<!DOCTYPE html>
<html class="writer-html5" lang="en" >
<head>
<meta charset="utf-8" /><meta name="generator" content="Docutils 0.19: https://docutils.sourceforge.io/" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<title>Movement Primitives Examples — Fancy Gym 0.2 documentation</title>
<link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
<link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
<link rel="stylesheet" href="../_static/style.css" type="text/css" />
<link rel="shortcut icon" href="../_static/icon.svg"/>
<!--[if lt IE 9]>
<script src="../_static/js/html5shiv.min.js"></script>
<![endif]-->
<script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
<script src="../_static/jquery.js"></script>
<script src="../_static/underscore.js"></script>
<script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
<script src="../_static/doctools.js"></script>
<script src="../_static/sphinx_highlight.js"></script>
<script src="../_static/js/theme.js"></script>
<link rel="index" title="Index" href="../genindex.html" />
<link rel="search" title="Search" href="../search.html" />
<link rel="next" title="MP Params Tuning Example" href="mp_params_tuning.html" />
<link rel="prev" title="OpenAI Envs Examples" href="open_ai.html" />
</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">
Fancy Gym
<img src="../_static/icon.svg" class="logo" alt="Logo"/>
</a>
<div class="version">
0.2
</div>
<div role="search">
<form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
<input type="text" name="q" placeholder="Search docs" aria-label="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="Navigation menu">
<p class="caption" role="heading"><span class="caption-text">User Guide</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../guide/installation.html">Installation</a></li>
<li class="toctree-l1"><a class="reference internal" href="../guide/episodic_rl.html">What is Episodic RL?</a></li>
<li class="toctree-l1"><a class="reference internal" href="../guide/basic_usage.html">Basic Usage</a></li>
<li class="toctree-l1"><a class="reference internal" href="../guide/upgrading_envs.html">Creating new MP Environments</a></li>
</ul>
<p class="caption" role="heading"><span class="caption-text">Environments</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../envs/fancy/index.html">Fancy</a></li>
<li class="toctree-l1"><a class="reference internal" href="../envs/dmc.html">DeepMind Control (DMC)</a></li>
<li class="toctree-l1"><a class="reference internal" href="../envs/meta.html">Metaworld</a></li>
<li class="toctree-l1"><a class="reference internal" href="../envs/open_ai.html">Gymnasium</a></li>
</ul>
<p class="caption" role="heading"><span class="caption-text">Examples</span></p>
<ul class="current">
<li class="toctree-l1"><a class="reference internal" href="general.html">General Usage Examples</a></li>
<li class="toctree-l1"><a class="reference internal" href="dmc.html">DeepMind Control Examples</a></li>
<li class="toctree-l1"><a class="reference internal" href="metaworld.html">Metaworld Examples</a></li>
<li class="toctree-l1"><a class="reference internal" href="open_ai.html">OpenAI Envs Examples</a></li>
<li class="toctree-l1 current"><a class="current reference internal" href="#">Movement Primitives Examples</a></li>
<li class="toctree-l1"><a class="reference internal" href="mp_params_tuning.html">MP Params Tuning Example</a></li>
<li class="toctree-l1"><a class="reference internal" href="pd_control_gain_tuning.html">PD Control Gain Tuning Example</a></li>
<li class="toctree-l1"><a class="reference internal" href="replanning_envs.html">Replanning Example</a></li>
</ul>
<p class="caption" role="heading"><span class="caption-text">API</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../api.html">API</a></li>
</ul>
</div>
</div>
</nav>
<section data-toggle="wy-nav-shift" class="wy-nav-content-wrap"><nav class="wy-nav-top" aria-label="Mobile navigation menu" >
<i data-toggle="wy-nav-top" class="fa fa-bars"></i>
<a href="../index.html">Fancy Gym</a>
</nav>
<div class="wy-nav-content">
<div class="rst-content">
<div role="navigation" aria-label="Page navigation">
<ul class="wy-breadcrumbs">
<li><a href="../index.html" class="icon icon-home" aria-label="Home"></a></li>
<li class="breadcrumb-item active">Movement Primitives Examples</li>
<li class="wy-breadcrumbs-aside">
<a href="https://github.com/ALRhub/fancy_gym/blob/release/docs/source/examples/movement_primitives.rst" class="fa fa-github"> Edit on GitHub</a>
</li>
</ul>
<hr/>
</div>
<div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
<div itemprop="articleBody">
<section id="movement-primitives-examples">
<span id="example-mp"></span><h1>Movement Primitives Examples<a class="headerlink" href="#movement-primitives-examples" title="Permalink to this heading"></a></h1>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="linenos"> 1</span><span class="kn">import</span> <span class="nn">gymnasium</span> <span class="k">as</span> <span class="nn">gym</span>
<span class="linenos"> 2</span><span class="kn">import</span> <span class="nn">fancy_gym</span>
<span class="linenos"> 3</span>
<span class="linenos"> 4</span>
<span class="linenos"> 5</span><span class="k">def</span> <span class="nf">example_mp</span><span class="p">(</span><span class="n">env_name</span><span class="o">=</span><span class="s2">"fancy_ProMP/HoleReacher-v0"</span><span class="p">,</span> <span class="n">seed</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">iterations</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">render</span><span class="o">=</span><span class="kc">True</span><span class="p">):</span>
<span class="linenos"> 6</span><span class="w"> </span><span class="sd">"""</span>
<span class="linenos"> 7</span><span class="sd"> Example for running a black box based environment, which is already registered</span>
<span class="linenos"> 8</span><span class="sd"> Args:</span>
<span class="linenos"> 9</span><span class="sd"> env_name: Black box env_id</span>
<span class="linenos"> 10</span><span class="sd"> seed: seed for deterministic behaviour</span>
<span class="linenos"> 11</span><span class="sd"> iterations: Number of rollout steps to run</span>
<span class="linenos"> 12</span><span class="sd"> render: Render the episode</span>
<span class="linenos"> 13</span>
<span class="linenos"> 14</span><span class="sd"> Returns:</span>
<span class="linenos"> 15</span>
<span class="linenos"> 16</span><span class="sd"> """</span>
<span class="linenos"> 17</span> <span class="c1"># Equivalent to gym, we have a make function which can be used to create environments.</span>
<span class="linenos"> 18</span> <span class="c1"># It takes care of seeding and enables the use of a variety of external environments using the gym interface.</span>
<span class="linenos"> 19</span> <span class="n">env</span> <span class="o">=</span> <span class="n">gym</span><span class="o">.</span><span class="n">make</span><span class="p">(</span><span class="n">env_name</span><span class="p">,</span> <span class="n">render_mode</span><span class="o">=</span><span class="s1">'human'</span> <span class="k">if</span> <span class="n">render</span> <span class="k">else</span> <span class="kc">None</span><span class="p">)</span>
<span class="linenos"> 20</span>
<span class="linenos"> 21</span> <span class="n">returns</span> <span class="o">=</span> <span class="mi">0</span>
<span class="linenos"> 22</span> <span class="c1"># env.render(mode=None)</span>
<span class="linenos"> 23</span> <span class="n">obs</span> <span class="o">=</span> <span class="n">env</span><span class="o">.</span><span class="n">reset</span><span class="p">(</span><span class="n">seed</span><span class="o">=</span><span class="n">seed</span><span class="p">)</span>
<span class="linenos"> 24</span>
<span class="linenos"> 25</span> <span class="c1"># number of samples/full trajectories (multiple environment steps)</span>
<span class="linenos"> 26</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">iterations</span><span class="p">):</span>
<span class="linenos"> 27</span>
<span class="linenos"> 28</span> <span class="k">if</span> <span class="n">render</span> <span class="ow">and</span> <span class="n">i</span> <span class="o">%</span> <span class="mi">1</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
<span class="linenos"> 29</span> <span class="n">env</span><span class="o">.</span><span class="n">render</span><span class="p">()</span>
<span class="linenos"> 30</span>
<span class="linenos"> 31</span> <span class="c1"># Now the action space is not the raw action but the parametrization of the trajectory generator,</span>
<span class="linenos"> 32</span> <span class="c1"># such as a ProMP</span>
<span class="linenos"> 33</span> <span class="n">ac</span> <span class="o">=</span> <span class="n">env</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="linenos"> 34</span> <span class="c1"># This executes a full trajectory and gives back the context (obs) of the last step in the trajectory, or the</span>
<span class="linenos"> 35</span> <span class="c1"># full observation space of the last step, if replanning/sub-trajectory learning is used. The 'reward' is equal</span>
<span class="linenos"> 36</span> <span class="c1"># to the return of a trajectory. Default is the sum over the step-wise rewards.</span>
<span class="linenos"> 37</span> <span class="n">obs</span><span class="p">,</span> <span class="n">reward</span><span class="p">,</span> <span class="n">terminated</span><span class="p">,</span> <span class="n">truncated</span><span class="p">,</span> <span class="n">info</span> <span class="o">=</span> <span class="n">env</span><span class="o">.</span><span class="n">step</span><span class="p">(</span><span class="n">ac</span><span class="p">)</span>
<span class="linenos"> 38</span> <span class="c1"># Aggregated returns</span>
<span class="linenos"> 39</span> <span class="n">returns</span> <span class="o">+=</span> <span class="n">reward</span>
<span class="linenos"> 40</span>
<span class="linenos"> 41</span> <span class="k">if</span> <span class="n">terminated</span> <span class="ow">or</span> <span class="n">truncated</span><span class="p">:</span>
<span class="linenos"> 42</span> <span class="nb">print</span><span class="p">(</span><span class="n">reward</span><span class="p">)</span>
<span class="linenos"> 43</span> <span class="n">obs</span> <span class="o">=</span> <span class="n">env</span><span class="o">.</span><span class="n">reset</span><span class="p">()</span>
<span class="linenos"> 44</span> <span class="n">env</span><span class="o">.</span><span class="n">close</span><span class="p">()</span>
<span class="linenos"> 45</span>
<span class="linenos"> 46</span>
<span class="linenos"> 47</span><span class="k">def</span> <span class="nf">example_custom_mp</span><span class="p">(</span><span class="n">env_name</span><span class="o">=</span><span class="s2">"fancy_ProMP/Reacher5d-v0"</span><span class="p">,</span> <span class="n">seed</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">iterations</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">render</span><span class="o">=</span><span class="kc">True</span><span class="p">):</span>
<span class="linenos"> 48</span><span class="w"> </span><span class="sd">"""</span>
<span class="linenos"> 49</span><span class="sd"> Example for running a custom movement primitive based environments.</span>
<span class="linenos"> 50</span><span class="sd"> Our already registered environments follow the same structure.</span>
<span class="linenos"> 51</span><span class="sd"> Hence, this also allows to adjust hyperparameters of the movement primitives.</span>
<span class="linenos"> 52</span><span class="sd"> Yet, we recommend the method above if you are just interested in changing those parameters for existing tasks.</span>
<span class="linenos"> 53</span><span class="sd"> We appreciate PRs for custom environments (especially MP wrappers of existing tasks) </span>
<span class="linenos"> 54</span><span class="sd"> for our repo: https://github.com/ALRhub/fancy_gym/</span>
<span class="linenos"> 55</span><span class="sd"> Args:</span>
<span class="linenos"> 56</span><span class="sd"> seed: seed</span>
<span class="linenos"> 57</span><span class="sd"> iterations: Number of rollout steps to run</span>
<span class="linenos"> 58</span><span class="sd"> render: Render the episode</span>
<span class="linenos"> 59</span>
<span class="linenos"> 60</span><span class="sd"> Returns:</span>
<span class="linenos"> 61</span>
<span class="linenos"> 62</span><span class="sd"> """</span>
<span class="linenos"> 63</span> <span class="c1"># Changing the arguments of the black box env is possible by providing them to gym through mp_config_override.</span>
<span class="linenos"> 64</span> <span class="c1"># E.g. here for way to many basis functions</span>
<span class="linenos"> 65</span> <span class="n">env</span> <span class="o">=</span> <span class="n">gym</span><span class="o">.</span><span class="n">make</span><span class="p">(</span><span class="n">env_name</span><span class="p">,</span> <span class="n">seed</span><span class="p">,</span> <span class="n">mp_config_override</span><span class="o">=</span><span class="p">{</span><span class="s1">'basis_generator_kwargs'</span><span class="p">:</span> <span class="p">{</span><span class="s1">'num_basis'</span><span class="p">:</span> <span class="mi">1000</span><span class="p">}},</span> <span class="n">render_mode</span><span class="o">=</span><span class="s1">'human'</span> <span class="k">if</span> <span class="n">render</span> <span class="k">else</span> <span class="kc">None</span><span class="p">)</span>
<span class="linenos"> 66</span>
<span class="linenos"> 67</span> <span class="n">returns</span> <span class="o">=</span> <span class="mi">0</span>
<span class="linenos"> 68</span> <span class="n">obs</span> <span class="o">=</span> <span class="n">env</span><span class="o">.</span><span class="n">reset</span><span class="p">()</span>
<span class="linenos"> 69</span>
<span class="linenos"> 70</span> <span class="c1"># This time rendering every trajectory</span>
<span class="linenos"> 71</span> <span class="k">if</span> <span class="n">render</span><span class="p">:</span>
<span class="linenos"> 72</span> <span class="n">env</span><span class="o">.</span><span class="n">render</span><span class="p">()</span>
<span class="linenos"> 73</span>
<span class="linenos"> 74</span> <span class="c1"># number of samples/full trajectories (multiple environment steps)</span>
<span class="linenos"> 75</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">iterations</span><span class="p">):</span>
<span class="linenos"> 76</span> <span class="n">ac</span> <span class="o">=</span> <span class="n">env</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="linenos"> 77</span> <span class="n">obs</span><span class="p">,</span> <span class="n">reward</span><span class="p">,</span> <span class="n">terminated</span><span class="p">,</span> <span class="n">truncated</span><span class="p">,</span> <span class="n">info</span> <span class="o">=</span> <span class="n">env</span><span class="o">.</span><span class="n">step</span><span class="p">(</span><span class="n">ac</span><span class="p">)</span>
<span class="linenos"> 78</span> <span class="n">returns</span> <span class="o">+=</span> <span class="n">reward</span>
<span class="linenos"> 79</span>
<span class="linenos"> 80</span> <span class="k">if</span> <span class="n">terminated</span> <span class="ow">or</span> <span class="n">truncated</span><span class="p">:</span>
<span class="linenos"> 81</span> <span class="nb">print</span><span class="p">(</span><span class="n">i</span><span class="p">,</span> <span class="n">reward</span><span class="p">)</span>
<span class="linenos"> 82</span> <span class="n">obs</span> <span class="o">=</span> <span class="n">env</span><span class="o">.</span><span class="n">reset</span><span class="p">()</span>
<span class="linenos"> 83</span>
<span class="linenos"> 84</span> <span class="n">env</span><span class="o">.</span><span class="n">close</span><span class="p">()</span>
<span class="linenos"> 85</span> <span class="k">return</span> <span class="n">obs</span>
<span class="linenos"> 86</span>
<span class="linenos"> 87</span><span class="k">class</span> <span class="nc">Custom_MPWrapper</span><span class="p">(</span><span class="n">fancy_gym</span><span class="o">.</span><span class="n">envs</span><span class="o">.</span><span class="n">mujoco</span><span class="o">.</span><span class="n">reacher</span><span class="o">.</span><span class="n">MPWrapper</span><span class="p">):</span>
<span class="linenos"> 88</span> <span class="n">mp_config</span> <span class="o">=</span> <span class="p">{</span>
<span class="linenos"> 89</span> <span class="s1">'ProMP'</span><span class="p">:</span> <span class="p">{</span>
<span class="linenos"> 90</span> <span class="s1">'trajectory_generator_kwargs'</span><span class="p">:</span> <span class="p">{</span>
<span class="linenos"> 91</span> <span class="s1">'trajectory_generator_type'</span><span class="p">:</span> <span class="s1">'promp'</span><span class="p">,</span>
<span class="linenos"> 92</span> <span class="s1">'weights_scale'</span><span class="p">:</span> <span class="mi">2</span>
<span class="linenos"> 93</span> <span class="p">},</span>
<span class="linenos"> 94</span> <span class="s1">'phase_generator_kwargs'</span><span class="p">:</span> <span class="p">{</span>
<span class="linenos"> 95</span> <span class="s1">'phase_generator_type'</span><span class="p">:</span> <span class="s1">'linear'</span>
<span class="linenos"> 96</span> <span class="p">},</span>
<span class="linenos"> 97</span> <span class="s1">'controller_kwargs'</span><span class="p">:</span> <span class="p">{</span>
<span class="linenos"> 98</span> <span class="s1">'controller_type'</span><span class="p">:</span> <span class="s1">'velocity'</span>
<span class="linenos"> 99</span> <span class="p">},</span>
<span class="linenos">100</span> <span class="s1">'basis_generator_kwargs'</span><span class="p">:</span> <span class="p">{</span>
<span class="linenos">101</span> <span class="s1">'basis_generator_type'</span><span class="p">:</span> <span class="s1">'zero_rbf'</span><span class="p">,</span>
<span class="linenos">102</span> <span class="s1">'num_basis'</span><span class="p">:</span> <span class="mi">5</span><span class="p">,</span>
<span class="linenos">103</span> <span class="s1">'num_basis_zero_start'</span><span class="p">:</span> <span class="mi">1</span>
<span class="linenos">104</span> <span class="p">}</span>
<span class="linenos">105</span> <span class="p">},</span>
<span class="linenos">106</span> <span class="s1">'DMP'</span><span class="p">:</span> <span class="p">{</span>
<span class="linenos">107</span> <span class="s1">'trajectory_generator_kwargs'</span><span class="p">:</span> <span class="p">{</span>
<span class="linenos">108</span> <span class="s1">'trajectory_generator_type'</span><span class="p">:</span> <span class="s1">'dmp'</span><span class="p">,</span>
<span class="linenos">109</span> <span class="s1">'weights_scale'</span><span class="p">:</span> <span class="mi">500</span>
<span class="linenos">110</span> <span class="p">},</span>
<span class="linenos">111</span> <span class="s1">'phase_generator_kwargs'</span><span class="p">:</span> <span class="p">{</span>
<span class="linenos">112</span> <span class="s1">'phase_generator_type'</span><span class="p">:</span> <span class="s1">'exp'</span><span class="p">,</span>
<span class="linenos">113</span> <span class="s1">'alpha_phase'</span><span class="p">:</span> <span class="mf">2.5</span>
<span class="linenos">114</span> <span class="p">},</span>
<span class="linenos">115</span> <span class="s1">'controller_kwargs'</span><span class="p">:</span> <span class="p">{</span>
<span class="linenos">116</span> <span class="s1">'controller_type'</span><span class="p">:</span> <span class="s1">'velocity'</span>
<span class="linenos">117</span> <span class="p">},</span>
<span class="linenos">118</span> <span class="s1">'basis_generator_kwargs'</span><span class="p">:</span> <span class="p">{</span>
<span class="linenos">119</span> <span class="s1">'basis_generator_type'</span><span class="p">:</span> <span class="s1">'rbf'</span><span class="p">,</span>
<span class="linenos">120</span> <span class="s1">'num_basis'</span><span class="p">:</span> <span class="mi">5</span>
<span class="linenos">121</span> <span class="p">}</span>
<span class="linenos">122</span> <span class="p">}</span>
<span class="linenos">123</span> <span class="p">}</span>
<span class="linenos">124</span>
<span class="linenos">125</span>
<span class="linenos">126</span><span class="k">def</span> <span class="nf">example_fully_custom_mp</span><span class="p">(</span><span class="n">seed</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">iterations</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">render</span><span class="o">=</span><span class="kc">True</span><span class="p">):</span>
<span class="linenos">127</span><span class="w"> </span><span class="sd">"""</span>
<span class="linenos">128</span><span class="sd"> Example for running a custom movement primitive based environments.</span>
<span class="linenos">129</span><span class="sd"> Our already registered environments follow the same structure.</span>
<span class="linenos">130</span><span class="sd"> Hence, this also allows to adjust hyperparameters of the movement primitives.</span>
<span class="linenos">131</span><span class="sd"> Yet, we recommend the method above if you are just interested in changing those parameters for existing tasks.</span>
<span class="linenos">132</span><span class="sd"> We appreciate PRs for custom environments (especially MP wrappers of existing tasks) </span>
<span class="linenos">133</span><span class="sd"> for our repo: https://github.com/ALRhub/fancy_gym/</span>
<span class="linenos">134</span><span class="sd"> Args:</span>
<span class="linenos">135</span><span class="sd"> seed: seed</span>
<span class="linenos">136</span><span class="sd"> iterations: Number of rollout steps to run</span>
<span class="linenos">137</span><span class="sd"> render: Render the episode</span>
<span class="linenos">138</span>
<span class="linenos">139</span><span class="sd"> Returns:</span>
<span class="linenos">140</span>
<span class="linenos">141</span><span class="sd"> """</span>
<span class="linenos">142</span>
<span class="linenos">143</span> <span class="n">base_env_id</span> <span class="o">=</span> <span class="s2">"fancy/Reacher5d-v0"</span>
<span class="linenos">144</span> <span class="n">custom_env_id</span> <span class="o">=</span> <span class="s2">"fancy/Reacher5d-Custom-v0"</span>
<span class="linenos">145</span> <span class="n">custom_env_id_DMP</span> <span class="o">=</span> <span class="s2">"fancy_DMP/Reacher5d-Custom-v0"</span>
<span class="linenos">146</span> <span class="n">custom_env_id_ProMP</span> <span class="o">=</span> <span class="s2">"fancy_ProMP/Reacher5d-Custom-v0"</span>
<span class="linenos">147</span>
<span class="linenos">148</span> <span class="n">fancy_gym</span><span class="o">.</span><span class="n">upgrade</span><span class="p">(</span><span class="n">custom_env_id</span><span class="p">,</span> <span class="n">mp_wrapper</span><span class="o">=</span><span class="n">Custom_MPWrapper</span><span class="p">,</span> <span class="n">add_mp_types</span><span class="o">=</span><span class="p">[</span><span class="s1">'ProMP'</span><span class="p">,</span> <span class="s1">'DMP'</span><span class="p">],</span> <span class="n">base_id</span><span class="o">=</span><span class="n">base_env_id</span><span class="p">)</span>
<span class="linenos">149</span>
<span class="linenos">150</span> <span class="n">env</span> <span class="o">=</span> <span class="n">gym</span><span class="o">.</span><span class="n">make</span><span class="p">(</span><span class="n">custom_env_id_ProMP</span><span class="p">,</span> <span class="n">render_mode</span><span class="o">=</span><span class="s1">'human'</span> <span class="k">if</span> <span class="n">render</span> <span class="k">else</span> <span class="kc">None</span><span class="p">)</span>
<span class="linenos">151</span>
<span class="linenos">152</span> <span class="n">rewards</span> <span class="o">=</span> <span class="mi">0</span>
<span class="linenos">153</span> <span class="n">obs</span> <span class="o">=</span> <span class="n">env</span><span class="o">.</span><span class="n">reset</span><span class="p">()</span>
<span class="linenos">154</span>
<span class="linenos">155</span> <span class="k">if</span> <span class="n">render</span><span class="p">:</span>
<span class="linenos">156</span> <span class="n">env</span><span class="o">.</span><span class="n">render</span><span class="p">()</span>
<span class="linenos">157</span>
<span class="linenos">158</span> <span class="c1"># number of samples/full trajectories (multiple environment steps)</span>
<span class="linenos">159</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">iterations</span><span class="p">):</span>
<span class="linenos">160</span> <span class="n">ac</span> <span class="o">=</span> <span class="n">env</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="linenos">161</span> <span class="n">obs</span><span class="p">,</span> <span class="n">reward</span><span class="p">,</span> <span class="n">terminated</span><span class="p">,</span> <span class="n">truncated</span><span class="p">,</span> <span class="n">info</span> <span class="o">=</span> <span class="n">env</span><span class="o">.</span><span class="n">step</span><span class="p">(</span><span class="n">ac</span><span class="p">)</span>
<span class="linenos">162</span> <span class="n">rewards</span> <span class="o">+=</span> <span class="n">reward</span>
<span class="linenos">163</span>
<span class="linenos">164</span> <span class="k">if</span> <span class="n">terminated</span> <span class="ow">or</span> <span class="n">truncated</span><span class="p">:</span>
<span class="linenos">165</span> <span class="nb">print</span><span class="p">(</span><span class="n">rewards</span><span class="p">)</span>
<span class="linenos">166</span> <span class="n">rewards</span> <span class="o">=</span> <span class="mi">0</span>
<span class="linenos">167</span> <span class="n">obs</span> <span class="o">=</span> <span class="n">env</span><span class="o">.</span><span class="n">reset</span><span class="p">()</span>
<span class="linenos">168</span>
<span class="linenos">169</span> <span class="k">try</span><span class="p">:</span> <span class="c1"># Some mujoco-based envs don't correlcty implement .close</span>
<span class="linenos">170</span> <span class="n">env</span><span class="o">.</span><span class="n">close</span><span class="p">()</span>
<span class="linenos">171</span> <span class="k">except</span><span class="p">:</span>
<span class="linenos">172</span> <span class="k">pass</span>
<span class="linenos">173</span>
<span class="linenos">174</span>
<span class="linenos">175</span><span class="k">def</span> <span class="nf">example_fully_custom_mp_alternative</span><span class="p">(</span><span class="n">seed</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">iterations</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">render</span><span class="o">=</span><span class="kc">True</span><span class="p">):</span>
<span class="linenos">176</span><span class="w"> </span><span class="sd">"""</span>
<span class="linenos">177</span><span class="sd"> Instead of defining the mp_args in a new custom MP_Wrapper, they can also be provided during registration.</span>
<span class="linenos">178</span><span class="sd"> Args:</span>
<span class="linenos">179</span><span class="sd"> seed: seed</span>
<span class="linenos">180</span><span class="sd"> iterations: Number of rollout steps to run</span>
<span class="linenos">181</span><span class="sd"> render: Render the episode</span>
<span class="linenos">182</span>
<span class="linenos">183</span><span class="sd"> Returns:</span>
<span class="linenos">184</span>
<span class="linenos">185</span><span class="sd"> """</span>
<span class="linenos">186</span>
<span class="linenos">187</span> <span class="n">base_env_id</span> <span class="o">=</span> <span class="s2">"fancy/Reacher5d-v0"</span>
<span class="linenos">188</span> <span class="n">custom_env_id</span> <span class="o">=</span> <span class="s2">"fancy/Reacher5d-Custom-v0"</span>
<span class="linenos">189</span> <span class="n">custom_env_id_ProMP</span> <span class="o">=</span> <span class="s2">"fancy_ProMP/Reacher5d-Custom-v0"</span>
<span class="linenos">190</span>
<span class="linenos">191</span> <span class="n">fancy_gym</span><span class="o">.</span><span class="n">upgrade</span><span class="p">(</span><span class="n">custom_env_id</span><span class="p">,</span> <span class="n">mp_wrapper</span><span class="o">=</span><span class="n">fancy_gym</span><span class="o">.</span><span class="n">envs</span><span class="o">.</span><span class="n">mujoco</span><span class="o">.</span><span class="n">reacher</span><span class="o">.</span><span class="n">MPWrapper</span><span class="p">,</span> <span class="n">add_mp_types</span><span class="o">=</span><span class="p">[</span><span class="s1">'ProMP'</span><span class="p">],</span> <span class="n">base_id</span><span class="o">=</span><span class="n">base_env_id</span><span class="p">,</span> <span class="n">mp_config_override</span><span class="o">=</span> <span class="p">{</span><span class="s1">'ProMP'</span><span class="p">:</span> <span class="p">{</span>
<span class="linenos">192</span> <span class="s1">'trajectory_generator_kwargs'</span><span class="p">:</span> <span class="p">{</span>
<span class="linenos">193</span> <span class="s1">'trajectory_generator_type'</span><span class="p">:</span> <span class="s1">'promp'</span><span class="p">,</span>
<span class="linenos">194</span> <span class="s1">'weights_scale'</span><span class="p">:</span> <span class="mi">2</span>
<span class="linenos">195</span> <span class="p">},</span>
<span class="linenos">196</span> <span class="s1">'phase_generator_kwargs'</span><span class="p">:</span> <span class="p">{</span>
<span class="linenos">197</span> <span class="s1">'phase_generator_type'</span><span class="p">:</span> <span class="s1">'linear'</span>
<span class="linenos">198</span> <span class="p">},</span>
<span class="linenos">199</span> <span class="s1">'controller_kwargs'</span><span class="p">:</span> <span class="p">{</span>
<span class="linenos">200</span> <span class="s1">'controller_type'</span><span class="p">:</span> <span class="s1">'velocity'</span>
<span class="linenos">201</span> <span class="p">},</span>
<span class="linenos">202</span> <span class="s1">'basis_generator_kwargs'</span><span class="p">:</span> <span class="p">{</span>
<span class="linenos">203</span> <span class="s1">'basis_generator_type'</span><span class="p">:</span> <span class="s1">'zero_rbf'</span><span class="p">,</span>
<span class="linenos">204</span> <span class="s1">'num_basis'</span><span class="p">:</span> <span class="mi">5</span><span class="p">,</span>
<span class="linenos">205</span> <span class="s1">'num_basis_zero_start'</span><span class="p">:</span> <span class="mi">1</span>
<span class="linenos">206</span> <span class="p">}</span>
<span class="linenos">207</span> <span class="p">}})</span>
<span class="linenos">208</span>
<span class="linenos">209</span> <span class="n">env</span> <span class="o">=</span> <span class="n">gym</span><span class="o">.</span><span class="n">make</span><span class="p">(</span><span class="n">custom_env_id_ProMP</span><span class="p">,</span> <span class="n">render_mode</span><span class="o">=</span><span class="s1">'human'</span> <span class="k">if</span> <span class="n">render</span> <span class="k">else</span> <span class="kc">None</span><span class="p">)</span>
<span class="linenos">210</span>
<span class="linenos">211</span> <span class="n">rewards</span> <span class="o">=</span> <span class="mi">0</span>
<span class="linenos">212</span> <span class="n">obs</span> <span class="o">=</span> <span class="n">env</span><span class="o">.</span><span class="n">reset</span><span class="p">()</span>
<span class="linenos">213</span>
<span class="linenos">214</span> <span class="k">if</span> <span class="n">render</span><span class="p">:</span>
<span class="linenos">215</span> <span class="n">env</span><span class="o">.</span><span class="n">render</span><span class="p">()</span>
<span class="linenos">216</span>
<span class="linenos">217</span> <span class="c1"># number of samples/full trajectories (multiple environment steps)</span>
<span class="linenos">218</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">iterations</span><span class="p">):</span>
<span class="linenos">219</span> <span class="n">ac</span> <span class="o">=</span> <span class="n">env</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="linenos">220</span> <span class="n">obs</span><span class="p">,</span> <span class="n">reward</span><span class="p">,</span> <span class="n">terminated</span><span class="p">,</span> <span class="n">truncated</span><span class="p">,</span> <span class="n">info</span> <span class="o">=</span> <span class="n">env</span><span class="o">.</span><span class="n">step</span><span class="p">(</span><span class="n">ac</span><span class="p">)</span>
<span class="linenos">221</span> <span class="n">rewards</span> <span class="o">+=</span> <span class="n">reward</span>
<span class="linenos">222</span>
<span class="linenos">223</span> <span class="k">if</span> <span class="n">terminated</span> <span class="ow">or</span> <span class="n">truncated</span><span class="p">:</span>
<span class="linenos">224</span> <span class="nb">print</span><span class="p">(</span><span class="n">rewards</span><span class="p">)</span>
<span class="linenos">225</span> <span class="n">rewards</span> <span class="o">=</span> <span class="mi">0</span>
<span class="linenos">226</span> <span class="n">obs</span> <span class="o">=</span> <span class="n">env</span><span class="o">.</span><span class="n">reset</span><span class="p">()</span>
<span class="linenos">227</span>
<span class="linenos">228</span> <span class="k">if</span> <span class="n">render</span><span class="p">:</span>
<span class="linenos">229</span> <span class="n">env</span><span class="o">.</span><span class="n">render</span><span class="p">()</span>
<span class="linenos">230</span>
<span class="linenos">231</span> <span class="n">rewards</span> <span class="o">=</span> <span class="mi">0</span>
<span class="linenos">232</span> <span class="n">obs</span> <span class="o">=</span> <span class="n">env</span><span class="o">.</span><span class="n">reset</span><span class="p">()</span>
<span class="linenos">233</span>
<span class="linenos">234</span> <span class="c1"># number of samples/full trajectories (multiple environment steps)</span>
<span class="linenos">235</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">iterations</span><span class="p">):</span>
<span class="linenos">236</span> <span class="n">ac</span> <span class="o">=</span> <span class="n">env</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="linenos">237</span> <span class="n">obs</span><span class="p">,</span> <span class="n">reward</span><span class="p">,</span> <span class="n">terminated</span><span class="p">,</span> <span class="n">truncated</span><span class="p">,</span> <span class="n">info</span> <span class="o">=</span> <span class="n">env</span><span class="o">.</span><span class="n">step</span><span class="p">(</span><span class="n">ac</span><span class="p">)</span>
<span class="linenos">238</span> <span class="n">rewards</span> <span class="o">+=</span> <span class="n">reward</span>
<span class="linenos">239</span>
<span class="linenos">240</span> <span class="k">if</span> <span class="n">terminated</span> <span class="ow">or</span> <span class="n">truncated</span><span class="p">:</span>
<span class="linenos">241</span> <span class="nb">print</span><span class="p">(</span><span class="n">rewards</span><span class="p">)</span>
<span class="linenos">242</span> <span class="n">rewards</span> <span class="o">=</span> <span class="mi">0</span>
<span class="linenos">243</span> <span class="n">obs</span> <span class="o">=</span> <span class="n">env</span><span class="o">.</span><span class="n">reset</span><span class="p">()</span>
<span class="linenos">244</span>
<span class="linenos">245</span> <span class="k">try</span><span class="p">:</span> <span class="c1"># Some mujoco-based envs don't correlcty implement .close</span>
<span class="linenos">246</span> <span class="n">env</span><span class="o">.</span><span class="n">close</span><span class="p">()</span>
<span class="linenos">247</span> <span class="k">except</span><span class="p">:</span>
<span class="linenos">248</span> <span class="k">pass</span>
<span class="linenos">249</span>
<span class="linenos">250</span>
<span class="linenos">251</span><span class="k">def</span> <span class="nf">main</span><span class="p">():</span>
<span class="linenos">252</span> <span class="n">render</span> <span class="o">=</span> <span class="kc">False</span>
<span class="linenos">253</span> <span class="c1"># DMP</span>
<span class="linenos">254</span> <span class="n">example_mp</span><span class="p">(</span><span class="s2">"fancy_DMP/HoleReacher-v0"</span><span class="p">,</span> <span class="n">seed</span><span class="o">=</span><span class="mi">10</span><span class="p">,</span> <span class="n">iterations</span><span class="o">=</span><span class="mi">5</span><span class="p">,</span> <span class="n">render</span><span class="o">=</span><span class="n">render</span><span class="p">)</span>
<span class="linenos">255</span>
<span class="linenos">256</span> <span class="c1"># ProMP</span>
<span class="linenos">257</span> <span class="n">example_mp</span><span class="p">(</span><span class="s2">"fancy_ProMP/HoleReacher-v0"</span><span class="p">,</span> <span class="n">seed</span><span class="o">=</span><span class="mi">10</span><span class="p">,</span> <span class="n">iterations</span><span class="o">=</span><span class="mi">5</span><span class="p">,</span> <span class="n">render</span><span class="o">=</span><span class="n">render</span><span class="p">)</span>
<span class="linenos">258</span> <span class="n">example_mp</span><span class="p">(</span><span class="s2">"fancy_ProMP/BoxPushingTemporalSparse-v0"</span><span class="p">,</span> <span class="n">seed</span><span class="o">=</span><span class="mi">10</span><span class="p">,</span> <span class="n">iterations</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">render</span><span class="o">=</span><span class="n">render</span><span class="p">)</span>
<span class="linenos">259</span> <span class="n">example_mp</span><span class="p">(</span><span class="s2">"fancy_ProMP/TableTennis4D-v0"</span><span class="p">,</span> <span class="n">seed</span><span class="o">=</span><span class="mi">10</span><span class="p">,</span> <span class="n">iterations</span><span class="o">=</span><span class="mi">20</span><span class="p">,</span> <span class="n">render</span><span class="o">=</span><span class="n">render</span><span class="p">)</span>
<span class="linenos">260</span>
<span class="linenos">261</span> <span class="c1"># ProDMP with Replanning</span>
<span class="linenos">262</span> <span class="n">example_mp</span><span class="p">(</span><span class="s2">"fancy_ProDMP/BoxPushingDenseReplan-v0"</span><span class="p">,</span> <span class="n">seed</span><span class="o">=</span><span class="mi">10</span><span class="p">,</span> <span class="n">iterations</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span> <span class="n">render</span><span class="o">=</span><span class="n">render</span><span class="p">)</span>
<span class="linenos">263</span> <span class="n">example_mp</span><span class="p">(</span><span class="s2">"fancy_ProDMP/TableTennis4DReplan-v0"</span><span class="p">,</span> <span class="n">seed</span><span class="o">=</span><span class="mi">10</span><span class="p">,</span> <span class="n">iterations</span><span class="o">=</span><span class="mi">20</span><span class="p">,</span> <span class="n">render</span><span class="o">=</span><span class="n">render</span><span class="p">)</span>
<span class="linenos">264</span> <span class="n">example_mp</span><span class="p">(</span><span class="s2">"fancy_ProDMP/TableTennisWindReplan-v0"</span><span class="p">,</span> <span class="n">seed</span><span class="o">=</span><span class="mi">10</span><span class="p">,</span> <span class="n">iterations</span><span class="o">=</span><span class="mi">20</span><span class="p">,</span> <span class="n">render</span><span class="o">=</span><span class="n">render</span><span class="p">)</span>
<span class="linenos">265</span>
<span class="linenos">266</span> <span class="c1"># Altered basis functions</span>
<span class="linenos">267</span> <span class="n">obs1</span> <span class="o">=</span> <span class="n">example_custom_mp</span><span class="p">(</span><span class="s2">"fancy_ProMP/Reacher5d-v0"</span><span class="p">,</span> <span class="n">seed</span><span class="o">=</span><span class="mi">10</span><span class="p">,</span> <span class="n">iterations</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">render</span><span class="o">=</span><span class="n">render</span><span class="p">)</span>
<span class="linenos">268</span>
<span class="linenos">269</span> <span class="c1"># Custom MP</span>
<span class="linenos">270</span> <span class="n">example_fully_custom_mp</span><span class="p">(</span><span class="n">seed</span><span class="o">=</span><span class="mi">10</span><span class="p">,</span> <span class="n">iterations</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">render</span><span class="o">=</span><span class="n">render</span><span class="p">)</span>
<span class="linenos">271</span> <span class="n">example_fully_custom_mp_alternative</span><span class="p">(</span><span class="n">seed</span><span class="o">=</span><span class="mi">10</span><span class="p">,</span> <span class="n">iterations</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">render</span><span class="o">=</span><span class="n">render</span><span class="p">)</span>
<span class="linenos">272</span>
<span class="linenos">273</span><span class="k">if</span> <span class="vm">__name__</span><span class="o">==</span><span class="s1">'__main__'</span><span class="p">:</span>
<span class="linenos">274</span> <span class="n">main</span><span class="p">()</span>
</pre></div>
</div>
</section>
</div>
</div>
<footer><div class="rst-footer-buttons" role="navigation" aria-label="Footer">
<a href="open_ai.html" class="btn btn-neutral float-left" title="OpenAI Envs Examples" accesskey="p" rel="prev"><span class="fa fa-arrow-circle-left" aria-hidden="true"></span> Previous</a>
<a href="mp_params_tuning.html" class="btn btn-neutral float-right" title="MP Params Tuning Example" accesskey="n" rel="next">Next <span class="fa fa-arrow-circle-right" aria-hidden="true"></span></a>
</div>
<hr/>
<div role="contentinfo">
<p>© Copyright 2020-2024, Fabian Otto, Onur Celik, Dominik Roth, Hongyi Zhou.</p>
</div>
Built with <a href="https://www.sphinx-doc.org/">Sphinx</a> using a
<a href="https://github.com/readthedocs/sphinx_rtd_theme">theme</a>
provided by <a href="https://readthedocs.org">Read the Docs</a>.
</footer>
</div>
</div>
</section>
</div>
<script>
jQuery(function () {
SphinxRtdTheme.Navigation.enable(true);
});
</script>
</body>
</html>