcement Learning Coach 0.11.0 documentation</title>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/css/custom.css" type="text/css" />
|
||||
<link rel="index" title="Index" href="../../../genindex.html" />
|
||||
<link rel="search" title="Search" href="../../../search.html" />
|
||||
<link href="../../../_static/css/custom.css" rel="stylesheet" type="text/css">
|
||||
|
||||
|
||||
|
||||
<script src="../../../_static/js/modernizr.min.js"></script>
|
||||
|
||||
</head>
|
||||
|
||||
<body class="wy-body-for-nav">
|
||||
|
||||
|
||||
<div class="wy-grid-for-nav">
|
||||
|
||||
|
||||
<nav data-toggle="wy-nav-shift" class="wy-nav-side">
|
||||
<div class="wy-side-scroll">
|
||||
<div class="wy-side-nav-search">
|
||||
|
||||
|
||||
|
||||
<a href="../../../index.html" class="icon icon-home"> Reinforcement Learning Coach
|
||||
|
||||
|
||||
|
||||
|
||||
<img src="../../../_static/dark_logo.png" class="logo" alt="Logo"/>
|
||||
|
||||
</a>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="search">
|
||||
<form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
|
||||
<input type="text" name="q" placeholder="Search docs" />
|
||||
<input type="hidden" name="check_keywords" value="yes" />
|
||||
<input type="hidden" name="area" value="default" />
|
||||
</form>
|
||||
</div>
|
||||
|
||||
|
||||
</div>
|
||||
|
||||
<div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<p class="caption"><span class="caption-text">Intro</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../usage.html">Usage</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../features/index.html">Features</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../selecting_an_algorithm.html">Selecting an Algorithm</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../dashboard.html">Coach Dashboard</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Design</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/control_flow.html">Control Flow</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/network.html">Network Design</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Contributing</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_agent.html">Adding a New Agent</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_env.html">Adding a New Environment</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Components</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/agents/index.html">Agents</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/architectures/index.html">Architectures</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/environments/index.html">Environments</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/exploration_policies/index.html">Exploration Policies</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/filters/index.html">Filters</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/memories/index.html">Memories</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/core_types.html">Core Types</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/spaces.html">Spaces</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/additional_parameters.html">Additional Parameters</a></li>
|
||||
</ul>
|
||||
|
||||
|
||||
|
||||
</div>
|
||||
</div>
|
||||
</nav>
|
||||
|
||||
<section data-toggle="wy-nav-shift" class="wy-nav-content-wrap">
|
||||
|
||||
|
||||
<nav class="wy-nav-top" aria-label="top navigation">
|
||||
|
||||
<i data-toggle="wy-nav-top" class="fa fa-bars"></i>
|
||||
<a href="../../../index.html">Reinforcement Learning Coach</a>
|
||||
|
||||
</nav>
|
||||
|
||||
|
||||
<div class="wy-nav-content">
|
||||
|
||||
<div class="rst-content">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="navigation" aria-label="breadcrumbs navigation">
|
||||
|
||||
<ul class="wy-breadcrumbs">
|
||||
|
||||
<li><a href="../../../index.html">Docs</a> »</li>
|
||||
|
||||
<li><a href="../../index.html">Module code</a> »</li>
|
||||
|
||||
<li>rl_coach.exploration_policies.boltzmann</li>
|
||||
|
||||
|
||||
<li class="wy-breadcrumbs-aside">
|
||||
|
||||
</li>
|
||||
|
||||
</ul>
|
||||
|
||||
|
||||
<hr/>
|
||||
</div>
|
||||
<div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
|
||||
<div itemprop="articleBody">
|
||||
|
||||
<h1>Source code for rl_coach.exploration_policies.boltzmann</h1><div class="highlight"><pre>
|
||||
<span></span><span class="c1">#</span>
|
||||
<span class="c1"># Copyright (c) 2017 Intel Corporation </span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Licensed under the Apache License, Version 2.0 (the "License");</span>
|
||||
<span class="c1"># you may not use this file except in compliance with the License.</span>
|
||||
<span class="c1"># You may obtain a copy of the License at</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># http://www.apache.org/licenses/LICENSE-2.0</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Unless required by applicable law or agreed to in writing, software</span>
|
||||
<span class="c1"># distributed under the License is distributed on an "AS IS" BASIS,</span>
|
||||
<span class="c1"># WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.</span>
|
||||
<span class="c1"># See the License for the specific language governing permissions and</span>
|
||||
<span class="c1"># limitations under the License.</span>
|
||||
<span class="c1">#</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">typing</span> <span class="k">import</span> <span class="n">List</span>
|
||||
|
||||
<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.core_types</span> <span class="k">import</span> <span class="n">RunPhase</span><span class="p">,</span> <span class="n">ActionType</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.exploration_policies.exploration_policy</span> <span class="k">import</span> <span class="n">ExplorationPolicy</span><span class="p">,</span> <span class="n">ExplorationParameters</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.schedules</span> <span class="k">import</span> <span class="n">Schedule</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.spaces</span> <span class="k">import</span> <span class="n">ActionSpace</span>
|
||||
|
||||
|
||||
<span class="k">class</span> <span class="nc">BoltzmannParameters</span><span class="p">(</span><span class="n">ExplorationParameters</span><span class="p">):</span>
|
||||
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">temperature_schedule</span> <span class="o">=</span> <span class="kc">None</span>
|
||||
|
||||
<span class="nd">@property</span>
|
||||
<span class="k">def</span> <span class="nf">path</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="s1">'rl_coach.exploration_policies.boltzmann:Boltzmann'</span>
|
||||
|
||||
|
||||
|
||||
<div class="viewcode-block" id="Boltzmann"><a class="viewcode-back" href="../../../components/exploration_policies/index.html#rl_coach.exploration_policies.Boltzmann">[docs]</a><span class="k">class</span> <span class="nc">Boltzmann</span><span class="p">(</span><span class="n">ExplorationPolicy</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> The Boltzmann exploration policy is intended for discrete action spaces. It assumes that each of the possible</span>
|
||||
<span class="sd"> actions has some value assigned to it (such as the Q value), and uses a softmax function to convert these values</span>
|
||||
<span class="sd"> into a distribution over the actions. It then samples the action for playing out of the calculated distribution.</span>
|
||||
<span class="sd"> An additional temperature schedule can be given by the user, and will control the steepness of the softmax function.</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">action_space</span><span class="p">:</span> <span class="n">ActionSpace</span><span class="p">,</span> <span class="n">temperature_schedule</span><span class="p">:</span> <span class="n">Schedule</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> :param action_space: the action space used by the environment</span>
|
||||
<span class="sd"> :param temperature_schedule: the schedule for the temperature parameter of the softmax</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">action_space</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">temperature_schedule</span> <span class="o">=</span> <span class="n">temperature_schedule</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">get_action</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">action_values</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">ActionType</span><span class="p">])</span> <span class="o">-></span> <span class="n">ActionType</span><span class="p">:</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">phase</span> <span class="o">==</span> <span class="n">RunPhase</span><span class="o">.</span><span class="n">TRAIN</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">temperature_schedule</span><span class="o">.</span><span class="n">step</span><span class="p">()</span>
|
||||
<span class="c1"># softmax calculation</span>
|
||||
<span class="n">exp_probabilities</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span><span class="n">action_values</span> <span class="o">/</span> <span class="bp">self</span><span class="o">.</span><span class="n">temperature_schedule</span><span class="o">.</span><span class="n">current_value</span><span class="p">)</span>
|
||||
<span class="n">probabilities</span> <span class="o">=</span> <span class="n">exp_probabilities</span> <span class="o">/</span> <span class="n">np</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">exp_probabilities</span><span class="p">)</span>
|
||||
<span class="c1"># make sure probs sum to 1</span>
|
||||
<span class="n">probabilities</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">-</span> <span class="n">np</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">probabilities</span><span class="p">[:</span><span class="o">-</span><span class="mi">1</span><span class="p">])</span>
|
||||
<span class="c1"># choose actions according to the probabilities</span>
|
||||
<span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">choice</span><span class="p">(</span><span class="nb">range</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="o">.</span><span class="n">shape</span><span class="p">),</span> <span class="n">p</span><span class="o">=</span><span class="n">probabilities</span><span class="p">)</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">get_control_param</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">temperature_schedule</span><span class="o">.</span><span class="n">current_value</span></div>
|
||||
</pre></div>
|
||||
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<footer>
|
||||
|
||||
|
||||
<hr/>
|
||||
|
||||
<div role="contentinfo">
|
||||
<p>
|
||||
© Copyright 2018, Intel AI Lab
|
||||
|
||||
</p>
|
||||
</div>
|
||||
Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
|
||||
|
||||
</footer>
|
||||
|
||||
</div>
|
||||
</div>
|
||||
|
||||
</section>
|
||||
|
||||
</div>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></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 async="async" type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.1/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script>
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" src="../../../_static/js/theme.js"></script>
|
||||
|
||||
<script type="text/javascript">
|
||||
jQuery(function () {
|
||||
SphinxRtdTheme.Navigation.enable(true);
|
||||
});
|
||||
</script>
|
||||
|
||||
</body>
|
||||
</html>
|
||||
315
docs/_modules/rl_coach/exploration_policies/bootstrapped.html
Normal file
315
docs/_modules/rl_coach/exploration_policies/bootstrapped.html
Normal file
@@ -0,0 +1,315 @@
|
||||
|
||||
|
||||
<!DOCTYPE html>
|
||||
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
|
||||
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
|
||||
<head>
|
||||
<meta charset="utf-8">
|
||||
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||||
|
||||
<title>rl_coach.exploration_policies.bootstrapped — Reinforcement Learning Coach 0.11.0 documentation</title>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/css/custom.css" type="text/css" />
|
||||
<link rel="index" title="Index" href="../../../genindex.html" />
|
||||
<link rel="search" title="Search" href="../../../search.html" />
|
||||
<link href="../../../_static/css/custom.css" rel="stylesheet" type="text/css">
|
||||
|
||||
|
||||
|
||||
<script src="../../../_static/js/modernizr.min.js"></script>
|
||||
|
||||
</head>
|
||||
|
||||
<body class="wy-body-for-nav">
|
||||
|
||||
|
||||
<div class="wy-grid-for-nav">
|
||||
|
||||
|
||||
<nav data-toggle="wy-nav-shift" class="wy-nav-side">
|
||||
<div class="wy-side-scroll">
|
||||
<div class="wy-side-nav-search">
|
||||
|
||||
|
||||
|
||||
<a href="../../../index.html" class="icon icon-home"> Reinforcement Learning Coach
|
||||
|
||||
|
||||
|
||||
|
||||
<img src="../../../_static/dark_logo.png" class="logo" alt="Logo"/>
|
||||
|
||||
</a>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="search">
|
||||
<form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
|
||||
<input type="text" name="q" placeholder="Search docs" />
|
||||
<input type="hidden" name="check_keywords" value="yes" />
|
||||
<input type="hidden" name="area" value="default" />
|
||||
</form>
|
||||
</div>
|
||||
|
||||
|
||||
</div>
|
||||
|
||||
<div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<p class="caption"><span class="caption-text">Intro</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../usage.html">Usage</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../features/index.html">Features</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../selecting_an_algorithm.html">Selecting an Algorithm</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../dashboard.html">Coach Dashboard</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Design</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/control_flow.html">Control Flow</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/network.html">Network Design</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Contributing</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_agent.html">Adding a New Agent</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_env.html">Adding a New Environment</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Components</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/agents/index.html">Agents</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/architectures/index.html">Architectures</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/environments/index.html">Environments</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/exploration_policies/index.html">Exploration Policies</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/filters/index.html">Filters</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/memories/index.html">Memories</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/core_types.html">Core Types</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/spaces.html">Spaces</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/additional_parameters.html">Additional Parameters</a></li>
|
||||
</ul>
|
||||
|
||||
|
||||
|
||||
</div>
|
||||
</div>
|
||||
</nav>
|
||||
|
||||
<section data-toggle="wy-nav-shift" class="wy-nav-content-wrap">
|
||||
|
||||
|
||||
<nav class="wy-nav-top" aria-label="top navigation">
|
||||
|
||||
<i data-toggle="wy-nav-top" class="fa fa-bars"></i>
|
||||
<a href="../../../index.html">Reinforcement Learning Coach</a>
|
||||
|
||||
</nav>
|
||||
|
||||
|
||||
<div class="wy-nav-content">
|
||||
|
||||
<div class="rst-content">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="navigation" aria-label="breadcrumbs navigation">
|
||||
|
||||
<ul class="wy-breadcrumbs">
|
||||
|
||||
<li><a href="../../../index.html">Docs</a> »</li>
|
||||
|
||||
<li><a href="../../index.html">Module code</a> »</li>
|
||||
|
||||
<li>rl_coach.exploration_policies.bootstrapped</li>
|
||||
|
||||
|
||||
<li class="wy-breadcrumbs-aside">
|
||||
|
||||
</li>
|
||||
|
||||
</ul>
|
||||
|
||||
|
||||
<hr/>
|
||||
</div>
|
||||
<div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
|
||||
<div itemprop="articleBody">
|
||||
|
||||
<h1>Source code for rl_coach.exploration_policies.bootstrapped</h1><div class="highlight"><pre>
|
||||
<span></span><span class="c1">#</span>
|
||||
<span class="c1"># Copyright (c) 2017 Intel Corporation </span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Licensed under the Apache License, Version 2.0 (the "License");</span>
|
||||
<span class="c1"># you may not use this file except in compliance with the License.</span>
|
||||
<span class="c1"># You may obtain a copy of the License at</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># http://www.apache.org/licenses/LICENSE-2.0</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Unless required by applicable law or agreed to in writing, software</span>
|
||||
<span class="c1"># distributed under the License is distributed on an "AS IS" BASIS,</span>
|
||||
<span class="c1"># WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.</span>
|
||||
<span class="c1"># See the License for the specific language governing permissions and</span>
|
||||
<span class="c1"># limitations under the License.</span>
|
||||
<span class="c1">#</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">typing</span> <span class="k">import</span> <span class="n">List</span>
|
||||
|
||||
<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.core_types</span> <span class="k">import</span> <span class="n">RunPhase</span><span class="p">,</span> <span class="n">ActionType</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.exploration_policies.additive_noise</span> <span class="k">import</span> <span class="n">AdditiveNoiseParameters</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.exploration_policies.e_greedy</span> <span class="k">import</span> <span class="n">EGreedy</span><span class="p">,</span> <span class="n">EGreedyParameters</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.exploration_policies.exploration_policy</span> <span class="k">import</span> <span class="n">ExplorationParameters</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.schedules</span> <span class="k">import</span> <span class="n">Schedule</span><span class="p">,</span> <span class="n">LinearSchedule</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.spaces</span> <span class="k">import</span> <span class="n">ActionSpace</span>
|
||||
|
||||
|
||||
<span class="k">class</span> <span class="nc">BootstrappedParameters</span><span class="p">(</span><span class="n">EGreedyParameters</span><span class="p">):</span>
|
||||
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">architecture_num_q_heads</span> <span class="o">=</span> <span class="mi">10</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">bootstrapped_data_sharing_probability</span> <span class="o">=</span> <span class="mf">1.0</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">epsilon_schedule</span> <span class="o">=</span> <span class="n">LinearSchedule</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mf">0.01</span><span class="p">,</span> <span class="mi">1000000</span><span class="p">)</span>
|
||||
|
||||
<span class="nd">@property</span>
|
||||
<span class="k">def</span> <span class="nf">path</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="s1">'rl_coach.exploration_policies.bootstrapped:Bootstrapped'</span>
|
||||
|
||||
|
||||
<div class="viewcode-block" id="Bootstrapped"><a class="viewcode-back" href="../../../components/exploration_policies/index.html#rl_coach.exploration_policies.Bootstrapped">[docs]</a><span class="k">class</span> <span class="nc">Bootstrapped</span><span class="p">(</span><span class="n">EGreedy</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> Bootstrapped exploration policy is currently only used for discrete action spaces along with the</span>
|
||||
<span class="sd"> Bootstrapped DQN agent. It assumes that there is an ensemble of network heads, where each one predicts the</span>
|
||||
<span class="sd"> values for all the possible actions. For each episode, a single head is selected to lead the agent, according</span>
|
||||
<span class="sd"> to its value predictions. In evaluation, the action is selected using a majority vote over all the heads</span>
|
||||
<span class="sd"> predictions.</span>
|
||||
|
||||
<span class="sd"> .. note::</span>
|
||||
<span class="sd"> This exploration policy will only work for Discrete action spaces with Bootstrapped DQN style agents,</span>
|
||||
<span class="sd"> since it requires the agent to have a network with multiple heads.</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">action_space</span><span class="p">:</span> <span class="n">ActionSpace</span><span class="p">,</span> <span class="n">epsilon_schedule</span><span class="p">:</span> <span class="n">Schedule</span><span class="p">,</span> <span class="n">evaluation_epsilon</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
|
||||
<span class="n">architecture_num_q_heads</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
|
||||
<span class="n">continuous_exploration_policy_parameters</span><span class="p">:</span> <span class="n">ExplorationParameters</span> <span class="o">=</span> <span class="n">AdditiveNoiseParameters</span><span class="p">(),):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> :param action_space: the action space used by the environment</span>
|
||||
<span class="sd"> :param epsilon_schedule: a schedule for the epsilon values</span>
|
||||
<span class="sd"> :param evaluation_epsilon: the epsilon value to use for evaluation phases</span>
|
||||
<span class="sd"> :param continuous_exploration_policy_parameters: the parameters of the continuous exploration policy to use</span>
|
||||
<span class="sd"> if the e-greedy is used for a continuous policy</span>
|
||||
<span class="sd"> :param architecture_num_q_heads: the number of q heads to select from</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">action_space</span><span class="p">,</span> <span class="n">epsilon_schedule</span><span class="p">,</span> <span class="n">evaluation_epsilon</span><span class="p">,</span> <span class="n">continuous_exploration_policy_parameters</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span> <span class="o">=</span> <span class="n">architecture_num_q_heads</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">selected_head</span> <span class="o">=</span> <span class="mi">0</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">last_action_values</span> <span class="o">=</span> <span class="mi">0</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">select_head</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">selected_head</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span><span class="p">)</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">get_action</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">action_values</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">ActionType</span><span class="p">])</span> <span class="o">-></span> <span class="n">ActionType</span><span class="p">:</span>
|
||||
<span class="c1"># action values are none in case the exploration policy is going to select a random action</span>
|
||||
<span class="k">if</span> <span class="n">action_values</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">phase</span> <span class="o">==</span> <span class="n">RunPhase</span><span class="o">.</span><span class="n">TRAIN</span><span class="p">:</span>
|
||||
<span class="n">action_values</span> <span class="o">=</span> <span class="n">action_values</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">selected_head</span><span class="p">]</span>
|
||||
<span class="k">else</span><span class="p">:</span>
|
||||
<span class="c1"># ensemble voting for evaluation</span>
|
||||
<span class="n">top_action_votings</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><span class="n">action_values</span><span class="p">,</span> <span class="n">axis</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
|
||||
<span class="n">counts</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">bincount</span><span class="p">(</span><span class="n">top_action_votings</span><span class="o">.</span><span class="n">squeeze</span><span class="p">())</span>
|
||||
<span class="n">top_action</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><span class="n">counts</span><span class="p">)</span>
|
||||
<span class="c1"># convert the top action to a one hot vector and pass it to e-greedy</span>
|
||||
<span class="n">action_values</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">eye</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="o">.</span><span class="n">actions</span><span class="p">))[[</span><span class="n">top_action</span><span class="p">]]</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">last_action_values</span> <span class="o">=</span> <span class="n">action_values</span>
|
||||
<span class="k">return</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">get_action</span><span class="p">(</span><span class="n">action_values</span><span class="p">)</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">get_control_param</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">selected_head</span></div>
|
||||
</pre></div>
|
||||
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<footer>
|
||||
|
||||
|
||||
<hr/>
|
||||
|
||||
<div role="contentinfo">
|
||||
<p>
|
||||
© Copyright 2018, Intel AI Lab
|
||||
|
||||
</p>
|
||||
</div>
|
||||
Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
|
||||
|
||||
</footer>
|
||||
|
||||
</div>
|
||||
</div>
|
||||
|
||||
</section>
|
||||
|
||||
</div>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></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 async="async" type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.1/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script>
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" src="../../../_static/js/theme.js"></script>
|
||||
|
||||
<script type="text/javascript">
|
||||
jQuery(function () {
|
||||
SphinxRtdTheme.Navigation.enable(true);
|
||||
});
|
||||
</script>
|
||||
|
||||
</body>
|
||||
</html>
|
||||
281
docs/_modules/rl_coach/exploration_policies/categorical.html
Normal file
281
docs/_modules/rl_coach/exploration_policies/categorical.html
Normal file
@@ -0,0 +1,281 @@
|
||||
|
||||
|
||||
<!DOCTYPE html>
|
||||
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
|
||||
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
|
||||
<head>
|
||||
<meta charset="utf-8">
|
||||
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||||
|
||||
<title>rl_coach.exploration_policies.categorical — Reinforcement Learning Coach 0.11.0 documentation</title>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/css/custom.css" type="text/css" />
|
||||
<link rel="index" title="Index" href="../../../genindex.html" />
|
||||
<link rel="search" title="Search" href="../../../search.html" />
|
||||
<link href="../../../_static/css/custom.css" rel="stylesheet" type="text/css">
|
||||
|
||||
|
||||
|
||||
<script src="../../../_static/js/modernizr.min.js"></script>
|
||||
|
||||
</head>
|
||||
|
||||
<body class="wy-body-for-nav">
|
||||
|
||||
|
||||
<div class="wy-grid-for-nav">
|
||||
|
||||
|
||||
<nav data-toggle="wy-nav-shift" class="wy-nav-side">
|
||||
<div class="wy-side-scroll">
|
||||
<div class="wy-side-nav-search">
|
||||
|
||||
|
||||
|
||||
<a href="../../../index.html" class="icon icon-home"> Reinforcement Learning Coach
|
||||
|
||||
|
||||
|
||||
|
||||
<img src="../../../_static/dark_logo.png" class="logo" alt="Logo"/>
|
||||
|
||||
</a>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="search">
|
||||
<form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
|
||||
<input type="text" name="q" placeholder="Search docs" />
|
||||
<input type="hidden" name="check_keywords" value="yes" />
|
||||
<input type="hidden" name="area" value="default" />
|
||||
</form>
|
||||
</div>
|
||||
|
||||
|
||||
</div>
|
||||
|
||||
<div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<p class="caption"><span class="caption-text">Intro</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../usage.html">Usage</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../features/index.html">Features</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../selecting_an_algorithm.html">Selecting an Algorithm</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../dashboard.html">Coach Dashboard</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Design</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/control_flow.html">Control Flow</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/network.html">Network Design</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Contributing</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_agent.html">Adding a New Agent</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_env.html">Adding a New Environment</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Components</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/agents/index.html">Agents</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/architectures/index.html">Architectures</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/environments/index.html">Environments</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/exploration_policies/index.html">Exploration Policies</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/filters/index.html">Filters</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/memories/index.html">Memories</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/core_types.html">Core Types</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/spaces.html">Spaces</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/additional_parameters.html">Additional Parameters</a></li>
|
||||
</ul>
|
||||
|
||||
|
||||
|
||||
</div>
|
||||
</div>
|
||||
</nav>
|
||||
|
||||
<section data-toggle="wy-nav-shift" class="wy-nav-content-wrap">
|
||||
|
||||
|
||||
<nav class="wy-nav-top" aria-label="top navigation">
|
||||
|
||||
<i data-toggle="wy-nav-top" class="fa fa-bars"></i>
|
||||
<a href="../../../index.html">Reinforcement Learning Coach</a>
|
||||
|
||||
</nav>
|
||||
|
||||
|
||||
<div class="wy-nav-content">
|
||||
|
||||
<div class="rst-content">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="navigation" aria-label="breadcrumbs navigation">
|
||||
|
||||
<ul class="wy-breadcrumbs">
|
||||
|
||||
<li><a href="../../../index.html">Docs</a> »</li>
|
||||
|
||||
<li><a href="../../index.html">Module code</a> »</li>
|
||||
|
||||
<li>rl_coach.exploration_policies.categorical</li>
|
||||
|
||||
|
||||
<li class="wy-breadcrumbs-aside">
|
||||
|
||||
</li>
|
||||
|
||||
</ul>
|
||||
|
||||
|
||||
<hr/>
|
||||
</div>
|
||||
<div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
|
||||
<div itemprop="articleBody">
|
||||
|
||||
<h1>Source code for rl_coach.exploration_policies.categorical</h1><div class="highlight"><pre>
|
||||
<span></span><span class="c1">#</span>
|
||||
<span class="c1"># Copyright (c) 2017 Intel Corporation </span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Licensed under the Apache License, Version 2.0 (the "License");</span>
|
||||
<span class="c1"># you may not use this file except in compliance with the License.</span>
|
||||
<span class="c1"># You may obtain a copy of the License at</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># http://www.apache.org/licenses/LICENSE-2.0</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Unless required by applicable law or agreed to in writing, software</span>
|
||||
<span class="c1"># distributed under the License is distributed on an "AS IS" BASIS,</span>
|
||||
<span class="c1"># WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.</span>
|
||||
<span class="c1"># See the License for the specific language governing permissions and</span>
|
||||
<span class="c1"># limitations under the License.</span>
|
||||
<span class="c1">#</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">typing</span> <span class="k">import</span> <span class="n">List</span>
|
||||
|
||||
<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.core_types</span> <span class="k">import</span> <span class="n">RunPhase</span><span class="p">,</span> <span class="n">ActionType</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.exploration_policies.exploration_policy</span> <span class="k">import</span> <span class="n">ExplorationPolicy</span><span class="p">,</span> <span class="n">ExplorationParameters</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.spaces</span> <span class="k">import</span> <span class="n">ActionSpace</span>
|
||||
|
||||
|
||||
<span class="k">class</span> <span class="nc">CategoricalParameters</span><span class="p">(</span><span class="n">ExplorationParameters</span><span class="p">):</span>
|
||||
<span class="nd">@property</span>
|
||||
<span class="k">def</span> <span class="nf">path</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="s1">'rl_coach.exploration_policies.categorical:Categorical'</span>
|
||||
|
||||
|
||||
<div class="viewcode-block" id="Categorical"><a class="viewcode-back" href="../../../components/exploration_policies/index.html#rl_coach.exploration_policies.Categorical">[docs]</a><span class="k">class</span> <span class="nc">Categorical</span><span class="p">(</span><span class="n">ExplorationPolicy</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> Categorical exploration policy is intended for discrete action spaces. It expects the action values to</span>
|
||||
<span class="sd"> represent a probability distribution over the action, from which a single action will be sampled.</span>
|
||||
<span class="sd"> In evaluation, the action that has the highest probability will be selected. This is particularly useful for</span>
|
||||
<span class="sd"> actor-critic schemes, where the actors output is a probability distribution over the actions.</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">action_space</span><span class="p">:</span> <span class="n">ActionSpace</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> :param action_space: the action space used by the environment</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">action_space</span><span class="p">)</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">get_action</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">action_values</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">ActionType</span><span class="p">])</span> <span class="o">-></span> <span class="n">ActionType</span><span class="p">:</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">phase</span> <span class="o">==</span> <span class="n">RunPhase</span><span class="o">.</span><span class="n">TRAIN</span><span class="p">:</span>
|
||||
<span class="c1"># choose actions according to the probabilities</span>
|
||||
<span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">choice</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="o">.</span><span class="n">actions</span><span class="p">,</span> <span class="n">p</span><span class="o">=</span><span class="n">action_values</span><span class="p">)</span>
|
||||
<span class="k">else</span><span class="p">:</span>
|
||||
<span class="c1"># take the action with the highest probability</span>
|
||||
<span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><span class="n">action_values</span><span class="p">)</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">get_control_param</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="mi">0</span></div>
|
||||
</pre></div>
|
||||
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<footer>
|
||||
|
||||
|
||||
<hr/>
|
||||
|
||||
<div role="contentinfo">
|
||||
<p>
|
||||
© Copyright 2018, Intel AI Lab
|
||||
|
||||
</p>
|
||||
</div>
|
||||
Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
|
||||
|
||||
</footer>
|
||||
|
||||
</div>
|
||||
</div>
|
||||
|
||||
</section>
|
||||
|
||||
</div>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></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 async="async" type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.1/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script>
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" src="../../../_static/js/theme.js"></script>
|
||||
|
||||
<script type="text/javascript">
|
||||
jQuery(function () {
|
||||
SphinxRtdTheme.Navigation.enable(true);
|
||||
});
|
||||
</script>
|
||||
|
||||
</body>
|
||||
</html>
|
||||
@@ -0,0 +1,265 @@
|
||||
|
||||
|
||||
<!DOCTYPE html>
|
||||
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
|
||||
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
|
||||
<head>
|
||||
<meta charset="utf-8">
|
||||
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||||
|
||||
<title>rl_coach.exploration_policies.continuous_entropy — Reinforcement Learning Coach 0.11.0 documentation</title>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/css/custom.css" type="text/css" />
|
||||
<link rel="index" title="Index" href="../../../genindex.html" />
|
||||
<link rel="search" title="Search" href="../../../search.html" />
|
||||
<link href="../../../_static/css/custom.css" rel="stylesheet" type="text/css">
|
||||
|
||||
|
||||
|
||||
<script src="../../../_static/js/modernizr.min.js"></script>
|
||||
|
||||
</head>
|
||||
|
||||
<body class="wy-body-for-nav">
|
||||
|
||||
|
||||
<div class="wy-grid-for-nav">
|
||||
|
||||
|
||||
<nav data-toggle="wy-nav-shift" class="wy-nav-side">
|
||||
<div class="wy-side-scroll">
|
||||
<div class="wy-side-nav-search">
|
||||
|
||||
|
||||
|
||||
<a href="../../../index.html" class="icon icon-home"> Reinforcement Learning Coach
|
||||
|
||||
|
||||
|
||||
|
||||
<img src="../../../_static/dark_logo.png" class="logo" alt="Logo"/>
|
||||
|
||||
</a>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="search">
|
||||
<form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
|
||||
<input type="text" name="q" placeholder="Search docs" />
|
||||
<input type="hidden" name="check_keywords" value="yes" />
|
||||
<input type="hidden" name="area" value="default" />
|
||||
</form>
|
||||
</div>
|
||||
|
||||
|
||||
</div>
|
||||
|
||||
<div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<p class="caption"><span class="caption-text">Intro</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../usage.html">Usage</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../features/index.html">Features</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../selecting_an_algorithm.html">Selecting an Algorithm</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../dashboard.html">Coach Dashboard</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Design</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/control_flow.html">Control Flow</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/network.html">Network Design</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Contributing</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_agent.html">Adding a New Agent</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_env.html">Adding a New Environment</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Components</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/agents/index.html">Agents</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/architectures/index.html">Architectures</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/environments/index.html">Environments</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/exploration_policies/index.html">Exploration Policies</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/filters/index.html">Filters</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/memories/index.html">Memories</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/core_types.html">Core Types</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/spaces.html">Spaces</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/additional_parameters.html">Additional Parameters</a></li>
|
||||
</ul>
|
||||
|
||||
|
||||
|
||||
</div>
|
||||
</div>
|
||||
</nav>
|
||||
|
||||
<section data-toggle="wy-nav-shift" class="wy-nav-content-wrap">
|
||||
|
||||
|
||||
<nav class="wy-nav-top" aria-label="top navigation">
|
||||
|
||||
<i data-toggle="wy-nav-top" class="fa fa-bars"></i>
|
||||
<a href="../../../index.html">Reinforcement Learning Coach</a>
|
||||
|
||||
</nav>
|
||||
|
||||
|
||||
<div class="wy-nav-content">
|
||||
|
||||
<div class="rst-content">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="navigation" aria-label="breadcrumbs navigation">
|
||||
|
||||
<ul class="wy-breadcrumbs">
|
||||
|
||||
<li><a href="../../../index.html">Docs</a> »</li>
|
||||
|
||||
<li><a href="../../index.html">Module code</a> »</li>
|
||||
|
||||
<li>rl_coach.exploration_policies.continuous_entropy</li>
|
||||
|
||||
|
||||
<li class="wy-breadcrumbs-aside">
|
||||
|
||||
</li>
|
||||
|
||||
</ul>
|
||||
|
||||
|
||||
<hr/>
|
||||
</div>
|
||||
<div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
|
||||
<div itemprop="articleBody">
|
||||
|
||||
<h1>Source code for rl_coach.exploration_policies.continuous_entropy</h1><div class="highlight"><pre>
|
||||
<span></span><span class="c1">#</span>
|
||||
<span class="c1"># Copyright (c) 2017 Intel Corporation </span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Licensed under the Apache License, Version 2.0 (the "License");</span>
|
||||
<span class="c1"># you may not use this file except in compliance with the License.</span>
|
||||
<span class="c1"># You may obtain a copy of the License at</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># http://www.apache.org/licenses/LICENSE-2.0</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Unless required by applicable law or agreed to in writing, software</span>
|
||||
<span class="c1"># distributed under the License is distributed on an "AS IS" BASIS,</span>
|
||||
<span class="c1"># WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.</span>
|
||||
<span class="c1"># See the License for the specific language governing permissions and</span>
|
||||
<span class="c1"># limitations under the License.</span>
|
||||
<span class="c1">#</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.exploration_policies.additive_noise</span> <span class="k">import</span> <span class="n">AdditiveNoise</span><span class="p">,</span> <span class="n">AdditiveNoiseParameters</span>
|
||||
|
||||
|
||||
<span class="k">class</span> <span class="nc">ContinuousEntropyParameters</span><span class="p">(</span><span class="n">AdditiveNoiseParameters</span><span class="p">):</span>
|
||||
<span class="nd">@property</span>
|
||||
<span class="k">def</span> <span class="nf">path</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="s1">'rl_coach.exploration_policies.continuous_entropy:ContinuousEntropy'</span>
|
||||
|
||||
|
||||
<div class="viewcode-block" id="ContinuousEntropy"><a class="viewcode-back" href="../../../components/exploration_policies/index.html#rl_coach.exploration_policies.ContinuousEntropy">[docs]</a><span class="k">class</span> <span class="nc">ContinuousEntropy</span><span class="p">(</span><span class="n">AdditiveNoise</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> Continuous entropy is an exploration policy that is actually implemented as part of the network.</span>
|
||||
<span class="sd"> The exploration policy class is only a placeholder for choosing this policy. The exploration policy is</span>
|
||||
<span class="sd"> implemented by adding a regularization factor to the network loss, which regularizes the entropy of the action.</span>
|
||||
<span class="sd"> This exploration policy is only intended for continuous action spaces, and assumes that the entire calculation</span>
|
||||
<span class="sd"> is implemented as part of the head.</span>
|
||||
|
||||
<span class="sd"> .. warning::</span>
|
||||
<span class="sd"> This exploration policy expects the agent or the network to implement the exploration functionality.</span>
|
||||
<span class="sd"> There are only a few heads that actually are relevant and implement the entropy regularization factor.</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="k">pass</span></div>
|
||||
</pre></div>
|
||||
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<footer>
|
||||
|
||||
|
||||
<hr/>
|
||||
|
||||
<div role="contentinfo">
|
||||
<p>
|
||||
© Copyright 2018, Intel AI Lab
|
||||
|
||||
</p>
|
||||
</div>
|
||||
Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
|
||||
|
||||
</footer>
|
||||
|
||||
</div>
|
||||
</div>
|
||||
|
||||
</section>
|
||||
|
||||
</div>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></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 async="async" type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.1/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script>
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" src="../../../_static/js/theme.js"></script>
|
||||
|
||||
<script type="text/javascript">
|
||||
jQuery(function () {
|
||||
SphinxRtdTheme.Navigation.enable(true);
|
||||
});
|
||||
</script>
|
||||
|
||||
</body>
|
||||
</html>
|
||||
342
docs/_modules/rl_coach/exploration_policies/e_greedy.html
Normal file
342
docs/_modules/rl_coach/exploration_policies/e_greedy.html
Normal file
@@ -0,0 +1,342 @@
|
||||
|
||||
|
||||
<!DOCTYPE html>
|
||||
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
|
||||
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
|
||||
<head>
|
||||
<meta charset="utf-8">
|
||||
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||||
|
||||
<title>rl_coach.exploration_policies.e_greedy — Reinforcement Learning Coach 0.11.0 documentation</title>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/css/custom.css" type="text/css" />
|
||||
<link rel="index" title="Index" href="../../../genindex.html" />
|
||||
<link rel="search" title="Search" href="../../../search.html" />
|
||||
<link href="../../../_static/css/custom.css" rel="stylesheet" type="text/css">
|
||||
|
||||
|
||||
|
||||
<script src="../../../_static/js/modernizr.min.js"></script>
|
||||
|
||||
</head>
|
||||
|
||||
<body class="wy-body-for-nav">
|
||||
|
||||
|
||||
<div class="wy-grid-for-nav">
|
||||
|
||||
|
||||
<nav data-toggle="wy-nav-shift" class="wy-nav-side">
|
||||
<div class="wy-side-scroll">
|
||||
<div class="wy-side-nav-search">
|
||||
|
||||
|
||||
|
||||
<a href="../../../index.html" class="icon icon-home"> Reinforcement Learning Coach
|
||||
|
||||
|
||||
|
||||
|
||||
<img src="../../../_static/dark_logo.png" class="logo" alt="Logo"/>
|
||||
|
||||
</a>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="search">
|
||||
<form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
|
||||
<input type="text" name="q" placeholder="Search docs" />
|
||||
<input type="hidden" name="check_keywords" value="yes" />
|
||||
<input type="hidden" name="area" value="default" />
|
||||
</form>
|
||||
</div>
|
||||
|
||||
|
||||
</div>
|
||||
|
||||
<div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<p class="caption"><span class="caption-text">Intro</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../usage.html">Usage</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../features/index.html">Features</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../selecting_an_algorithm.html">Selecting an Algorithm</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../dashboard.html">Coach Dashboard</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Design</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/control_flow.html">Control Flow</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/network.html">Network Design</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Contributing</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_agent.html">Adding a New Agent</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_env.html">Adding a New Environment</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Components</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/agents/index.html">Agents</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/architectures/index.html">Architectures</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/environments/index.html">Environments</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/exploration_policies/index.html">Exploration Policies</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/filters/index.html">Filters</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/memories/index.html">Memories</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/core_types.html">Core Types</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/spaces.html">Spaces</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/additional_parameters.html">Additional Parameters</a></li>
|
||||
</ul>
|
||||
|
||||
|
||||
|
||||
</div>
|
||||
</div>
|
||||
</nav>
|
||||
|
||||
<section data-toggle="wy-nav-shift" class="wy-nav-content-wrap">
|
||||
|
||||
|
||||
<nav class="wy-nav-top" aria-label="top navigation">
|
||||
|
||||
<i data-toggle="wy-nav-top" class="fa fa-bars"></i>
|
||||
<a href="../../../index.html">Reinforcement Learning Coach</a>
|
||||
|
||||
</nav>
|
||||
|
||||
|
||||
<div class="wy-nav-content">
|
||||
|
||||
<div class="rst-content">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="navigation" aria-label="breadcrumbs navigation">
|
||||
|
||||
<ul class="wy-breadcrumbs">
|
||||
|
||||
<li><a href="../../../index.html">Docs</a> »</li>
|
||||
|
||||
<li><a href="../../index.html">Module code</a> »</li>
|
||||
|
||||
<li>rl_coach.exploration_policies.e_greedy</li>
|
||||
|
||||
|
||||
<li class="wy-breadcrumbs-aside">
|
||||
|
||||
</li>
|
||||
|
||||
</ul>
|
||||
|
||||
|
||||
<hr/>
|
||||
</div>
|
||||
<div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
|
||||
<div itemprop="articleBody">
|
||||
|
||||
<h1>Source code for rl_coach.exploration_policies.e_greedy</h1><div class="highlight"><pre>
|
||||
<span></span><span class="c1">#</span>
|
||||
<span class="c1"># Copyright (c) 2017 Intel Corporation </span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Licensed under the Apache License, Version 2.0 (the "License");</span>
|
||||
<span class="c1"># you may not use this file except in compliance with the License.</span>
|
||||
<span class="c1"># You may obtain a copy of the License at</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># http://www.apache.org/licenses/LICENSE-2.0</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Unless required by applicable law or agreed to in writing, software</span>
|
||||
<span class="c1"># distributed under the License is distributed on an "AS IS" BASIS,</span>
|
||||
<span class="c1"># WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.</span>
|
||||
<span class="c1"># See the License for the specific language governing permissions and</span>
|
||||
<span class="c1"># limitations under the License.</span>
|
||||
<span class="c1">#</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">typing</span> <span class="k">import</span> <span class="n">List</span>
|
||||
|
||||
<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.core_types</span> <span class="k">import</span> <span class="n">RunPhase</span><span class="p">,</span> <span class="n">ActionType</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.exploration_policies.additive_noise</span> <span class="k">import</span> <span class="n">AdditiveNoiseParameters</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.exploration_policies.exploration_policy</span> <span class="k">import</span> <span class="n">ExplorationParameters</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.exploration_policies.exploration_policy</span> <span class="k">import</span> <span class="n">ExplorationPolicy</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.schedules</span> <span class="k">import</span> <span class="n">Schedule</span><span class="p">,</span> <span class="n">LinearSchedule</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.spaces</span> <span class="k">import</span> <span class="n">ActionSpace</span><span class="p">,</span> <span class="n">DiscreteActionSpace</span><span class="p">,</span> <span class="n">BoxActionSpace</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.utils</span> <span class="k">import</span> <span class="n">dynamic_import_and_instantiate_module_from_params</span>
|
||||
|
||||
|
||||
<span class="k">class</span> <span class="nc">EGreedyParameters</span><span class="p">(</span><span class="n">ExplorationParameters</span><span class="p">):</span>
|
||||
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">epsilon_schedule</span> <span class="o">=</span> <span class="n">LinearSchedule</span><span class="p">(</span><span class="mf">0.5</span><span class="p">,</span> <span class="mf">0.01</span><span class="p">,</span> <span class="mi">50000</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">evaluation_epsilon</span> <span class="o">=</span> <span class="mf">0.05</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">continuous_exploration_policy_parameters</span> <span class="o">=</span> <span class="n">AdditiveNoiseParameters</span><span class="p">()</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">continuous_exploration_policy_parameters</span><span class="o">.</span><span class="n">noise_percentage_schedule</span> <span class="o">=</span> <span class="n">LinearSchedule</span><span class="p">(</span><span class="mf">0.1</span><span class="p">,</span> <span class="mf">0.1</span><span class="p">,</span> <span class="mi">50000</span><span class="p">)</span>
|
||||
<span class="c1"># for continuous control -</span>
|
||||
<span class="c1"># (see http://www.cs.ubc.ca/~van/papers/2017-TOG-deepLoco/2017-TOG-deepLoco.pdf)</span>
|
||||
|
||||
<span class="nd">@property</span>
|
||||
<span class="k">def</span> <span class="nf">path</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="s1">'rl_coach.exploration_policies.e_greedy:EGreedy'</span>
|
||||
|
||||
|
||||
<div class="viewcode-block" id="EGreedy"><a class="viewcode-back" href="../../../components/exploration_policies/index.html#rl_coach.exploration_policies.EGreedy">[docs]</a><span class="k">class</span> <span class="nc">EGreedy</span><span class="p">(</span><span class="n">ExplorationPolicy</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> e-greedy is an exploration policy that is intended for both discrete and continuous action spaces.</span>
|
||||
|
||||
<span class="sd"> For discrete action spaces, it assumes that each action is assigned a value, and it selects the action with the</span>
|
||||
<span class="sd"> highest value with probability 1 - epsilon. Otherwise, it selects a action sampled uniformly out of all the</span>
|
||||
<span class="sd"> possible actions. The epsilon value is given by the user and can be given as a schedule.</span>
|
||||
<span class="sd"> In evaluation, a different epsilon value can be specified.</span>
|
||||
|
||||
<span class="sd"> For continuous action spaces, it assumes that the mean action is given by the agent. With probability epsilon,</span>
|
||||
<span class="sd"> it samples a random action out of the action space bounds. Otherwise, it selects the action according to a</span>
|
||||
<span class="sd"> given continuous exploration policy, which is set to AdditiveNoise by default. In evaluation, the action is</span>
|
||||
<span class="sd"> always selected according to the given continuous exploration policy (where its phase is set to evaluation as well).</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">action_space</span><span class="p">:</span> <span class="n">ActionSpace</span><span class="p">,</span> <span class="n">epsilon_schedule</span><span class="p">:</span> <span class="n">Schedule</span><span class="p">,</span>
|
||||
<span class="n">evaluation_epsilon</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
|
||||
<span class="n">continuous_exploration_policy_parameters</span><span class="p">:</span> <span class="n">ExplorationParameters</span><span class="o">=</span><span class="n">AdditiveNoiseParameters</span><span class="p">()):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> :param action_space: the action space used by the environment</span>
|
||||
<span class="sd"> :param epsilon_schedule: a schedule for the epsilon values</span>
|
||||
<span class="sd"> :param evaluation_epsilon: the epsilon value to use for evaluation phases</span>
|
||||
<span class="sd"> :param continuous_exploration_policy_parameters: the parameters of the continuous exploration policy to use</span>
|
||||
<span class="sd"> if the e-greedy is used for a continuous policy</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">action_space</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">epsilon_schedule</span> <span class="o">=</span> <span class="n">epsilon_schedule</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">evaluation_epsilon</span> <span class="o">=</span> <span class="n">evaluation_epsilon</span>
|
||||
|
||||
<span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="p">,</span> <span class="n">BoxActionSpace</span><span class="p">):</span>
|
||||
<span class="c1"># for continuous e-greedy (see http://www.cs.ubc.ca/~van/papers/2017-TOG-deepLoco/2017-TOG-deepLoco.pdf)</span>
|
||||
<span class="n">continuous_exploration_policy_parameters</span><span class="o">.</span><span class="n">action_space</span> <span class="o">=</span> <span class="n">action_space</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">continuous_exploration_policy</span> <span class="o">=</span> \
|
||||
<span class="n">dynamic_import_and_instantiate_module_from_params</span><span class="p">(</span><span class="n">continuous_exploration_policy_parameters</span><span class="p">)</span>
|
||||
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">current_random_value</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">rand</span><span class="p">()</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">requires_action_values</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="n">epsilon</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">evaluation_epsilon</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">phase</span> <span class="o">==</span> <span class="n">RunPhase</span><span class="o">.</span><span class="n">TEST</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">epsilon_schedule</span><span class="o">.</span><span class="n">current_value</span>
|
||||
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">current_random_value</span> <span class="o">>=</span> <span class="n">epsilon</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">get_action</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">action_values</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">ActionType</span><span class="p">])</span> <span class="o">-></span> <span class="n">ActionType</span><span class="p">:</span>
|
||||
<span class="n">epsilon</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">evaluation_epsilon</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">phase</span> <span class="o">==</span> <span class="n">RunPhase</span><span class="o">.</span><span class="n">TEST</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">epsilon_schedule</span><span class="o">.</span><span class="n">current_value</span>
|
||||
|
||||
<span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="p">,</span> <span class="n">DiscreteActionSpace</span><span class="p">):</span>
|
||||
<span class="n">top_action</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><span class="n">action_values</span><span class="p">)</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">current_random_value</span> <span class="o"><</span> <span class="n">epsilon</span><span class="p">:</span>
|
||||
<span class="n">chosen_action</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="o">.</span><span class="n">sample</span><span class="p">()</span>
|
||||
<span class="k">else</span><span class="p">:</span>
|
||||
<span class="n">chosen_action</span> <span class="o">=</span> <span class="n">top_action</span>
|
||||
<span class="k">else</span><span class="p">:</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">current_random_value</span> <span class="o"><</span> <span class="n">epsilon</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">phase</span> <span class="o">==</span> <span class="n">RunPhase</span><span class="o">.</span><span class="n">TRAIN</span><span class="p">:</span>
|
||||
<span class="n">chosen_action</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="o">.</span><span class="n">sample</span><span class="p">()</span>
|
||||
<span class="k">else</span><span class="p">:</span>
|
||||
<span class="n">chosen_action</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">continuous_exploration_policy</span><span class="o">.</span><span class="n">get_action</span><span class="p">(</span><span class="n">action_values</span><span class="p">)</span>
|
||||
|
||||
<span class="c1"># step the epsilon schedule and generate a new random value for next time</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">phase</span> <span class="o">==</span> <span class="n">RunPhase</span><span class="o">.</span><span class="n">TRAIN</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">epsilon_schedule</span><span class="o">.</span><span class="n">step</span><span class="p">()</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">current_random_value</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">rand</span><span class="p">()</span>
|
||||
<span class="k">return</span> <span class="n">chosen_action</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">get_control_param</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="p">,</span> <span class="n">DiscreteActionSpace</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">evaluation_epsilon</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">phase</span> <span class="o">==</span> <span class="n">RunPhase</span><span class="o">.</span><span class="n">TEST</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">epsilon_schedule</span><span class="o">.</span><span class="n">current_value</span>
|
||||
<span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="p">,</span> <span class="n">BoxActionSpace</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">continuous_exploration_policy</span><span class="o">.</span><span class="n">get_control_param</span><span class="p">()</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">change_phase</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">phase</span><span class="p">):</span>
|
||||
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">change_phase</span><span class="p">(</span><span class="n">phase</span><span class="p">)</span>
|
||||
<span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="p">,</span> <span class="n">BoxActionSpace</span><span class="p">):</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">continuous_exploration_policy</span><span class="o">.</span><span class="n">change_phase</span><span class="p">(</span><span class="n">phase</span><span class="p">)</span></div>
|
||||
</pre></div>
|
||||
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<footer>
|
||||
|
||||
|
||||
<hr/>
|
||||
|
||||
<div role="contentinfo">
|
||||
<p>
|
||||
© Copyright 2018, Intel AI Lab
|
||||
|
||||
</p>
|
||||
</div>
|
||||
Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
|
||||
|
||||
</footer>
|
||||
|
||||
</div>
|
||||
</div>
|
||||
|
||||
</section>
|
||||
|
||||
</div>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></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 async="async" type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.1/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script>
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" src="../../../_static/js/theme.js"></script>
|
||||
|
||||
<script type="text/javascript">
|
||||
jQuery(function () {
|
||||
SphinxRtdTheme.Navigation.enable(true);
|
||||
});
|
||||
</script>
|
||||
|
||||
</body>
|
||||
</html>
|
||||
@@ -0,0 +1,311 @@
|
||||
|
||||
|
||||
<!DOCTYPE html>
|
||||
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
|
||||
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
|
||||
<head>
|
||||
<meta charset="utf-8">
|
||||
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||||
|
||||
<title>rl_coach.exploration_policies.exploration_policy — Reinforcement Learning Coach 0.11.0 documentation</title>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/css/custom.css" type="text/css" />
|
||||
<link rel="index" title="Index" href="../../../genindex.html" />
|
||||
<link rel="search" title="Search" href="../../../search.html" />
|
||||
<link href="../../../_static/css/custom.css" rel="stylesheet" type="text/css">
|
||||
|
||||
|
||||
|
||||
<script src="../../../_static/js/modernizr.min.js"></script>
|
||||
|
||||
</head>
|
||||
|
||||
<body class="wy-body-for-nav">
|
||||
|
||||
|
||||
<div class="wy-grid-for-nav">
|
||||
|
||||
|
||||
<nav data-toggle="wy-nav-shift" class="wy-nav-side">
|
||||
<div class="wy-side-scroll">
|
||||
<div class="wy-side-nav-search">
|
||||
|
||||
|
||||
|
||||
<a href="../../../index.html" class="icon icon-home"> Reinforcement Learning Coach
|
||||
|
||||
|
||||
|
||||
|
||||
<img src="../../../_static/dark_logo.png" class="logo" alt="Logo"/>
|
||||
|
||||
</a>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="search">
|
||||
<form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
|
||||
<input type="text" name="q" placeholder="Search docs" />
|
||||
<input type="hidden" name="check_keywords" value="yes" />
|
||||
<input type="hidden" name="area" value="default" />
|
||||
</form>
|
||||
</div>
|
||||
|
||||
|
||||
</div>
|
||||
|
||||
<div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<p class="caption"><span class="caption-text">Intro</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../usage.html">Usage</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../features/index.html">Features</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../selecting_an_algorithm.html">Selecting an Algorithm</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../dashboard.html">Coach Dashboard</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Design</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/control_flow.html">Control Flow</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/network.html">Network Design</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Contributing</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_agent.html">Adding a New Agent</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_env.html">Adding a New Environment</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Components</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/agents/index.html">Agents</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/architectures/index.html">Architectures</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/environments/index.html">Environments</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/exploration_policies/index.html">Exploration Policies</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/filters/index.html">Filters</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/memories/index.html">Memories</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/core_types.html">Core Types</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/spaces.html">Spaces</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/additional_parameters.html">Additional Parameters</a></li>
|
||||
</ul>
|
||||
|
||||
|
||||
|
||||
</div>
|
||||
</div>
|
||||
</nav>
|
||||
|
||||
<section data-toggle="wy-nav-shift" class="wy-nav-content-wrap">
|
||||
|
||||
|
||||
<nav class="wy-nav-top" aria-label="top navigation">
|
||||
|
||||
<i data-toggle="wy-nav-top" class="fa fa-bars"></i>
|
||||
<a href="../../../index.html">Reinforcement Learning Coach</a>
|
||||
|
||||
</nav>
|
||||
|
||||
|
||||
<div class="wy-nav-content">
|
||||
|
||||
<div class="rst-content">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="navigation" aria-label="breadcrumbs navigation">
|
||||
|
||||
<ul class="wy-breadcrumbs">
|
||||
|
||||
<li><a href="../../../index.html">Docs</a> »</li>
|
||||
|
||||
<li><a href="../../index.html">Module code</a> »</li>
|
||||
|
||||
<li>rl_coach.exploration_policies.exploration_policy</li>
|
||||
|
||||
|
||||
<li class="wy-breadcrumbs-aside">
|
||||
|
||||
</li>
|
||||
|
||||
</ul>
|
||||
|
||||
|
||||
<hr/>
|
||||
</div>
|
||||
<div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
|
||||
<div itemprop="articleBody">
|
||||
|
||||
<h1>Source code for rl_coach.exploration_policies.exploration_policy</h1><div class="highlight"><pre>
|
||||
<span></span><span class="c1">#</span>
|
||||
<span class="c1"># Copyright (c) 2017 Intel Corporation </span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Licensed under the Apache License, Version 2.0 (the "License");</span>
|
||||
<span class="c1"># you may not use this file except in compliance with the License.</span>
|
||||
<span class="c1"># You may obtain a copy of the License at</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># http://www.apache.org/licenses/LICENSE-2.0</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Unless required by applicable law or agreed to in writing, software</span>
|
||||
<span class="c1"># distributed under the License is distributed on an "AS IS" BASIS,</span>
|
||||
<span class="c1"># WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.</span>
|
||||
<span class="c1"># See the License for the specific language governing permissions and</span>
|
||||
<span class="c1"># limitations under the License.</span>
|
||||
<span class="c1">#</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">typing</span> <span class="k">import</span> <span class="n">List</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.base_parameters</span> <span class="k">import</span> <span class="n">Parameters</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.core_types</span> <span class="k">import</span> <span class="n">RunPhase</span><span class="p">,</span> <span class="n">ActionType</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.spaces</span> <span class="k">import</span> <span class="n">ActionSpace</span>
|
||||
|
||||
|
||||
<span class="k">class</span> <span class="nc">ExplorationParameters</span><span class="p">(</span><span class="n">Parameters</span><span class="p">):</span>
|
||||
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">action_space</span> <span class="o">=</span> <span class="kc">None</span>
|
||||
|
||||
<span class="nd">@property</span>
|
||||
<span class="k">def</span> <span class="nf">path</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="s1">'rl_coach.exploration_policies.exploration_policy:ExplorationPolicy'</span>
|
||||
|
||||
|
||||
<div class="viewcode-block" id="ExplorationPolicy"><a class="viewcode-back" href="../../../components/exploration_policies/index.html#rl_coach.exploration_policies.ExplorationPolicy">[docs]</a><span class="k">class</span> <span class="nc">ExplorationPolicy</span><span class="p">(</span><span class="nb">object</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> An exploration policy takes the predicted actions or action values from the agent, and selects the action to</span>
|
||||
<span class="sd"> actually apply to the environment using some predefined algorithm.</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">action_space</span><span class="p">:</span> <span class="n">ActionSpace</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> :param action_space: the action space used by the environment</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">phase</span> <span class="o">=</span> <span class="n">RunPhase</span><span class="o">.</span><span class="n">HEATUP</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">action_space</span> <span class="o">=</span> <span class="n">action_space</span>
|
||||
|
||||
<div class="viewcode-block" id="ExplorationPolicy.reset"><a class="viewcode-back" href="../../../components/exploration_policies/index.html#rl_coach.exploration_policies.ExplorationPolicy.reset">[docs]</a> <span class="k">def</span> <span class="nf">reset</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> Used for resetting the exploration policy parameters when needed</span>
|
||||
<span class="sd"> :return: None</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="k">pass</span></div>
|
||||
|
||||
<div class="viewcode-block" id="ExplorationPolicy.get_action"><a class="viewcode-back" href="../../../components/exploration_policies/index.html#rl_coach.exploration_policies.ExplorationPolicy.get_action">[docs]</a> <span class="k">def</span> <span class="nf">get_action</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">action_values</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">ActionType</span><span class="p">])</span> <span class="o">-></span> <span class="n">ActionType</span><span class="p">:</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> Given a list of values corresponding to each action, </span>
|
||||
<span class="sd"> choose one actions according to the exploration policy</span>
|
||||
<span class="sd"> :param action_values: A list of action values</span>
|
||||
<span class="sd"> :return: The chosen action</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span> <span class="o">==</span> <span class="n">ExplorationPolicy</span><span class="p">:</span>
|
||||
<span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"The ExplorationPolicy class is an abstract class and should not be used directly. "</span>
|
||||
<span class="s2">"Please set the exploration parameters to point to an inheriting class like EGreedy or "</span>
|
||||
<span class="s2">"AdditiveNoise"</span><span class="p">)</span>
|
||||
<span class="k">else</span><span class="p">:</span>
|
||||
<span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"The get_action function should be overridden in the inheriting exploration class"</span><span class="p">)</span></div>
|
||||
|
||||
<div class="viewcode-block" id="ExplorationPolicy.change_phase"><a class="viewcode-back" href="../../../components/exploration_policies/index.html#rl_coach.exploration_policies.ExplorationPolicy.change_phase">[docs]</a> <span class="k">def</span> <span class="nf">change_phase</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">phase</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> Change between running phases of the algorithm</span>
|
||||
<span class="sd"> :param phase: Either Heatup or Train</span>
|
||||
<span class="sd"> :return: none</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">phase</span> <span class="o">=</span> <span class="n">phase</span></div>
|
||||
|
||||
<div class="viewcode-block" id="ExplorationPolicy.requires_action_values"><a class="viewcode-back" href="../../../components/exploration_policies/index.html#rl_coach.exploration_policies.ExplorationPolicy.requires_action_values">[docs]</a> <span class="k">def</span> <span class="nf">requires_action_values</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-></span> <span class="nb">bool</span><span class="p">:</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> Allows exploration policies to define if they require the action values for the current step.</span>
|
||||
<span class="sd"> This can save up a lot of computation. For example in e-greedy, if the random value generated is smaller</span>
|
||||
<span class="sd"> than epsilon, the action is completely random, and the action values don't need to be calculated</span>
|
||||
<span class="sd"> :return: True if the action values are required. False otherwise</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="k">return</span> <span class="kc">True</span></div>
|
||||
|
||||
<span class="k">def</span> <span class="nf">get_control_param</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="mi">0</span></div>
|
||||
</pre></div>
|
||||
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<footer>
|
||||
|
||||
|
||||
<hr/>
|
||||
|
||||
<div role="contentinfo">
|
||||
<p>
|
||||
© Copyright 2018, Intel AI Lab
|
||||
|
||||
</p>
|
||||
</div>
|
||||
Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
|
||||
|
||||
</footer>
|
||||
|
||||
</div>
|
||||
</div>
|
||||
|
||||
</section>
|
||||
|
||||
</div>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></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 async="async" type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.1/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script>
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" src="../../../_static/js/theme.js"></script>
|
||||
|
||||
<script type="text/javascript">
|
||||
jQuery(function () {
|
||||
SphinxRtdTheme.Navigation.enable(true);
|
||||
});
|
||||
</script>
|
||||
|
||||
</body>
|
||||
</html>
|
||||
278
docs/_modules/rl_coach/exploration_policies/greedy.html
Normal file
278
docs/_modules/rl_coach/exploration_policies/greedy.html
Normal file
@@ -0,0 +1,278 @@
|
||||
|
||||
|
||||
<!DOCTYPE html>
|
||||
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
|
||||
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
|
||||
<head>
|
||||
<meta charset="utf-8">
|
||||
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||||
|
||||
<title>rl_coach.exploration_policies.greedy — Reinforcement Learning Coach 0.11.0 documentation</title>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/css/custom.css" type="text/css" />
|
||||
<link rel="index" title="Index" href="../../../genindex.html" />
|
||||
<link rel="search" title="Search" href="../../../search.html" />
|
||||
<link href="../../../_static/css/custom.css" rel="stylesheet" type="text/css">
|
||||
|
||||
|
||||
|
||||
<script src="../../../_static/js/modernizr.min.js"></script>
|
||||
|
||||
</head>
|
||||
|
||||
<body class="wy-body-for-nav">
|
||||
|
||||
|
||||
<div class="wy-grid-for-nav">
|
||||
|
||||
|
||||
<nav data-toggle="wy-nav-shift" class="wy-nav-side">
|
||||
<div class="wy-side-scroll">
|
||||
<div class="wy-side-nav-search">
|
||||
|
||||
|
||||
|
||||
<a href="../../../index.html" class="icon icon-home"> Reinforcement Learning Coach
|
||||
|
||||
|
||||
|
||||
|
||||
<img src="../../../_static/dark_logo.png" class="logo" alt="Logo"/>
|
||||
|
||||
</a>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="search">
|
||||
<form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
|
||||
<input type="text" name="q" placeholder="Search docs" />
|
||||
<input type="hidden" name="check_keywords" value="yes" />
|
||||
<input type="hidden" name="area" value="default" />
|
||||
</form>
|
||||
</div>
|
||||
|
||||
|
||||
</div>
|
||||
|
||||
<div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<p class="caption"><span class="caption-text">Intro</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../usage.html">Usage</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../features/index.html">Features</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../selecting_an_algorithm.html">Selecting an Algorithm</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../dashboard.html">Coach Dashboard</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Design</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/control_flow.html">Control Flow</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/network.html">Network Design</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Contributing</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_agent.html">Adding a New Agent</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_env.html">Adding a New Environment</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Components</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/agents/index.html">Agents</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/architectures/index.html">Architectures</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/environments/index.html">Environments</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/exploration_policies/index.html">Exploration Policies</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/filters/index.html">Filters</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/memories/index.html">Memories</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/core_types.html">Core Types</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/spaces.html">Spaces</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/additional_parameters.html">Additional Parameters</a></li>
|
||||
</ul>
|
||||
|
||||
|
||||
|
||||
</div>
|
||||
</div>
|
||||
</nav>
|
||||
|
||||
<section data-toggle="wy-nav-shift" class="wy-nav-content-wrap">
|
||||
|
||||
|
||||
<nav class="wy-nav-top" aria-label="top navigation">
|
||||
|
||||
<i data-toggle="wy-nav-top" class="fa fa-bars"></i>
|
||||
<a href="../../../index.html">Reinforcement Learning Coach</a>
|
||||
|
||||
</nav>
|
||||
|
||||
|
||||
<div class="wy-nav-content">
|
||||
|
||||
<div class="rst-content">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="navigation" aria-label="breadcrumbs navigation">
|
||||
|
||||
<ul class="wy-breadcrumbs">
|
||||
|
||||
<li><a href="../../../index.html">Docs</a> »</li>
|
||||
|
||||
<li><a href="../../index.html">Module code</a> »</li>
|
||||
|
||||
<li>rl_coach.exploration_policies.greedy</li>
|
||||
|
||||
|
||||
<li class="wy-breadcrumbs-aside">
|
||||
|
||||
</li>
|
||||
|
||||
</ul>
|
||||
|
||||
|
||||
<hr/>
|
||||
</div>
|
||||
<div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
|
||||
<div itemprop="articleBody">
|
||||
|
||||
<h1>Source code for rl_coach.exploration_policies.greedy</h1><div class="highlight"><pre>
|
||||
<span></span><span class="c1">#</span>
|
||||
<span class="c1"># Copyright (c) 2017 Intel Corporation </span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Licensed under the Apache License, Version 2.0 (the "License");</span>
|
||||
<span class="c1"># you may not use this file except in compliance with the License.</span>
|
||||
<span class="c1"># You may obtain a copy of the License at</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># http://www.apache.org/licenses/LICENSE-2.0</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Unless required by applicable law or agreed to in writing, software</span>
|
||||
<span class="c1"># distributed under the License is distributed on an "AS IS" BASIS,</span>
|
||||
<span class="c1"># WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.</span>
|
||||
<span class="c1"># See the License for the specific language governing permissions and</span>
|
||||
<span class="c1"># limitations under the License.</span>
|
||||
<span class="c1">#</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">typing</span> <span class="k">import</span> <span class="n">List</span>
|
||||
|
||||
<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.core_types</span> <span class="k">import</span> <span class="n">ActionType</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.exploration_policies.exploration_policy</span> <span class="k">import</span> <span class="n">ExplorationPolicy</span><span class="p">,</span> <span class="n">ExplorationParameters</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.spaces</span> <span class="k">import</span> <span class="n">ActionSpace</span><span class="p">,</span> <span class="n">DiscreteActionSpace</span><span class="p">,</span> <span class="n">BoxActionSpace</span>
|
||||
|
||||
|
||||
<span class="k">class</span> <span class="nc">GreedyParameters</span><span class="p">(</span><span class="n">ExplorationParameters</span><span class="p">):</span>
|
||||
<span class="nd">@property</span>
|
||||
<span class="k">def</span> <span class="nf">path</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="s1">'rl_coach.exploration_policies.greedy:Greedy'</span>
|
||||
|
||||
|
||||
<div class="viewcode-block" id="Greedy"><a class="viewcode-back" href="../../../components/exploration_policies/index.html#rl_coach.exploration_policies.Greedy">[docs]</a><span class="k">class</span> <span class="nc">Greedy</span><span class="p">(</span><span class="n">ExplorationPolicy</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> The Greedy exploration policy is intended for both discrete and continuous action spaces.</span>
|
||||
<span class="sd"> For discrete action spaces, it always selects the action with the maximum value, as given by the agent.</span>
|
||||
<span class="sd"> For continuous action spaces, it always return the exact action, as it was given by the agent.</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">action_space</span><span class="p">:</span> <span class="n">ActionSpace</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> :param action_space: the action space used by the environment</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">action_space</span><span class="p">)</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">get_action</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">action_values</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">ActionType</span><span class="p">])</span> <span class="o">-></span> <span class="n">ActionType</span><span class="p">:</span>
|
||||
<span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="p">)</span> <span class="o">==</span> <span class="n">DiscreteActionSpace</span><span class="p">:</span>
|
||||
<span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><span class="n">action_values</span><span class="p">)</span>
|
||||
<span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="p">)</span> <span class="o">==</span> <span class="n">BoxActionSpace</span><span class="p">:</span>
|
||||
<span class="k">return</span> <span class="n">action_values</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">get_control_param</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="mi">0</span></div>
|
||||
</pre></div>
|
||||
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<footer>
|
||||
|
||||
|
||||
<hr/>
|
||||
|
||||
<div role="contentinfo">
|
||||
<p>
|
||||
© Copyright 2018, Intel AI Lab
|
||||
|
||||
</p>
|
||||
</div>
|
||||
Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
|
||||
|
||||
</footer>
|
||||
|
||||
</div>
|
||||
</div>
|
||||
|
||||
</section>
|
||||
|
||||
</div>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></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 async="async" type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.1/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script>
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" src="../../../_static/js/theme.js"></script>
|
||||
|
||||
<script type="text/javascript">
|
||||
jQuery(function () {
|
||||
SphinxRtdTheme.Navigation.enable(true);
|
||||
});
|
||||
</script>
|
||||
|
||||
</body>
|
||||
</html>
|
||||
313
docs/_modules/rl_coach/exploration_policies/ou_process.html
Normal file
313
docs/_modules/rl_coach/exploration_policies/ou_process.html
Normal file
@@ -0,0 +1,313 @@
|
||||
|
||||
|
||||
<!DOCTYPE html>
|
||||
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
|
||||
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
|
||||
<head>
|
||||
<meta charset="utf-8">
|
||||
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||||
|
||||
<title>rl_coach.exploration_policies.ou_process — Reinforcement Learning Coach 0.11.0 documentation</title>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/css/custom.css" type="text/css" />
|
||||
<link rel="index" title="Index" href="../../../genindex.html" />
|
||||
<link rel="search" title="Search" href="../../../search.html" />
|
||||
<link href="../../../_static/css/custom.css" rel="stylesheet" type="text/css">
|
||||
|
||||
|
||||
|
||||
<script src="../../../_static/js/modernizr.min.js"></script>
|
||||
|
||||
</head>
|
||||
|
||||
<body class="wy-body-for-nav">
|
||||
|
||||
|
||||
<div class="wy-grid-for-nav">
|
||||
|
||||
|
||||
<nav data-toggle="wy-nav-shift" class="wy-nav-side">
|
||||
<div class="wy-side-scroll">
|
||||
<div class="wy-side-nav-search">
|
||||
|
||||
|
||||
|
||||
<a href="../../../index.html" class="icon icon-home"> Reinforcement Learning Coach
|
||||
|
||||
|
||||
|
||||
|
||||
<img src="../../../_static/dark_logo.png" class="logo" alt="Logo"/>
|
||||
|
||||
</a>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="search">
|
||||
<form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
|
||||
<input type="text" name="q" placeholder="Search docs" />
|
||||
<input type="hidden" name="check_keywords" value="yes" />
|
||||
<input type="hidden" name="area" value="default" />
|
||||
</form>
|
||||
</div>
|
||||
|
||||
|
||||
</div>
|
||||
|
||||
<div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<p class="caption"><span class="caption-text">Intro</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../usage.html">Usage</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../features/index.html">Features</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../selecting_an_algorithm.html">Selecting an Algorithm</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../dashboard.html">Coach Dashboard</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Design</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/control_flow.html">Control Flow</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/network.html">Network Design</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Contributing</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_agent.html">Adding a New Agent</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_env.html">Adding a New Environment</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Components</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/agents/index.html">Agents</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/architectures/index.html">Architectures</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/environments/index.html">Environments</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/exploration_policies/index.html">Exploration Policies</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/filters/index.html">Filters</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/memories/index.html">Memories</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/core_types.html">Core Types</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/spaces.html">Spaces</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/additional_parameters.html">Additional Parameters</a></li>
|
||||
</ul>
|
||||
|
||||
|
||||
|
||||
</div>
|
||||
</div>
|
||||
</nav>
|
||||
|
||||
<section data-toggle="wy-nav-shift" class="wy-nav-content-wrap">
|
||||
|
||||
|
||||
<nav class="wy-nav-top" aria-label="top navigation">
|
||||
|
||||
<i data-toggle="wy-nav-top" class="fa fa-bars"></i>
|
||||
<a href="../../../index.html">Reinforcement Learning Coach</a>
|
||||
|
||||
</nav>
|
||||
|
||||
|
||||
<div class="wy-nav-content">
|
||||
|
||||
<div class="rst-content">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="navigation" aria-label="breadcrumbs navigation">
|
||||
|
||||
<ul class="wy-breadcrumbs">
|
||||
|
||||
<li><a href="../../../index.html">Docs</a> »</li>
|
||||
|
||||
<li><a href="../../index.html">Module code</a> »</li>
|
||||
|
||||
<li>rl_coach.exploration_policies.ou_process</li>
|
||||
|
||||
|
||||
<li class="wy-breadcrumbs-aside">
|
||||
|
||||
</li>
|
||||
|
||||
</ul>
|
||||
|
||||
|
||||
<hr/>
|
||||
</div>
|
||||
<div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
|
||||
<div itemprop="articleBody">
|
||||
|
||||
<h1>Source code for rl_coach.exploration_policies.ou_process</h1><div class="highlight"><pre>
|
||||
<span></span><span class="c1">#</span>
|
||||
<span class="c1"># Copyright (c) 2017 Intel Corporation </span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Licensed under the Apache License, Version 2.0 (the "License");</span>
|
||||
<span class="c1"># you may not use this file except in compliance with the License.</span>
|
||||
<span class="c1"># You may obtain a copy of the License at</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># http://www.apache.org/licenses/LICENSE-2.0</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Unless required by applicable law or agreed to in writing, software</span>
|
||||
<span class="c1"># distributed under the License is distributed on an "AS IS" BASIS,</span>
|
||||
<span class="c1"># WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.</span>
|
||||
<span class="c1"># See the License for the specific language governing permissions and</span>
|
||||
<span class="c1"># limitations under the License.</span>
|
||||
<span class="c1">#</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">typing</span> <span class="k">import</span> <span class="n">List</span>
|
||||
|
||||
<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.core_types</span> <span class="k">import</span> <span class="n">RunPhase</span><span class="p">,</span> <span class="n">ActionType</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.exploration_policies.exploration_policy</span> <span class="k">import</span> <span class="n">ExplorationPolicy</span><span class="p">,</span> <span class="n">ExplorationParameters</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.spaces</span> <span class="k">import</span> <span class="n">ActionSpace</span><span class="p">,</span> <span class="n">BoxActionSpace</span><span class="p">,</span> <span class="n">GoalsSpace</span>
|
||||
|
||||
|
||||
<span class="c1"># Based on on the description in:</span>
|
||||
<span class="c1"># https://math.stackexchange.com/questions/1287634/implementing-ornstein-uhlenbeck-in-matlab</span>
|
||||
<span class="k">class</span> <span class="nc">OUProcessParameters</span><span class="p">(</span><span class="n">ExplorationParameters</span><span class="p">):</span>
|
||||
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">mu</span> <span class="o">=</span> <span class="mi">0</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">theta</span> <span class="o">=</span> <span class="mf">0.15</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">sigma</span> <span class="o">=</span> <span class="mf">0.2</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">dt</span> <span class="o">=</span> <span class="mf">0.01</span>
|
||||
|
||||
<span class="nd">@property</span>
|
||||
<span class="k">def</span> <span class="nf">path</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="s1">'rl_coach.exploration_policies.ou_process:OUProcess'</span>
|
||||
|
||||
|
||||
<span class="c1"># Ornstein-Uhlenbeck process</span>
|
||||
<div class="viewcode-block" id="OUProcess"><a class="viewcode-back" href="../../../components/exploration_policies/index.html#rl_coach.exploration_policies.OUProcess">[docs]</a><span class="k">class</span> <span class="nc">OUProcess</span><span class="p">(</span><span class="n">ExplorationPolicy</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> OUProcess exploration policy is intended for continuous action spaces, and selects the action according to</span>
|
||||
<span class="sd"> an Ornstein-Uhlenbeck process. The Ornstein-Uhlenbeck process implements the action as a Gaussian process, where</span>
|
||||
<span class="sd"> the samples are correlated between consequent time steps.</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">action_space</span><span class="p">:</span> <span class="n">ActionSpace</span><span class="p">,</span> <span class="n">mu</span><span class="p">:</span> <span class="nb">float</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">theta</span><span class="p">:</span> <span class="nb">float</span><span class="o">=</span><span class="mf">0.15</span><span class="p">,</span> <span class="n">sigma</span><span class="p">:</span> <span class="nb">float</span><span class="o">=</span><span class="mf">0.2</span><span class="p">,</span> <span class="n">dt</span><span class="p">:</span> <span class="nb">float</span><span class="o">=</span><span class="mf">0.01</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> :param action_space: the action space used by the environment</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">action_space</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">mu</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">mu</span><span class="p">)</span> <span class="o">*</span> <span class="n">np</span><span class="o">.</span><span class="n">ones</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">theta</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">theta</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">sigma</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">sigma</span><span class="p">)</span> <span class="o">*</span> <span class="n">np</span><span class="o">.</span><span class="n">ones</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">state</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">dt</span> <span class="o">=</span> <span class="n">dt</span>
|
||||
|
||||
<span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">action_space</span><span class="p">,</span> <span class="n">BoxActionSpace</span><span class="p">)</span> <span class="ow">or</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">action_space</span><span class="p">,</span> <span class="n">GoalsSpace</span><span class="p">)):</span>
|
||||
<span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"OU process exploration works only for continuous controls."</span>
|
||||
<span class="s2">"The given action space is of type: </span><span class="si">{}</span><span class="s2">"</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">action_space</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="p">))</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">reset</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">state</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">noise</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span>
|
||||
<span class="n">dx</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">theta</span> <span class="o">*</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">mu</span> <span class="o">-</span> <span class="n">x</span><span class="p">)</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">dt</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">sigma</span> <span class="o">*</span> <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">randn</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">x</span><span class="p">))</span> <span class="o">*</span> <span class="n">np</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">dt</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">state</span> <span class="o">=</span> <span class="n">x</span> <span class="o">+</span> <span class="n">dx</span>
|
||||
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">get_action</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">action_values</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">ActionType</span><span class="p">])</span> <span class="o">-></span> <span class="n">ActionType</span><span class="p">:</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">phase</span> <span class="o">==</span> <span class="n">RunPhase</span><span class="o">.</span><span class="n">TRAIN</span><span class="p">:</span>
|
||||
<span class="n">noise</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">noise</span><span class="p">()</span>
|
||||
<span class="k">else</span><span class="p">:</span>
|
||||
<span class="n">noise</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span>
|
||||
|
||||
<span class="n">action</span> <span class="o">=</span> <span class="n">action_values</span><span class="o">.</span><span class="n">squeeze</span><span class="p">()</span> <span class="o">+</span> <span class="n">noise</span>
|
||||
|
||||
<span class="k">return</span> <span class="n">action</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">get_control_param</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">phase</span> <span class="o">==</span> <span class="n">RunPhase</span><span class="o">.</span><span class="n">TRAIN</span><span class="p">:</span>
|
||||
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span>
|
||||
<span class="k">else</span><span class="p">:</span>
|
||||
<span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span></div>
|
||||
</pre></div>
|
||||
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<footer>
|
||||
|
||||
|
||||
<hr/>
|
||||
|
||||
<div role="contentinfo">
|
||||
<p>
|
||||
© Copyright 2018, Intel AI Lab
|
||||
|
||||
</p>
|
||||
</div>
|
||||
Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
|
||||
|
||||
</footer>
|
||||
|
||||
</div>
|
||||
</div>
|
||||
|
||||
</section>
|
||||
|
||||
</div>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></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 async="async" type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.1/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script>
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" src="../../../_static/js/theme.js"></script>
|
||||
|
||||
<script type="text/javascript">
|
||||
jQuery(function () {
|
||||
SphinxRtdTheme.Navigation.enable(true);
|
||||
});
|
||||
</script>
|
||||
|
||||
</body>
|
||||
</html>
|
||||
314
docs/_modules/rl_coach/exploration_policies/parameter_noise.html
Normal file
314
docs/_modules/rl_coach/exploration_policies/parameter_noise.html
Normal file
@@ -0,0 +1,314 @@
|
||||
|
||||
|
||||
<!DOCTYPE html>
|
||||
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
|
||||
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
|
||||
<head>
|
||||
<meta charset="utf-8">
|
||||
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||||
|
||||
<title>rl_coach.exploration_policies.parameter_noise — Reinforcement Learning Coach 0.11.0 documentation</title>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/css/custom.css" type="text/css" />
|
||||
<link rel="index" title="Index" href="../../../genindex.html" />
|
||||
<link rel="search" title="Search" href="../../../search.html" />
|
||||
<link href="../../../_static/css/custom.css" rel="stylesheet" type="text/css">
|
||||
|
||||
|
||||
|
||||
<script src="../../../_static/js/modernizr.min.js"></script>
|
||||
|
||||
</head>
|
||||
|
||||
<body class="wy-body-for-nav">
|
||||
|
||||
|
||||
<div class="wy-grid-for-nav">
|
||||
|
||||
|
||||
<nav data-toggle="wy-nav-shift" class="wy-nav-side">
|
||||
<div class="wy-side-scroll">
|
||||
<div class="wy-side-nav-search">
|
||||
|
||||
|
||||
|
||||
<a href="../../../index.html" class="icon icon-home"> Reinforcement Learning Coach
|
||||
|
||||
|
||||
|
||||
|
||||
<img src="../../../_static/dark_logo.png" class="logo" alt="Logo"/>
|
||||
|
||||
</a>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="search">
|
||||
<form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
|
||||
<input type="text" name="q" placeholder="Search docs" />
|
||||
<input type="hidden" name="check_keywords" value="yes" />
|
||||
<input type="hidden" name="area" value="default" />
|
||||
</form>
|
||||
</div>
|
||||
|
||||
|
||||
</div>
|
||||
|
||||
<div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<p class="caption"><span class="caption-text">Intro</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../usage.html">Usage</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../features/index.html">Features</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../selecting_an_algorithm.html">Selecting an Algorithm</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../dashboard.html">Coach Dashboard</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Design</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/control_flow.html">Control Flow</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/network.html">Network Design</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Contributing</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_agent.html">Adding a New Agent</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_env.html">Adding a New Environment</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Components</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/agents/index.html">Agents</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/architectures/index.html">Architectures</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/environments/index.html">Environments</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/exploration_policies/index.html">Exploration Policies</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/filters/index.html">Filters</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/memories/index.html">Memories</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/core_types.html">Core Types</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/spaces.html">Spaces</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/additional_parameters.html">Additional Parameters</a></li>
|
||||
</ul>
|
||||
|
||||
|
||||
|
||||
</div>
|
||||
</div>
|
||||
</nav>
|
||||
|
||||
<section data-toggle="wy-nav-shift" class="wy-nav-content-wrap">
|
||||
|
||||
|
||||
<nav class="wy-nav-top" aria-label="top navigation">
|
||||
|
||||
<i data-toggle="wy-nav-top" class="fa fa-bars"></i>
|
||||
<a href="../../../index.html">Reinforcement Learning Coach</a>
|
||||
|
||||
</nav>
|
||||
|
||||
|
||||
<div class="wy-nav-content">
|
||||
|
||||
<div class="rst-content">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="navigation" aria-label="breadcrumbs navigation">
|
||||
|
||||
<ul class="wy-breadcrumbs">
|
||||
|
||||
<li><a href="../../../index.html">Docs</a> »</li>
|
||||
|
||||
<li><a href="../../index.html">Module code</a> »</li>
|
||||
|
||||
<li>rl_coach.exploration_policies.parameter_noise</li>
|
||||
|
||||
|
||||
<li class="wy-breadcrumbs-aside">
|
||||
|
||||
</li>
|
||||
|
||||
</ul>
|
||||
|
||||
|
||||
<hr/>
|
||||
</div>
|
||||
<div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
|
||||
<div itemprop="articleBody">
|
||||
|
||||
<h1>Source code for rl_coach.exploration_policies.parameter_noise</h1><div class="highlight"><pre>
|
||||
<span></span><span class="c1">#</span>
|
||||
<span class="c1"># Copyright (c) 2017 Intel Corporation</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Licensed under the Apache License, Version 2.0 (the "License");</span>
|
||||
<span class="c1"># you may not use this file except in compliance with the License.</span>
|
||||
<span class="c1"># You may obtain a copy of the License at</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># http://www.apache.org/licenses/LICENSE-2.0</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Unless required by applicable law or agreed to in writing, software</span>
|
||||
<span class="c1"># distributed under the License is distributed on an "AS IS" BASIS,</span>
|
||||
<span class="c1"># WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.</span>
|
||||
<span class="c1"># See the License for the specific language governing permissions and</span>
|
||||
<span class="c1"># limitations under the License.</span>
|
||||
<span class="c1">#</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">typing</span> <span class="k">import</span> <span class="n">List</span><span class="p">,</span> <span class="n">Dict</span>
|
||||
|
||||
<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.agents.dqn_agent</span> <span class="k">import</span> <span class="n">DQNAgentParameters</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.architectures.tensorflow_components.layers</span> <span class="k">import</span> <span class="n">NoisyNetDense</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.base_parameters</span> <span class="k">import</span> <span class="n">AgentParameters</span><span class="p">,</span> <span class="n">NetworkParameters</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.spaces</span> <span class="k">import</span> <span class="n">ActionSpace</span><span class="p">,</span> <span class="n">BoxActionSpace</span><span class="p">,</span> <span class="n">DiscreteActionSpace</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.core_types</span> <span class="k">import</span> <span class="n">ActionType</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.exploration_policies.exploration_policy</span> <span class="k">import</span> <span class="n">ExplorationPolicy</span><span class="p">,</span> <span class="n">ExplorationParameters</span>
|
||||
|
||||
|
||||
<span class="k">class</span> <span class="nc">ParameterNoiseParameters</span><span class="p">(</span><span class="n">ExplorationParameters</span><span class="p">):</span>
|
||||
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">agent_params</span><span class="p">:</span> <span class="n">AgentParameters</span><span class="p">):</span>
|
||||
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
|
||||
<span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">agent_params</span><span class="p">,</span> <span class="n">DQNAgentParameters</span><span class="p">):</span>
|
||||
<span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"Currently only DQN variants are supported for using an exploration type of "</span>
|
||||
<span class="s2">"ParameterNoise."</span><span class="p">)</span>
|
||||
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">network_params</span> <span class="o">=</span> <span class="n">agent_params</span><span class="o">.</span><span class="n">network_wrappers</span>
|
||||
|
||||
<span class="nd">@property</span>
|
||||
<span class="k">def</span> <span class="nf">path</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="s1">'rl_coach.exploration_policies.parameter_noise:ParameterNoise'</span>
|
||||
|
||||
|
||||
<div class="viewcode-block" id="ParameterNoise"><a class="viewcode-back" href="../../../components/exploration_policies/index.html#rl_coach.exploration_policies.ParameterNoise">[docs]</a><span class="k">class</span> <span class="nc">ParameterNoise</span><span class="p">(</span><span class="n">ExplorationPolicy</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> The ParameterNoise exploration policy is intended for both discrete and continuous action spaces.</span>
|
||||
<span class="sd"> It applies the exploration policy by replacing all the dense network layers with noisy layers.</span>
|
||||
<span class="sd"> The noisy layers have both weight means and weight standard deviations, and for each forward pass of the network</span>
|
||||
<span class="sd"> the weights are sampled from a normal distribution that follows the learned weights mean and standard deviation</span>
|
||||
<span class="sd"> values.</span>
|
||||
|
||||
<span class="sd"> Warning: currently supported only by DQN variants</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">network_params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">NetworkParameters</span><span class="p">],</span> <span class="n">action_space</span><span class="p">:</span> <span class="n">ActionSpace</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> :param action_space: the action space used by the environment</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">action_space</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">network_params</span> <span class="o">=</span> <span class="n">network_params</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">_replace_network_dense_layers</span><span class="p">()</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">get_action</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">action_values</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">ActionType</span><span class="p">])</span> <span class="o">-></span> <span class="n">ActionType</span><span class="p">:</span>
|
||||
<span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="p">)</span> <span class="o">==</span> <span class="n">DiscreteActionSpace</span><span class="p">:</span>
|
||||
<span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><span class="n">action_values</span><span class="p">)</span>
|
||||
<span class="k">elif</span> <span class="nb">type</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="p">)</span> <span class="o">==</span> <span class="n">BoxActionSpace</span><span class="p">:</span>
|
||||
<span class="n">action_values_mean</span> <span class="o">=</span> <span class="n">action_values</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">squeeze</span><span class="p">()</span>
|
||||
<span class="n">action_values_std</span> <span class="o">=</span> <span class="n">action_values</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">squeeze</span><span class="p">()</span>
|
||||
<span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">normal</span><span class="p">(</span><span class="n">action_values_mean</span><span class="p">,</span> <span class="n">action_values_std</span><span class="p">)</span>
|
||||
<span class="k">else</span><span class="p">:</span>
|
||||
<span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"ActionSpace type </span><span class="si">{}</span><span class="s2"> is not supported for ParameterNoise."</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="nb">type</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="p">)))</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">get_control_param</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="mi">0</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">_replace_network_dense_layers</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="c1"># replace the dense type for all the networks components (embedders, mw, heads) with a NoisyNetDense</span>
|
||||
|
||||
<span class="c1"># NOTE: we are changing network params in a non-params class (an already instantiated class), this could have</span>
|
||||
<span class="c1"># been prone to a bug, but since the networks are created very late in the game</span>
|
||||
<span class="c1"># (after agent.init_environment_dependent()_modules is called) - then we are fine.</span>
|
||||
|
||||
<span class="k">for</span> <span class="n">network_wrapper_params</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">network_params</span><span class="o">.</span><span class="n">values</span><span class="p">():</span>
|
||||
<span class="k">for</span> <span class="n">component_params</span> <span class="ow">in</span> <span class="nb">list</span><span class="p">(</span><span class="n">network_wrapper_params</span><span class="o">.</span><span class="n">input_embedders_parameters</span><span class="o">.</span><span class="n">values</span><span class="p">())</span> <span class="o">+</span> \
|
||||
<span class="p">[</span><span class="n">network_wrapper_params</span><span class="o">.</span><span class="n">middleware_parameters</span><span class="p">]</span> <span class="o">+</span> \
|
||||
<span class="n">network_wrapper_params</span><span class="o">.</span><span class="n">heads_parameters</span><span class="p">:</span>
|
||||
<span class="n">component_params</span><span class="o">.</span><span class="n">dense_layer</span> <span class="o">=</span> <span class="n">NoisyNetDense</span></div>
|
||||
|
||||
</pre></div>
|
||||
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<footer>
|
||||
|
||||
|
||||
<hr/>
|
||||
|
||||
<div role="contentinfo">
|
||||
<p>
|
||||
© Copyright 2018, Intel AI Lab
|
||||
|
||||
</p>
|
||||
</div>
|
||||
Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
|
||||
|
||||
</footer>
|
||||
|
||||
</div>
|
||||
</div>
|
||||
|
||||
</section>
|
||||
|
||||
</div>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></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 async="async" type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.1/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script>
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" src="../../../_static/js/theme.js"></script>
|
||||
|
||||
<script type="text/javascript">
|
||||
jQuery(function () {
|
||||
SphinxRtdTheme.Navigation.enable(true);
|
||||
});
|
||||
</script>
|
||||
|
||||
</body>
|
||||
</html>
|
||||
@@ -0,0 +1,337 @@
|
||||
|
||||
|
||||
<!DOCTYPE html>
|
||||
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
|
||||
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
|
||||
<head>
|
||||
<meta charset="utf-8">
|
||||
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||||
|
||||
<title>rl_coach.exploration_policies.truncated_normal — Reinforcement Learning Coach 0.11.0 documentation</title>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/css/custom.css" type="text/css" />
|
||||
<link rel="index" title="Index" href="../../../genindex.html" />
|
||||
<link rel="search" title="Search" href="../../../search.html" />
|
||||
<link href="../../../_static/css/custom.css" rel="stylesheet" type="text/css">
|
||||
|
||||
|
||||
|
||||
<script src="../../../_static/js/modernizr.min.js"></script>
|
||||
|
||||
</head>
|
||||
|
||||
<body class="wy-body-for-nav">
|
||||
|
||||
|
||||
<div class="wy-grid-for-nav">
|
||||
|
||||
|
||||
<nav data-toggle="wy-nav-shift" class="wy-nav-side">
|
||||
<div class="wy-side-scroll">
|
||||
<div class="wy-side-nav-search">
|
||||
|
||||
|
||||
|
||||
<a href="../../../index.html" class="icon icon-home"> Reinforcement Learning Coach
|
||||
|
||||
|
||||
|
||||
|
||||
<img src="../../../_static/dark_logo.png" class="logo" alt="Logo"/>
|
||||
|
||||
</a>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="search">
|
||||
<form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
|
||||
<input type="text" name="q" placeholder="Search docs" />
|
||||
<input type="hidden" name="check_keywords" value="yes" />
|
||||
<input type="hidden" name="area" value="default" />
|
||||
</form>
|
||||
</div>
|
||||
|
||||
|
||||
</div>
|
||||
|
||||
<div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<p class="caption"><span class="caption-text">Intro</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../usage.html">Usage</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../features/index.html">Features</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../selecting_an_algorithm.html">Selecting an Algorithm</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../dashboard.html">Coach Dashboard</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Design</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/control_flow.html">Control Flow</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/network.html">Network Design</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Contributing</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_agent.html">Adding a New Agent</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_env.html">Adding a New Environment</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Components</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/agents/index.html">Agents</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/architectures/index.html">Architectures</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/environments/index.html">Environments</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/exploration_policies/index.html">Exploration Policies</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/filters/index.html">Filters</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/memories/index.html">Memories</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/core_types.html">Core Types</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/spaces.html">Spaces</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/additional_parameters.html">Additional Parameters</a></li>
|
||||
</ul>
|
||||
|
||||
|
||||
|
||||
</div>
|
||||
</div>
|
||||
</nav>
|
||||
|
||||
<section data-toggle="wy-nav-shift" class="wy-nav-content-wrap">
|
||||
|
||||
|
||||
<nav class="wy-nav-top" aria-label="top navigation">
|
||||
|
||||
<i data-toggle="wy-nav-top" class="fa fa-bars"></i>
|
||||
<a href="../../../index.html">Reinforcement Learning Coach</a>
|
||||
|
||||
</nav>
|
||||
|
||||
|
||||
<div class="wy-nav-content">
|
||||
|
||||
<div class="rst-content">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="navigation" aria-label="breadcrumbs navigation">
|
||||
|
||||
<ul class="wy-breadcrumbs">
|
||||
|
||||
<li><a href="../../../index.html">Docs</a> »</li>
|
||||
|
||||
<li><a href="../../index.html">Module code</a> »</li>
|
||||
|
||||
<li>rl_coach.exploration_policies.truncated_normal</li>
|
||||
|
||||
|
||||
<li class="wy-breadcrumbs-aside">
|
||||
|
||||
</li>
|
||||
|
||||
</ul>
|
||||
|
||||
|
||||
<hr/>
|
||||
</div>
|
||||
<div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
|
||||
<div itemprop="articleBody">
|
||||
|
||||
<h1>Source code for rl_coach.exploration_policies.truncated_normal</h1><div class="highlight"><pre>
|
||||
<span></span><span class="c1">#</span>
|
||||
<span class="c1"># Copyright (c) 2017 Intel Corporation </span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Licensed under the Apache License, Version 2.0 (the "License");</span>
|
||||
<span class="c1"># you may not use this file except in compliance with the License.</span>
|
||||
<span class="c1"># You may obtain a copy of the License at</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># http://www.apache.org/licenses/LICENSE-2.0</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Unless required by applicable law or agreed to in writing, software</span>
|
||||
<span class="c1"># distributed under the License is distributed on an "AS IS" BASIS,</span>
|
||||
<span class="c1"># WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.</span>
|
||||
<span class="c1"># See the License for the specific language governing permissions and</span>
|
||||
<span class="c1"># limitations under the License.</span>
|
||||
<span class="c1">#</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">typing</span> <span class="k">import</span> <span class="n">List</span>
|
||||
|
||||
<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
|
||||
<span class="kn">from</span> <span class="nn">scipy.stats</span> <span class="k">import</span> <span class="n">truncnorm</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.core_types</span> <span class="k">import</span> <span class="n">RunPhase</span><span class="p">,</span> <span class="n">ActionType</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.exploration_policies.exploration_policy</span> <span class="k">import</span> <span class="n">ExplorationPolicy</span><span class="p">,</span> <span class="n">ExplorationParameters</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.schedules</span> <span class="k">import</span> <span class="n">Schedule</span><span class="p">,</span> <span class="n">LinearSchedule</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.spaces</span> <span class="k">import</span> <span class="n">ActionSpace</span><span class="p">,</span> <span class="n">BoxActionSpace</span>
|
||||
|
||||
|
||||
<span class="k">class</span> <span class="nc">TruncatedNormalParameters</span><span class="p">(</span><span class="n">ExplorationParameters</span><span class="p">):</span>
|
||||
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">noise_percentage_schedule</span> <span class="o">=</span> <span class="n">LinearSchedule</span><span class="p">(</span><span class="mf">0.1</span><span class="p">,</span> <span class="mf">0.1</span><span class="p">,</span> <span class="mi">50000</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">evaluation_noise_percentage</span> <span class="o">=</span> <span class="mf">0.05</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">clip_low</span> <span class="o">=</span> <span class="mi">0</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">clip_high</span> <span class="o">=</span> <span class="mi">1</span>
|
||||
|
||||
<span class="nd">@property</span>
|
||||
<span class="k">def</span> <span class="nf">path</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="s1">'rl_coach.exploration_policies.truncated_normal:TruncatedNormal'</span>
|
||||
|
||||
|
||||
<div class="viewcode-block" id="TruncatedNormal"><a class="viewcode-back" href="../../../components/exploration_policies/index.html#rl_coach.exploration_policies.TruncatedNormal">[docs]</a><span class="k">class</span> <span class="nc">TruncatedNormal</span><span class="p">(</span><span class="n">ExplorationPolicy</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> The TruncatedNormal exploration policy is intended for continuous action spaces. It samples the action from a</span>
|
||||
<span class="sd"> normal distribution, where the mean action is given by the agent, and the standard deviation can be given in t</span>
|
||||
<span class="sd"> wo different ways:</span>
|
||||
<span class="sd"> 1. Specified by the user as a noise schedule which is taken in percentiles out of the action space size</span>
|
||||
<span class="sd"> 2. Specified by the agents action. In case the agents action is a list with 2 values, the 1st one is assumed to</span>
|
||||
<span class="sd"> be the mean of the action, and 2nd is assumed to be its standard deviation.</span>
|
||||
<span class="sd"> When the sampled action is outside of the action bounds given by the user, it is sampled again and again, until it</span>
|
||||
<span class="sd"> is within the bounds.</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">action_space</span><span class="p">:</span> <span class="n">ActionSpace</span><span class="p">,</span> <span class="n">noise_percentage_schedule</span><span class="p">:</span> <span class="n">Schedule</span><span class="p">,</span>
|
||||
<span class="n">evaluation_noise_percentage</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">clip_low</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">clip_high</span><span class="p">:</span> <span class="nb">float</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> :param action_space: the action space used by the environment</span>
|
||||
<span class="sd"> :param noise_percentage_schedule: the schedule for the noise variance percentage relative to the absolute range</span>
|
||||
<span class="sd"> of the action space</span>
|
||||
<span class="sd"> :param evaluation_noise_percentage: the noise variance percentage that will be used during evaluation phases</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">action_space</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">noise_percentage_schedule</span> <span class="o">=</span> <span class="n">noise_percentage_schedule</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">evaluation_noise_percentage</span> <span class="o">=</span> <span class="n">evaluation_noise_percentage</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">clip_low</span> <span class="o">=</span> <span class="n">clip_low</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">clip_high</span> <span class="o">=</span> <span class="n">clip_high</span>
|
||||
|
||||
<span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">action_space</span><span class="p">,</span> <span class="n">BoxActionSpace</span><span class="p">):</span>
|
||||
<span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"Truncated normal exploration works only for continuous controls."</span>
|
||||
<span class="s2">"The given action space is of type: </span><span class="si">{}</span><span class="s2">"</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">action_space</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="p">))</span>
|
||||
|
||||
<span class="k">if</span> <span class="ow">not</span> <span class="n">np</span><span class="o">.</span><span class="n">all</span><span class="p">(</span><span class="o">-</span><span class="n">np</span><span class="o">.</span><span class="n">inf</span> <span class="o"><</span> <span class="n">action_space</span><span class="o">.</span><span class="n">high</span><span class="p">)</span> <span class="ow">or</span> <span class="ow">not</span> <span class="n">np</span><span class="o">.</span><span class="n">all</span><span class="p">(</span><span class="n">action_space</span><span class="o">.</span><span class="n">high</span> <span class="o"><</span> <span class="n">np</span><span class="o">.</span><span class="n">inf</span><span class="p">)</span>\
|
||||
<span class="ow">or</span> <span class="ow">not</span> <span class="n">np</span><span class="o">.</span><span class="n">all</span><span class="p">(</span><span class="o">-</span><span class="n">np</span><span class="o">.</span><span class="n">inf</span> <span class="o"><</span> <span class="n">action_space</span><span class="o">.</span><span class="n">low</span><span class="p">)</span> <span class="ow">or</span> <span class="ow">not</span> <span class="n">np</span><span class="o">.</span><span class="n">all</span><span class="p">(</span><span class="n">action_space</span><span class="o">.</span><span class="n">low</span> <span class="o"><</span> <span class="n">np</span><span class="o">.</span><span class="n">inf</span><span class="p">):</span>
|
||||
<span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"Additive noise exploration requires bounded actions"</span><span class="p">)</span>
|
||||
|
||||
<span class="c1"># TODO: allow working with unbounded actions by defining the noise in terms of range and not percentage</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">get_action</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">action_values</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">ActionType</span><span class="p">])</span> <span class="o">-></span> <span class="n">ActionType</span><span class="p">:</span>
|
||||
<span class="c1"># set the current noise percentage</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">phase</span> <span class="o">==</span> <span class="n">RunPhase</span><span class="o">.</span><span class="n">TEST</span><span class="p">:</span>
|
||||
<span class="n">current_noise_precentage</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">evaluation_noise_percentage</span>
|
||||
<span class="k">else</span><span class="p">:</span>
|
||||
<span class="n">current_noise_precentage</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">noise_percentage_schedule</span><span class="o">.</span><span class="n">current_value</span>
|
||||
|
||||
<span class="c1"># scale the noise to the action space range</span>
|
||||
<span class="n">action_values_std</span> <span class="o">=</span> <span class="n">current_noise_precentage</span> <span class="o">*</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="o">.</span><span class="n">high</span> <span class="o">-</span> <span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="o">.</span><span class="n">low</span><span class="p">)</span>
|
||||
|
||||
<span class="c1"># extract the mean values</span>
|
||||
<span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">action_values</span><span class="p">,</span> <span class="nb">list</span><span class="p">):</span>
|
||||
<span class="c1"># the action values are expected to be a list with the action mean and optionally the action stdev</span>
|
||||
<span class="n">action_values_mean</span> <span class="o">=</span> <span class="n">action_values</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">squeeze</span><span class="p">()</span>
|
||||
<span class="k">else</span><span class="p">:</span>
|
||||
<span class="c1"># the action values are expected to be a numpy array representing the action mean</span>
|
||||
<span class="n">action_values_mean</span> <span class="o">=</span> <span class="n">action_values</span><span class="o">.</span><span class="n">squeeze</span><span class="p">()</span>
|
||||
|
||||
<span class="c1"># step the noise schedule</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">phase</span> <span class="o">==</span> <span class="n">RunPhase</span><span class="o">.</span><span class="n">TRAIN</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">noise_percentage_schedule</span><span class="o">.</span><span class="n">step</span><span class="p">()</span>
|
||||
<span class="c1"># the second element of the list is assumed to be the standard deviation</span>
|
||||
<span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">action_values</span><span class="p">,</span> <span class="nb">list</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">action_values</span><span class="p">)</span> <span class="o">></span> <span class="mi">1</span><span class="p">:</span>
|
||||
<span class="n">action_values_std</span> <span class="o">=</span> <span class="n">action_values</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">squeeze</span><span class="p">()</span>
|
||||
|
||||
<span class="c1"># sample from truncated normal distribution</span>
|
||||
<span class="n">normalized_low</span> <span class="o">=</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">clip_low</span> <span class="o">-</span> <span class="n">action_values_mean</span><span class="p">)</span> <span class="o">/</span> <span class="n">action_values_std</span>
|
||||
<span class="n">normalized_high</span> <span class="o">=</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">clip_high</span> <span class="o">-</span> <span class="n">action_values_mean</span><span class="p">)</span> <span class="o">/</span> <span class="n">action_values_std</span>
|
||||
<span class="n">distribution</span> <span class="o">=</span> <span class="n">truncnorm</span><span class="p">(</span><span class="n">normalized_low</span><span class="p">,</span> <span class="n">normalized_high</span><span class="p">,</span> <span class="n">loc</span><span class="o">=</span><span class="n">action_values_mean</span><span class="p">,</span> <span class="n">scale</span><span class="o">=</span><span class="n">action_values_std</span><span class="p">)</span>
|
||||
<span class="n">action</span> <span class="o">=</span> <span class="n">distribution</span><span class="o">.</span><span class="n">rvs</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
|
||||
|
||||
<span class="k">return</span> <span class="n">action</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">get_control_param</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">ones</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">action_space</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span><span class="o">*</span><span class="bp">self</span><span class="o">.</span><span class="n">noise_percentage_schedule</span><span class="o">.</span><span class="n">current_value</span></div>
|
||||
</pre></div>
|
||||
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<footer>
|
||||
|
||||
|
||||
<hr/>
|
||||
|
||||
<div role="contentinfo">
|
||||
<p>
|
||||
© Copyright 2018, Intel AI Lab
|
||||
|
||||
</p>
|
||||
</div>
|
||||
Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
|
||||
|
||||
</footer>
|
||||
|
||||
</div>
|
||||
</div>
|
||||
|
||||
</section>
|
||||
|
||||
</div>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></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 async="async" type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.1/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script>
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" src="../../../_static/js/theme.js"></script>
|
||||
|
||||
<script type="text/javascript">
|
||||
jQuery(function () {
|
||||
SphinxRtdTheme.Navigation.enable(true);
|
||||
});
|
||||
</script>
|
||||
|
||||
</body>
|
||||
</html>
|
||||
319
docs/_modules/rl_coach/exploration_policies/ucb.html
Normal file
319
docs/_modules/rl_coach/exploration_policies/ucb.html
Normal file
@@ -0,0 +1,319 @@
|
||||
|
||||
|
||||
<!DOCTYPE html>
|
||||
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
|
||||
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
|
||||
<head>
|
||||
<meta charset="utf-8">
|
||||
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||||
|
||||
<title>rl_coach.exploration_policies.ucb — Reinforcement Learning Coach 0.11.0 documentation</title>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
|
||||
<link rel="stylesheet" href="../../../_static/css/custom.css" type="text/css" />
|
||||
<link rel="index" title="Index" href="../../../genindex.html" />
|
||||
<link rel="search" title="Search" href="../../../search.html" />
|
||||
<link href="../../../_static/css/custom.css" rel="stylesheet" type="text/css">
|
||||
|
||||
|
||||
|
||||
<script src="../../../_static/js/modernizr.min.js"></script>
|
||||
|
||||
</head>
|
||||
|
||||
<body class="wy-body-for-nav">
|
||||
|
||||
|
||||
<div class="wy-grid-for-nav">
|
||||
|
||||
|
||||
<nav data-toggle="wy-nav-shift" class="wy-nav-side">
|
||||
<div class="wy-side-scroll">
|
||||
<div class="wy-side-nav-search">
|
||||
|
||||
|
||||
|
||||
<a href="../../../index.html" class="icon icon-home"> Reinforcement Learning Coach
|
||||
|
||||
|
||||
|
||||
|
||||
<img src="../../../_static/dark_logo.png" class="logo" alt="Logo"/>
|
||||
|
||||
</a>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="search">
|
||||
<form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
|
||||
<input type="text" name="q" placeholder="Search docs" />
|
||||
<input type="hidden" name="check_keywords" value="yes" />
|
||||
<input type="hidden" name="area" value="default" />
|
||||
</form>
|
||||
</div>
|
||||
|
||||
|
||||
</div>
|
||||
|
||||
<div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<p class="caption"><span class="caption-text">Intro</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../usage.html">Usage</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../features/index.html">Features</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../selecting_an_algorithm.html">Selecting an Algorithm</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../dashboard.html">Coach Dashboard</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Design</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/control_flow.html">Control Flow</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../design/network.html">Network Design</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Contributing</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_agent.html">Adding a New Agent</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_env.html">Adding a New Environment</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Components</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/agents/index.html">Agents</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/architectures/index.html">Architectures</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/environments/index.html">Environments</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/exploration_policies/index.html">Exploration Policies</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/filters/index.html">Filters</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/memories/index.html">Memories</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/core_types.html">Core Types</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/spaces.html">Spaces</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../../components/additional_parameters.html">Additional Parameters</a></li>
|
||||
</ul>
|
||||
|
||||
|
||||
|
||||
</div>
|
||||
</div>
|
||||
</nav>
|
||||
|
||||
<section data-toggle="wy-nav-shift" class="wy-nav-content-wrap">
|
||||
|
||||
|
||||
<nav class="wy-nav-top" aria-label="top navigation">
|
||||
|
||||
<i data-toggle="wy-nav-top" class="fa fa-bars"></i>
|
||||
<a href="../../../index.html">Reinforcement Learning Coach</a>
|
||||
|
||||
</nav>
|
||||
|
||||
|
||||
<div class="wy-nav-content">
|
||||
|
||||
<div class="rst-content">
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<div role="navigation" aria-label="breadcrumbs navigation">
|
||||
|
||||
<ul class="wy-breadcrumbs">
|
||||
|
||||
<li><a href="../../../index.html">Docs</a> »</li>
|
||||
|
||||
<li><a href="../../index.html">Module code</a> »</li>
|
||||
|
||||
<li>rl_coach.exploration_policies.ucb</li>
|
||||
|
||||
|
||||
<li class="wy-breadcrumbs-aside">
|
||||
|
||||
</li>
|
||||
|
||||
</ul>
|
||||
|
||||
|
||||
<hr/>
|
||||
</div>
|
||||
<div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
|
||||
<div itemprop="articleBody">
|
||||
|
||||
<h1>Source code for rl_coach.exploration_policies.ucb</h1><div class="highlight"><pre>
|
||||
<span></span><span class="c1">#</span>
|
||||
<span class="c1"># Copyright (c) 2017 Intel Corporation </span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Licensed under the Apache License, Version 2.0 (the "License");</span>
|
||||
<span class="c1"># you may not use this file except in compliance with the License.</span>
|
||||
<span class="c1"># You may obtain a copy of the License at</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># http://www.apache.org/licenses/LICENSE-2.0</span>
|
||||
<span class="c1">#</span>
|
||||
<span class="c1"># Unless required by applicable law or agreed to in writing, software</span>
|
||||
<span class="c1"># distributed under the License is distributed on an "AS IS" BASIS,</span>
|
||||
<span class="c1"># WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.</span>
|
||||
<span class="c1"># See the License for the specific language governing permissions and</span>
|
||||
<span class="c1"># limitations under the License.</span>
|
||||
<span class="c1">#</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">typing</span> <span class="k">import</span> <span class="n">List</span>
|
||||
|
||||
<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.core_types</span> <span class="k">import</span> <span class="n">RunPhase</span><span class="p">,</span> <span class="n">ActionType</span><span class="p">,</span> <span class="n">EnvironmentSteps</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.exploration_policies.additive_noise</span> <span class="k">import</span> <span class="n">AdditiveNoiseParameters</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.exploration_policies.e_greedy</span> <span class="k">import</span> <span class="n">EGreedy</span><span class="p">,</span> <span class="n">EGreedyParameters</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.exploration_policies.exploration_policy</span> <span class="k">import</span> <span class="n">ExplorationParameters</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.schedules</span> <span class="k">import</span> <span class="n">Schedule</span><span class="p">,</span> <span class="n">LinearSchedule</span><span class="p">,</span> <span class="n">PieceWiseSchedule</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.spaces</span> <span class="k">import</span> <span class="n">ActionSpace</span>
|
||||
|
||||
|
||||
<span class="k">class</span> <span class="nc">UCBParameters</span><span class="p">(</span><span class="n">EGreedyParameters</span><span class="p">):</span>
|
||||
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">architecture_num_q_heads</span> <span class="o">=</span> <span class="mi">10</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">bootstrapped_data_sharing_probability</span> <span class="o">=</span> <span class="mf">1.0</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">epsilon_schedule</span> <span class="o">=</span> <span class="n">PieceWiseSchedule</span><span class="p">([</span>
|
||||
<span class="p">(</span><span class="n">LinearSchedule</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mf">0.1</span><span class="p">,</span> <span class="mi">1000000</span><span class="p">),</span> <span class="n">EnvironmentSteps</span><span class="p">(</span><span class="mi">1000000</span><span class="p">)),</span>
|
||||
<span class="p">(</span><span class="n">LinearSchedule</span><span class="p">(</span><span class="mf">0.1</span><span class="p">,</span> <span class="mf">0.01</span><span class="p">,</span> <span class="mi">4000000</span><span class="p">),</span> <span class="n">EnvironmentSteps</span><span class="p">(</span><span class="mi">4000000</span><span class="p">))</span>
|
||||
<span class="p">])</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">lamb</span> <span class="o">=</span> <span class="mf">0.1</span>
|
||||
|
||||
<span class="nd">@property</span>
|
||||
<span class="k">def</span> <span class="nf">path</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="s1">'rl_coach.exploration_policies.ucb:UCB'</span>
|
||||
|
||||
|
||||
<div class="viewcode-block" id="UCB"><a class="viewcode-back" href="../../../components/exploration_policies/index.html#rl_coach.exploration_policies.UCB">[docs]</a><span class="k">class</span> <span class="nc">UCB</span><span class="p">(</span><span class="n">EGreedy</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> UCB exploration policy is following the upper confidence bound heuristic to sample actions in discrete action spaces.</span>
|
||||
<span class="sd"> It assumes that there are multiple network heads that are predicting action values, and that the standard deviation</span>
|
||||
<span class="sd"> between the heads predictions represents the uncertainty of the agent in each of the actions.</span>
|
||||
<span class="sd"> It then updates the action value estimates to by mean(actions)+lambda*stdev(actions), where lambda is</span>
|
||||
<span class="sd"> given by the user. This exploration policy aims to take advantage of the uncertainty of the agent in its predictions,</span>
|
||||
<span class="sd"> and select the action according to the tradeoff between how uncertain the agent is, and how large it predicts</span>
|
||||
<span class="sd"> the outcome from those actions to be.</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">action_space</span><span class="p">:</span> <span class="n">ActionSpace</span><span class="p">,</span> <span class="n">epsilon_schedule</span><span class="p">:</span> <span class="n">Schedule</span><span class="p">,</span> <span class="n">evaluation_epsilon</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
|
||||
<span class="n">architecture_num_q_heads</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">lamb</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
|
||||
<span class="n">continuous_exploration_policy_parameters</span><span class="p">:</span> <span class="n">ExplorationParameters</span> <span class="o">=</span> <span class="n">AdditiveNoiseParameters</span><span class="p">()):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> :param action_space: the action space used by the environment</span>
|
||||
<span class="sd"> :param epsilon_schedule: a schedule for the epsilon values</span>
|
||||
<span class="sd"> :param evaluation_epsilon: the epsilon value to use for evaluation phases</span>
|
||||
<span class="sd"> :param architecture_num_q_heads: the number of q heads to select from</span>
|
||||
<span class="sd"> :param lamb: lambda coefficient for taking the standard deviation into account</span>
|
||||
<span class="sd"> :param continuous_exploration_policy_parameters: the parameters of the continuous exploration policy to use</span>
|
||||
<span class="sd"> if the e-greedy is used for a continuous policy</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">action_space</span><span class="p">,</span> <span class="n">epsilon_schedule</span><span class="p">,</span> <span class="n">evaluation_epsilon</span><span class="p">,</span> <span class="n">continuous_exploration_policy_parameters</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span> <span class="o">=</span> <span class="n">architecture_num_q_heads</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">lamb</span> <span class="o">=</span> <span class="n">lamb</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">std</span> <span class="o">=</span> <span class="mi">0</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">last_action_values</span> <span class="o">=</span> <span class="mi">0</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">select_head</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">pass</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">get_action</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">action_values</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">ActionType</span><span class="p">])</span> <span class="o">-></span> <span class="n">ActionType</span><span class="p">:</span>
|
||||
<span class="c1"># action values are none in case the exploration policy is going to select a random action</span>
|
||||
<span class="k">if</span> <span class="n">action_values</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">requires_action_values</span><span class="p">():</span>
|
||||
<span class="n">mean</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="n">action_values</span><span class="p">,</span> <span class="n">axis</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">phase</span> <span class="o">==</span> <span class="n">RunPhase</span><span class="o">.</span><span class="n">TRAIN</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">std</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">std</span><span class="p">(</span><span class="n">action_values</span><span class="p">,</span> <span class="n">axis</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">last_action_values</span> <span class="o">=</span> <span class="n">mean</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">lamb</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">std</span>
|
||||
<span class="k">else</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">last_action_values</span> <span class="o">=</span> <span class="n">mean</span>
|
||||
<span class="k">return</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">get_action</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">last_action_values</span><span class="p">)</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">get_control_param</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">phase</span> <span class="o">==</span> <span class="n">RunPhase</span><span class="o">.</span><span class="n">TRAIN</span><span class="p">:</span>
|
||||
<span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">std</span><span class="p">)</span>
|
||||
<span class="k">else</span><span class="p">:</span>
|
||||
<span class="k">return</span> <span class="mi">0</span></div>
|
||||
</pre></div>
|
||||
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<footer>
|
||||
|
||||
|
||||
<hr/>
|
||||
|
||||
<div role="contentinfo">
|
||||
<p>
|
||||
© Copyright 2018, Intel AI Lab
|
||||
|
||||
</p>
|
||||
</div>
|
||||
Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
|
||||
|
||||
</footer>
|
||||
|
||||
</div>
|
||||
</div>
|
||||
|
||||
</section>
|
||||
|
||||
</div>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></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 async="async" type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.1/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script>
|
||||
|
||||
|
||||
|
||||
|
||||
<script type="text/javascript" src="../../../_static/js/theme.js"></script>
|
||||
|
||||
<script type="text/javascript">
|
||||
jQuery(function () {
|
||||
SphinxRtdTheme.Navigation.enable(true);
|
||||
});
|
||||
</script>
|
||||
|
||||
</body>
|
||||
</html>
|
||||
Reference in New Issue
Block a user