mirror of
https://github.com/gryf/coach.git
synced 2025-12-17 19:20:19 +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
703
docs/_modules/rl_coach/environments/gym_environment.html
Normal file
703
docs/_modules/rl_coach/environments/gym_environment.html
Normal file
@@ -0,0 +1,703 @@
|
||||
|
||||
|
||||
<!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.environments.gym_environment — 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.environments.gym_environment</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.environments.gym_environment</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">import</span> <span class="nn">gym</span>
|
||||
<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
|
||||
<span class="kn">import</span> <span class="nn">scipy.ndimage</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.graph_managers.graph_manager</span> <span class="k">import</span> <span class="n">ScheduleParameters</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.utils</span> <span class="k">import</span> <span class="n">lower_under_to_upper</span><span class="p">,</span> <span class="n">short_dynamic_import</span>
|
||||
|
||||
<span class="k">try</span><span class="p">:</span>
|
||||
<span class="kn">import</span> <span class="nn">roboschool</span>
|
||||
<span class="kn">from</span> <span class="nn">OpenGL</span> <span class="k">import</span> <span class="n">GL</span>
|
||||
<span class="k">except</span> <span class="ne">ImportError</span><span class="p">:</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.logger</span> <span class="k">import</span> <span class="n">failed_imports</span>
|
||||
<span class="n">failed_imports</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="s2">"RoboSchool"</span><span class="p">)</span>
|
||||
|
||||
<span class="k">try</span><span class="p">:</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.gym_extensions.continuous</span> <span class="k">import</span> <span class="n">mujoco</span>
|
||||
<span class="k">except</span><span class="p">:</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.logger</span> <span class="k">import</span> <span class="n">failed_imports</span>
|
||||
<span class="n">failed_imports</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="s2">"GymExtensions"</span><span class="p">)</span>
|
||||
|
||||
<span class="k">try</span><span class="p">:</span>
|
||||
<span class="kn">import</span> <span class="nn">pybullet_envs</span>
|
||||
<span class="k">except</span> <span class="ne">ImportError</span><span class="p">:</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.logger</span> <span class="k">import</span> <span class="n">failed_imports</span>
|
||||
<span class="n">failed_imports</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="s2">"PyBullet"</span><span class="p">)</span>
|
||||
|
||||
<span class="kn">from</span> <span class="nn">typing</span> <span class="k">import</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Union</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">EnvironmentSteps</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.environments.environment</span> <span class="k">import</span> <span class="n">Environment</span><span class="p">,</span> <span class="n">EnvironmentParameters</span><span class="p">,</span> <span class="n">LevelSelection</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.spaces</span> <span class="k">import</span> <span class="n">DiscreteActionSpace</span><span class="p">,</span> <span class="n">BoxActionSpace</span><span class="p">,</span> <span class="n">ImageObservationSpace</span><span class="p">,</span> <span class="n">VectorObservationSpace</span><span class="p">,</span> \
|
||||
<span class="n">StateSpace</span><span class="p">,</span> <span class="n">RewardSpace</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.filters.filter</span> <span class="k">import</span> <span class="n">NoInputFilter</span><span class="p">,</span> <span class="n">NoOutputFilter</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.filters.reward.reward_clipping_filter</span> <span class="k">import</span> <span class="n">RewardClippingFilter</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.filters.observation.observation_rescale_to_size_filter</span> <span class="k">import</span> <span class="n">ObservationRescaleToSizeFilter</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.filters.observation.observation_stacking_filter</span> <span class="k">import</span> <span class="n">ObservationStackingFilter</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.filters.observation.observation_rgb_to_y_filter</span> <span class="k">import</span> <span class="n">ObservationRGBToYFilter</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.filters.observation.observation_to_uint8_filter</span> <span class="k">import</span> <span class="n">ObservationToUInt8Filter</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.filters.filter</span> <span class="k">import</span> <span class="n">InputFilter</span>
|
||||
<span class="kn">import</span> <span class="nn">random</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.base_parameters</span> <span class="k">import</span> <span class="n">VisualizationParameters</span>
|
||||
<span class="kn">from</span> <span class="nn">rl_coach.logger</span> <span class="k">import</span> <span class="n">screen</span>
|
||||
|
||||
|
||||
<span class="c1"># Parameters</span>
|
||||
<span class="k">class</span> <span class="nc">GymEnvironmentParameters</span><span class="p">(</span><span class="n">EnvironmentParameters</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">level</span><span class="o">=</span><span class="kc">None</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="n">level</span><span class="o">=</span><span class="n">level</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">random_initialization_steps</span> <span class="o">=</span> <span class="mi">0</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">max_over_num_frames</span> <span class="o">=</span> <span class="mi">1</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">additional_simulator_parameters</span> <span class="o">=</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.environments.gym_environment:GymEnvironment'</span>
|
||||
|
||||
|
||||
<span class="c1"># Generic parameters for vector environments such as mujoco, roboschool, bullet, etc.</span>
|
||||
<span class="k">class</span> <span class="nc">GymVectorEnvironment</span><span class="p">(</span><span class="n">GymEnvironmentParameters</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">level</span><span class="o">=</span><span class="kc">None</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="n">level</span><span class="o">=</span><span class="n">level</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">frame_skip</span> <span class="o">=</span> <span class="mi">1</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">default_input_filter</span> <span class="o">=</span> <span class="n">NoInputFilter</span><span class="p">()</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">default_output_filter</span> <span class="o">=</span> <span class="n">NoOutputFilter</span><span class="p">()</span>
|
||||
|
||||
|
||||
<span class="c1"># Roboschool</span>
|
||||
<span class="n">gym_roboschool_envs</span> <span class="o">=</span> <span class="p">[</span><span class="s1">'inverted_pendulum'</span><span class="p">,</span> <span class="s1">'inverted_pendulum_swingup'</span><span class="p">,</span> <span class="s1">'inverted_double_pendulum'</span><span class="p">,</span> <span class="s1">'reacher'</span><span class="p">,</span>
|
||||
<span class="s1">'hopper'</span><span class="p">,</span> <span class="s1">'walker2d'</span><span class="p">,</span> <span class="s1">'half_cheetah'</span><span class="p">,</span> <span class="s1">'ant'</span><span class="p">,</span> <span class="s1">'humanoid'</span><span class="p">,</span> <span class="s1">'humanoid_flagrun'</span><span class="p">,</span>
|
||||
<span class="s1">'humanoid_flagrun_harder'</span><span class="p">,</span> <span class="s1">'pong'</span><span class="p">]</span>
|
||||
<span class="n">roboschool_v0</span> <span class="o">=</span> <span class="p">{</span><span class="n">e</span><span class="p">:</span> <span class="s2">"</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">lower_under_to_upper</span><span class="p">(</span><span class="n">e</span><span class="p">)</span> <span class="o">+</span> <span class="s1">'-v0'</span><span class="p">)</span> <span class="k">for</span> <span class="n">e</span> <span class="ow">in</span> <span class="n">gym_roboschool_envs</span><span class="p">}</span>
|
||||
|
||||
<span class="c1"># Mujoco</span>
|
||||
<span class="n">gym_mujoco_envs</span> <span class="o">=</span> <span class="p">[</span><span class="s1">'inverted_pendulum'</span><span class="p">,</span> <span class="s1">'inverted_double_pendulum'</span><span class="p">,</span> <span class="s1">'reacher'</span><span class="p">,</span> <span class="s1">'hopper'</span><span class="p">,</span> <span class="s1">'walker2d'</span><span class="p">,</span> <span class="s1">'half_cheetah'</span><span class="p">,</span>
|
||||
<span class="s1">'ant'</span><span class="p">,</span> <span class="s1">'swimmer'</span><span class="p">,</span> <span class="s1">'humanoid'</span><span class="p">,</span> <span class="s1">'humanoid_standup'</span><span class="p">,</span> <span class="s1">'pusher'</span><span class="p">,</span> <span class="s1">'thrower'</span><span class="p">,</span> <span class="s1">'striker'</span><span class="p">]</span>
|
||||
|
||||
<span class="n">mujoco_v2</span> <span class="o">=</span> <span class="p">{</span><span class="n">e</span><span class="p">:</span> <span class="s2">"</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">lower_under_to_upper</span><span class="p">(</span><span class="n">e</span><span class="p">)</span> <span class="o">+</span> <span class="s1">'-v2'</span><span class="p">)</span> <span class="k">for</span> <span class="n">e</span> <span class="ow">in</span> <span class="n">gym_mujoco_envs</span><span class="p">}</span>
|
||||
<span class="n">mujoco_v2</span><span class="p">[</span><span class="s1">'walker2d'</span><span class="p">]</span> <span class="o">=</span> <span class="s1">'Walker2d-v2'</span>
|
||||
|
||||
<span class="c1"># Fetch</span>
|
||||
<span class="n">gym_fetch_envs</span> <span class="o">=</span> <span class="p">[</span><span class="s1">'reach'</span><span class="p">,</span> <span class="s1">'slide'</span><span class="p">,</span> <span class="s1">'push'</span><span class="p">,</span> <span class="s1">'pick_and_place'</span><span class="p">]</span>
|
||||
<span class="n">fetch_v1</span> <span class="o">=</span> <span class="p">{</span><span class="n">e</span><span class="p">:</span> <span class="s2">"</span><span class="si">{}</span><span class="s2">"</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="s1">'Fetch'</span> <span class="o">+</span> <span class="n">lower_under_to_upper</span><span class="p">(</span><span class="n">e</span><span class="p">)</span> <span class="o">+</span> <span class="s1">'-v1'</span><span class="p">)</span> <span class="k">for</span> <span class="n">e</span> <span class="ow">in</span> <span class="n">gym_fetch_envs</span><span class="p">}</span>
|
||||
|
||||
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd">Atari Environment Components</span>
|
||||
<span class="sd">"""</span>
|
||||
|
||||
<span class="n">AtariInputFilter</span> <span class="o">=</span> <span class="n">InputFilter</span><span class="p">(</span><span class="n">is_a_reference_filter</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
|
||||
<span class="n">AtariInputFilter</span><span class="o">.</span><span class="n">add_reward_filter</span><span class="p">(</span><span class="s1">'clipping'</span><span class="p">,</span> <span class="n">RewardClippingFilter</span><span class="p">(</span><span class="o">-</span><span class="mf">1.0</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">))</span>
|
||||
<span class="n">AtariInputFilter</span><span class="o">.</span><span class="n">add_observation_filter</span><span class="p">(</span><span class="s1">'observation'</span><span class="p">,</span> <span class="s1">'rescaling'</span><span class="p">,</span>
|
||||
<span class="n">ObservationRescaleToSizeFilter</span><span class="p">(</span><span class="n">ImageObservationSpace</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">([</span><span class="mi">84</span><span class="p">,</span> <span class="mi">84</span><span class="p">,</span> <span class="mi">3</span><span class="p">]),</span>
|
||||
<span class="n">high</span><span class="o">=</span><span class="mi">255</span><span class="p">)))</span>
|
||||
<span class="n">AtariInputFilter</span><span class="o">.</span><span class="n">add_observation_filter</span><span class="p">(</span><span class="s1">'observation'</span><span class="p">,</span> <span class="s1">'to_grayscale'</span><span class="p">,</span> <span class="n">ObservationRGBToYFilter</span><span class="p">())</span>
|
||||
<span class="n">AtariInputFilter</span><span class="o">.</span><span class="n">add_observation_filter</span><span class="p">(</span><span class="s1">'observation'</span><span class="p">,</span> <span class="s1">'to_uint8'</span><span class="p">,</span> <span class="n">ObservationToUInt8Filter</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">255</span><span class="p">))</span>
|
||||
<span class="n">AtariInputFilter</span><span class="o">.</span><span class="n">add_observation_filter</span><span class="p">(</span><span class="s1">'observation'</span><span class="p">,</span> <span class="s1">'stacking'</span><span class="p">,</span> <span class="n">ObservationStackingFilter</span><span class="p">(</span><span class="mi">4</span><span class="p">))</span>
|
||||
<span class="n">AtariOutputFilter</span> <span class="o">=</span> <span class="n">NoOutputFilter</span><span class="p">()</span>
|
||||
|
||||
|
||||
<span class="k">class</span> <span class="nc">Atari</span><span class="p">(</span><span class="n">GymEnvironmentParameters</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">level</span><span class="o">=</span><span class="kc">None</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="n">level</span><span class="o">=</span><span class="n">level</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">frame_skip</span> <span class="o">=</span> <span class="mi">4</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">max_over_num_frames</span> <span class="o">=</span> <span class="mi">2</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">random_initialization_steps</span> <span class="o">=</span> <span class="mi">30</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">default_input_filter</span> <span class="o">=</span> <span class="n">AtariInputFilter</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">default_output_filter</span> <span class="o">=</span> <span class="n">AtariOutputFilter</span>
|
||||
|
||||
|
||||
<span class="n">gym_atari_envs</span> <span class="o">=</span> <span class="p">[</span><span class="s1">'air_raid'</span><span class="p">,</span> <span class="s1">'alien'</span><span class="p">,</span> <span class="s1">'amidar'</span><span class="p">,</span> <span class="s1">'assault'</span><span class="p">,</span> <span class="s1">'asterix'</span><span class="p">,</span> <span class="s1">'asteroids'</span><span class="p">,</span> <span class="s1">'atlantis'</span><span class="p">,</span>
|
||||
<span class="s1">'bank_heist'</span><span class="p">,</span> <span class="s1">'battle_zone'</span><span class="p">,</span> <span class="s1">'beam_rider'</span><span class="p">,</span> <span class="s1">'berzerk'</span><span class="p">,</span> <span class="s1">'bowling'</span><span class="p">,</span> <span class="s1">'boxing'</span><span class="p">,</span> <span class="s1">'breakout'</span><span class="p">,</span> <span class="s1">'carnival'</span><span class="p">,</span>
|
||||
<span class="s1">'centipede'</span><span class="p">,</span> <span class="s1">'chopper_command'</span><span class="p">,</span> <span class="s1">'crazy_climber'</span><span class="p">,</span> <span class="s1">'demon_attack'</span><span class="p">,</span> <span class="s1">'double_dunk'</span><span class="p">,</span>
|
||||
<span class="s1">'elevator_action'</span><span class="p">,</span> <span class="s1">'enduro'</span><span class="p">,</span> <span class="s1">'fishing_derby'</span><span class="p">,</span> <span class="s1">'freeway'</span><span class="p">,</span> <span class="s1">'frostbite'</span><span class="p">,</span> <span class="s1">'gopher'</span><span class="p">,</span> <span class="s1">'gravitar'</span><span class="p">,</span>
|
||||
<span class="s1">'hero'</span><span class="p">,</span> <span class="s1">'ice_hockey'</span><span class="p">,</span> <span class="s1">'jamesbond'</span><span class="p">,</span> <span class="s1">'journey_escape'</span><span class="p">,</span> <span class="s1">'kangaroo'</span><span class="p">,</span> <span class="s1">'krull'</span><span class="p">,</span> <span class="s1">'kung_fu_master'</span><span class="p">,</span>
|
||||
<span class="s1">'montezuma_revenge'</span><span class="p">,</span> <span class="s1">'ms_pacman'</span><span class="p">,</span> <span class="s1">'name_this_game'</span><span class="p">,</span> <span class="s1">'phoenix'</span><span class="p">,</span> <span class="s1">'pitfall'</span><span class="p">,</span> <span class="s1">'pong'</span><span class="p">,</span> <span class="s1">'pooyan'</span><span class="p">,</span>
|
||||
<span class="s1">'private_eye'</span><span class="p">,</span> <span class="s1">'qbert'</span><span class="p">,</span> <span class="s1">'riverraid'</span><span class="p">,</span> <span class="s1">'road_runner'</span><span class="p">,</span> <span class="s1">'robotank'</span><span class="p">,</span> <span class="s1">'seaquest'</span><span class="p">,</span> <span class="s1">'skiing'</span><span class="p">,</span>
|
||||
<span class="s1">'solaris'</span><span class="p">,</span> <span class="s1">'space_invaders'</span><span class="p">,</span> <span class="s1">'star_gunner'</span><span class="p">,</span> <span class="s1">'tennis'</span><span class="p">,</span> <span class="s1">'time_pilot'</span><span class="p">,</span> <span class="s1">'tutankham'</span><span class="p">,</span> <span class="s1">'up_n_down'</span><span class="p">,</span>
|
||||
<span class="s1">'venture'</span><span class="p">,</span> <span class="s1">'video_pinball'</span><span class="p">,</span> <span class="s1">'wizard_of_wor'</span><span class="p">,</span> <span class="s1">'yars_revenge'</span><span class="p">,</span> <span class="s1">'zaxxon'</span><span class="p">]</span>
|
||||
<span class="n">atari_deterministic_v4</span> <span class="o">=</span> <span class="p">{</span><span class="n">e</span><span class="p">:</span> <span class="s2">"</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">lower_under_to_upper</span><span class="p">(</span><span class="n">e</span><span class="p">)</span> <span class="o">+</span> <span class="s1">'Deterministic-v4'</span><span class="p">)</span> <span class="k">for</span> <span class="n">e</span> <span class="ow">in</span> <span class="n">gym_atari_envs</span><span class="p">}</span>
|
||||
<span class="n">atari_no_frameskip_v4</span> <span class="o">=</span> <span class="p">{</span><span class="n">e</span><span class="p">:</span> <span class="s2">"</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">lower_under_to_upper</span><span class="p">(</span><span class="n">e</span><span class="p">)</span> <span class="o">+</span> <span class="s1">'NoFrameskip-v4'</span><span class="p">)</span> <span class="k">for</span> <span class="n">e</span> <span class="ow">in</span> <span class="n">gym_atari_envs</span><span class="p">}</span>
|
||||
|
||||
|
||||
<span class="c1"># default atari schedule used in the DeepMind papers</span>
|
||||
<span class="n">atari_schedule</span> <span class="o">=</span> <span class="n">ScheduleParameters</span><span class="p">()</span>
|
||||
<span class="n">atari_schedule</span><span class="o">.</span><span class="n">improve_steps</span> <span class="o">=</span> <span class="n">EnvironmentSteps</span><span class="p">(</span><span class="mi">50000000</span><span class="p">)</span>
|
||||
<span class="n">atari_schedule</span><span class="o">.</span><span class="n">steps_between_evaluation_periods</span> <span class="o">=</span> <span class="n">EnvironmentSteps</span><span class="p">(</span><span class="mi">250000</span><span class="p">)</span>
|
||||
<span class="n">atari_schedule</span><span class="o">.</span><span class="n">evaluation_steps</span> <span class="o">=</span> <span class="n">EnvironmentSteps</span><span class="p">(</span><span class="mi">135000</span><span class="p">)</span>
|
||||
<span class="n">atari_schedule</span><span class="o">.</span><span class="n">heatup_steps</span> <span class="o">=</span> <span class="n">EnvironmentSteps</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
|
||||
|
||||
|
||||
<span class="k">class</span> <span class="nc">MaxOverFramesAndFrameskipEnvWrapper</span><span class="p">(</span><span class="n">gym</span><span class="o">.</span><span class="n">Wrapper</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">env</span><span class="p">,</span> <span class="n">frameskip</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span> <span class="n">max_over_num_frames</span><span class="o">=</span><span class="mi">2</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="n">env</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">max_over_num_frames</span> <span class="o">=</span> <span class="n">max_over_num_frames</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">observations_stack</span> <span class="o">=</span> <span class="p">[]</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">frameskip</span> <span class="o">=</span> <span class="n">frameskip</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">first_frame_to_max_over</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">frameskip</span> <span class="o">-</span> <span class="bp">self</span><span class="o">.</span><span class="n">max_over_num_frames</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="k">return</span> <span class="bp">self</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="k">def</span> <span class="nf">step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">action</span><span class="p">):</span>
|
||||
<span class="n">total_reward</span> <span class="o">=</span> <span class="mf">0.0</span>
|
||||
<span class="n">done</span> <span class="o">=</span> <span class="kc">None</span>
|
||||
<span class="n">info</span> <span class="o">=</span> <span class="kc">None</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">observations_stack</span> <span class="o">=</span> <span class="p">[]</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="bp">self</span><span class="o">.</span><span class="n">frameskip</span><span class="p">):</span>
|
||||
<span class="n">observation</span><span class="p">,</span> <span class="n">reward</span><span class="p">,</span> <span class="n">done</span><span class="p">,</span> <span class="n">info</span> <span class="o">=</span> <span class="bp">self</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">action</span><span class="p">)</span>
|
||||
<span class="k">if</span> <span class="n">i</span> <span class="o">>=</span> <span class="bp">self</span><span class="o">.</span><span class="n">first_frame_to_max_over</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">observations_stack</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">observation</span><span class="p">)</span>
|
||||
<span class="n">total_reward</span> <span class="o">+=</span> <span class="n">reward</span>
|
||||
<span class="k">if</span> <span class="n">done</span><span class="p">:</span>
|
||||
<span class="c1"># deal with last state in episode</span>
|
||||
<span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">observations_stack</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">observations_stack</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">observation</span><span class="p">)</span>
|
||||
<span class="k">break</span>
|
||||
|
||||
<span class="n">max_over_frames_observation</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">observations_stack</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">return</span> <span class="n">max_over_frames_observation</span><span class="p">,</span> <span class="n">total_reward</span><span class="p">,</span> <span class="n">done</span><span class="p">,</span> <span class="n">info</span>
|
||||
|
||||
|
||||
<span class="c1"># Environment</span>
|
||||
<div class="viewcode-block" id="GymEnvironment"><a class="viewcode-back" href="../../../components/environments/index.html#rl_coach.environments.gym_environment.GymEnvironment">[docs]</a><span class="k">class</span> <span class="nc">GymEnvironment</span><span class="p">(</span><span class="n">Environment</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">level</span><span class="p">:</span> <span class="n">LevelSelection</span><span class="p">,</span> <span class="n">frame_skip</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">visualization_parameters</span><span class="p">:</span> <span class="n">VisualizationParameters</span><span class="p">,</span>
|
||||
<span class="n">target_success_rate</span><span class="p">:</span> <span class="nb">float</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span> <span class="n">additional_simulator_parameters</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">Any</span><span class="p">]</span> <span class="o">=</span> <span class="p">{},</span> <span class="n">seed</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
|
||||
<span class="n">human_control</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">custom_reward_threshold</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">]</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
|
||||
<span class="n">random_initialization_steps</span><span class="p">:</span> <span class="nb">int</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">max_over_num_frames</span><span class="p">:</span> <span class="nb">int</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> :param level: (str)</span>
|
||||
<span class="sd"> A string representing the gym level to run. This can also be a LevelSelection object.</span>
|
||||
<span class="sd"> For example, BreakoutDeterministic-v0</span>
|
||||
|
||||
<span class="sd"> :param frame_skip: (int)</span>
|
||||
<span class="sd"> The number of frames to skip between any two actions given by the agent. The action will be repeated</span>
|
||||
<span class="sd"> for all the skipped frames.</span>
|
||||
|
||||
<span class="sd"> :param visualization_parameters: (VisualizationParameters)</span>
|
||||
<span class="sd"> The parameters used for visualizing the environment, such as the render flag, storing videos etc.</span>
|
||||
|
||||
<span class="sd"> :param additional_simulator_parameters: (Dict[str, Any])</span>
|
||||
<span class="sd"> Any additional parameters that the user can pass to the Gym environment. These parameters should be</span>
|
||||
<span class="sd"> accepted by the __init__ function of the implemented Gym environment.</span>
|
||||
|
||||
<span class="sd"> :param seed: (int)</span>
|
||||
<span class="sd"> A seed to use for the random number generator when running the environment.</span>
|
||||
|
||||
<span class="sd"> :param human_control: (bool)</span>
|
||||
<span class="sd"> A flag that allows controlling the environment using the keyboard keys.</span>
|
||||
|
||||
<span class="sd"> :param custom_reward_threshold: (float)</span>
|
||||
<span class="sd"> Allows defining a custom reward that will be used to decide when the agent succeeded in passing the environment.</span>
|
||||
<span class="sd"> If not set, this value will be taken from the Gym environment definition.</span>
|
||||
|
||||
<span class="sd"> :param random_initialization_steps: (int)</span>
|
||||
<span class="sd"> The number of random steps that will be taken in the environment after each reset.</span>
|
||||
<span class="sd"> This is a feature presented in the DQN paper, which improves the variability of the episodes the agent sees.</span>
|
||||
|
||||
<span class="sd"> :param max_over_num_frames: (int)</span>
|
||||
<span class="sd"> This value will be used for merging multiple frames into a single frame by taking the maximum value for each</span>
|
||||
<span class="sd"> of the pixels in the frame. This is particularly used in Atari games, where the frames flicker, and objects</span>
|
||||
<span class="sd"> can be seen in one frame but disappear in the next.</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">level</span><span class="p">,</span> <span class="n">seed</span><span class="p">,</span> <span class="n">frame_skip</span><span class="p">,</span> <span class="n">human_control</span><span class="p">,</span> <span class="n">custom_reward_threshold</span><span class="p">,</span>
|
||||
<span class="n">visualization_parameters</span><span class="p">,</span> <span class="n">target_success_rate</span><span class="p">)</span>
|
||||
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">random_initialization_steps</span> <span class="o">=</span> <span class="n">random_initialization_steps</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">max_over_num_frames</span> <span class="o">=</span> <span class="n">max_over_num_frames</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">additional_simulator_parameters</span> <span class="o">=</span> <span class="n">additional_simulator_parameters</span>
|
||||
|
||||
<span class="c1"># hide warnings</span>
|
||||
<span class="n">gym</span><span class="o">.</span><span class="n">logger</span><span class="o">.</span><span class="n">set_level</span><span class="p">(</span><span class="mi">40</span><span class="p">)</span>
|
||||
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> load and initialize environment</span>
|
||||
<span class="sd"> environment ids can be defined in 3 ways:</span>
|
||||
<span class="sd"> 1. Native gym environments like BreakoutDeterministic-v0 for example</span>
|
||||
<span class="sd"> 2. Custom gym environments written and installed as python packages.</span>
|
||||
<span class="sd"> This environments should have a python module with a class inheriting gym.Env, implementing the</span>
|
||||
<span class="sd"> relevant functions (_reset, _step, _render) and defining the observation and action space</span>
|
||||
<span class="sd"> For example: my_environment_package:MyEnvironmentClass will run an environment defined in the</span>
|
||||
<span class="sd"> MyEnvironmentClass class</span>
|
||||
<span class="sd"> 3. Custom gym environments written as an independent module which is not installed.</span>
|
||||
<span class="sd"> This environments should have a python module with a class inheriting gym.Env, implementing the</span>
|
||||
<span class="sd"> relevant functions (_reset, _step, _render) and defining the observation and action space.</span>
|
||||
<span class="sd"> For example: path_to_my_environment.sub_directory.my_module:MyEnvironmentClass will run an</span>
|
||||
<span class="sd"> environment defined in the MyEnvironmentClass class which is located in the module in the relative path</span>
|
||||
<span class="sd"> path_to_my_environment.sub_directory.my_module</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="k">if</span> <span class="s1">':'</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">env_id</span><span class="p">:</span>
|
||||
<span class="c1"># custom environments</span>
|
||||
<span class="k">if</span> <span class="s1">'/'</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">env_id</span> <span class="ow">or</span> <span class="s1">'.'</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">env_id</span><span class="p">:</span>
|
||||
<span class="c1"># environment in a an absolute path module written as a unix path or in a relative path module</span>
|
||||
<span class="c1"># written as a python import path</span>
|
||||
<span class="n">env_class</span> <span class="o">=</span> <span class="n">short_dynamic_import</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">env_id</span><span class="p">)</span>
|
||||
<span class="k">else</span><span class="p">:</span>
|
||||
<span class="c1"># environment in a python package</span>
|
||||
<span class="n">env_class</span> <span class="o">=</span> <span class="n">gym</span><span class="o">.</span><span class="n">envs</span><span class="o">.</span><span class="n">registration</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">env_id</span><span class="p">)</span>
|
||||
|
||||
<span class="c1"># instantiate the environment</span>
|
||||
<span class="k">try</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">env</span> <span class="o">=</span> <span class="n">env_class</span><span class="p">(</span><span class="o">**</span><span class="bp">self</span><span class="o">.</span><span class="n">additional_simulator_parameters</span><span class="p">)</span>
|
||||
<span class="k">except</span><span class="p">:</span>
|
||||
<span class="n">screen</span><span class="o">.</span><span class="n">error</span><span class="p">(</span><span class="s2">"Failed to instantiate Gym environment class </span><span class="si">%s</span><span class="s2"> with arguments </span><span class="si">%s</span><span class="s2">"</span> <span class="o">%</span>
|
||||
<span class="p">(</span><span class="n">env_class</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">additional_simulator_parameters</span><span class="p">),</span> <span class="n">crash</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
|
||||
<span class="k">raise</span>
|
||||
<span class="k">else</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</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="bp">self</span><span class="o">.</span><span class="n">env_id</span><span class="p">)</span>
|
||||
|
||||
<span class="c1"># for classic control we want to use the native renderer because otherwise we will get 2 renderer windows</span>
|
||||
<span class="n">environment_to_always_use_with_native_rendering</span> <span class="o">=</span> <span class="p">[</span><span class="s1">'classic_control'</span><span class="p">,</span> <span class="s1">'mujoco'</span><span class="p">,</span> <span class="s1">'robotics'</span><span class="p">]</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">native_rendering</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">native_rendering</span> <span class="ow">or</span> \
|
||||
<span class="nb">any</span><span class="p">([</span><span class="n">env</span> <span class="ow">in</span> <span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">unwrapped</span><span class="o">.</span><span class="vm">__class__</span><span class="p">)</span>
|
||||
<span class="k">for</span> <span class="n">env</span> <span class="ow">in</span> <span class="n">environment_to_always_use_with_native_rendering</span><span class="p">])</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">native_rendering</span><span class="p">:</span>
|
||||
<span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="s1">'renderer'</span><span class="p">):</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">renderer</span><span class="o">.</span><span class="n">close</span><span class="p">()</span>
|
||||
|
||||
<span class="c1"># seed</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">seed</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">seed</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">seed</span><span class="p">)</span>
|
||||
<span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">seed</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">seed</span><span class="p">)</span>
|
||||
<span class="n">random</span><span class="o">.</span><span class="n">seed</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">seed</span><span class="p">)</span>
|
||||
|
||||
<span class="c1"># frame skip and max between consecutive frames</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">is_robotics_env</span> <span class="o">=</span> <span class="s1">'robotics'</span> <span class="ow">in</span> <span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">unwrapped</span><span class="o">.</span><span class="vm">__class__</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">is_mujoco_env</span> <span class="o">=</span> <span class="s1">'mujoco'</span> <span class="ow">in</span> <span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">unwrapped</span><span class="o">.</span><span class="vm">__class__</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">is_atari_env</span> <span class="o">=</span> <span class="s1">'Atari'</span> <span class="ow">in</span> <span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">unwrapped</span><span class="o">.</span><span class="vm">__class__</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">timelimit_env_wrapper</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">env</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_atari_env</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">unwrapped</span><span class="o">.</span><span class="n">frameskip</span> <span class="o">=</span> <span class="mi">1</span> <span class="c1"># this accesses the atari env that is wrapped with a timelimit wrapper env</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">env_id</span> <span class="o">==</span> <span class="s2">"SpaceInvadersDeterministic-v4"</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">frame_skip</span> <span class="o">==</span> <span class="mi">4</span><span class="p">:</span>
|
||||
<span class="n">screen</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">"Warning: The frame-skip for Space Invaders was automatically updated from 4 to 3. "</span>
|
||||
<span class="s2">"This is following the DQN paper where it was noticed that a frame-skip of 3 makes the "</span>
|
||||
<span class="s2">"laser rays disappear. To force frame-skip of 4, please use SpaceInvadersNoFrameskip-v4."</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">frame_skip</span> <span class="o">=</span> <span class="mi">3</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">env</span> <span class="o">=</span> <span class="n">MaxOverFramesAndFrameskipEnvWrapper</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="p">,</span>
|
||||
<span class="n">frameskip</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">frame_skip</span><span class="p">,</span>
|
||||
<span class="n">max_over_num_frames</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">max_over_num_frames</span><span class="p">)</span>
|
||||
<span class="k">else</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">unwrapped</span><span class="o">.</span><span class="n">frameskip</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">frame_skip</span>
|
||||
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">state_space</span> <span class="o">=</span> <span class="n">StateSpace</span><span class="p">({})</span>
|
||||
|
||||
<span class="c1"># observations</span>
|
||||
<span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">observation_space</span><span class="p">,</span> <span class="n">gym</span><span class="o">.</span><span class="n">spaces</span><span class="o">.</span><span class="n">dict_space</span><span class="o">.</span><span class="n">Dict</span><span class="p">):</span>
|
||||
<span class="n">state_space</span> <span class="o">=</span> <span class="p">{</span><span class="s1">'observation'</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">observation_space</span><span class="p">}</span>
|
||||
<span class="k">else</span><span class="p">:</span>
|
||||
<span class="n">state_space</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">observation_space</span><span class="o">.</span><span class="n">spaces</span>
|
||||
|
||||
<span class="k">for</span> <span class="n">observation_space_name</span><span class="p">,</span> <span class="n">observation_space</span> <span class="ow">in</span> <span class="n">state_space</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
|
||||
<span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">observation_space</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> <span class="o">==</span> <span class="mi">3</span><span class="p">:</span>
|
||||
<span class="c1"># we assume gym has image observations (with arbitrary number of channels) where their values are</span>
|
||||
<span class="c1"># within 0-255, and where the channel dimension is the last dimension</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">state_space</span><span class="p">[</span><span class="n">observation_space_name</span><span class="p">]</span> <span class="o">=</span> <span class="n">ImageObservationSpace</span><span class="p">(</span>
|
||||
<span class="n">shape</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">(</span><span class="n">observation_space</span><span class="o">.</span><span class="n">shape</span><span class="p">),</span>
|
||||
<span class="n">high</span><span class="o">=</span><span class="mi">255</span><span class="p">,</span>
|
||||
<span class="n">channels_axis</span><span class="o">=-</span><span class="mi">1</span>
|
||||
<span class="p">)</span>
|
||||
<span class="k">else</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">state_space</span><span class="p">[</span><span class="n">observation_space_name</span><span class="p">]</span> <span class="o">=</span> <span class="n">VectorObservationSpace</span><span class="p">(</span>
|
||||
<span class="n">shape</span><span class="o">=</span><span class="n">observation_space</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span>
|
||||
<span class="n">low</span><span class="o">=</span><span class="n">observation_space</span><span class="o">.</span><span class="n">low</span><span class="p">,</span>
|
||||
<span class="n">high</span><span class="o">=</span><span class="n">observation_space</span><span class="o">.</span><span class="n">high</span>
|
||||
<span class="p">)</span>
|
||||
<span class="k">if</span> <span class="s1">'desired_goal'</span> <span class="ow">in</span> <span class="n">state_space</span><span class="o">.</span><span class="n">keys</span><span class="p">():</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">goal_space</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state_space</span><span class="p">[</span><span class="s1">'desired_goal'</span><span class="p">]</span>
|
||||
|
||||
<span class="c1"># actions</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">env</span><span class="o">.</span><span class="n">action_space</span><span class="p">)</span> <span class="o">==</span> <span class="n">gym</span><span class="o">.</span><span class="n">spaces</span><span class="o">.</span><span class="n">box</span><span class="o">.</span><span class="n">Box</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">BoxActionSpace</span><span class="p">(</span>
|
||||
<span class="n">shape</span><span class="o">=</span><span class="bp">self</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">shape</span><span class="p">,</span>
|
||||
<span class="n">low</span><span class="o">=</span><span class="bp">self</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">low</span><span class="p">,</span>
|
||||
<span class="n">high</span><span class="o">=</span><span class="bp">self</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">high</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">env</span><span class="o">.</span><span class="n">action_space</span><span class="p">)</span> <span class="o">==</span> <span class="n">gym</span><span class="o">.</span><span class="n">spaces</span><span class="o">.</span><span class="n">discrete</span><span class="o">.</span><span class="n">Discrete</span><span class="p">:</span>
|
||||
<span class="n">actions_description</span> <span class="o">=</span> <span class="p">[]</span>
|
||||
<span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">unwrapped</span><span class="p">,</span> <span class="s1">'get_action_meanings'</span><span class="p">):</span>
|
||||
<span class="n">actions_description</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">unwrapped</span><span class="o">.</span><span class="n">get_action_meanings</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">DiscreteActionSpace</span><span class="p">(</span>
|
||||
<span class="n">num_actions</span><span class="o">=</span><span class="bp">self</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">n</span><span class="p">,</span>
|
||||
<span class="n">descriptions</span><span class="o">=</span><span class="n">actions_description</span>
|
||||
<span class="p">)</span>
|
||||
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">human_control</span><span class="p">:</span>
|
||||
<span class="c1"># TODO: add this to the action space</span>
|
||||
<span class="c1"># map keyboard keys to actions</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">key_to_action</span> <span class="o">=</span> <span class="p">{}</span>
|
||||
<span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">unwrapped</span><span class="p">,</span> <span class="s1">'get_keys_to_action'</span><span class="p">):</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">key_to_action</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">unwrapped</span><span class="o">.</span><span class="n">get_keys_to_action</span><span class="p">()</span>
|
||||
<span class="k">else</span><span class="p">:</span>
|
||||
<span class="n">screen</span><span class="o">.</span><span class="n">error</span><span class="p">(</span><span class="s2">"Error: Environment </span><span class="si">{}</span><span class="s2"> does not support human control."</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="p">),</span> <span class="n">crash</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
|
||||
|
||||
<span class="c1"># initialize the state by getting a new state from the environment</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">reset_internal_state</span><span class="p">(</span><span class="kc">True</span><span class="p">)</span>
|
||||
|
||||
<span class="c1"># render</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_rendered</span><span class="p">:</span>
|
||||
<span class="n">image</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_rendered_image</span><span class="p">()</span>
|
||||
<span class="n">scale</span> <span class="o">=</span> <span class="mi">1</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">human_control</span><span class="p">:</span>
|
||||
<span class="n">scale</span> <span class="o">=</span> <span class="mi">2</span>
|
||||
<span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">native_rendering</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">renderer</span><span class="o">.</span><span class="n">create_screen</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span><span class="o">*</span><span class="n">scale</span><span class="p">,</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">*</span><span class="n">scale</span><span class="p">)</span>
|
||||
|
||||
<span class="c1"># measurements</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">spec</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">timestep_limit</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">spec</span><span class="o">.</span><span class="n">timestep_limit</span>
|
||||
<span class="k">else</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">timestep_limit</span> <span class="o">=</span> <span class="kc">None</span>
|
||||
|
||||
<span class="c1"># the info is only updated after the first step</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">state</span> <span class="o">=</span> <span class="bp">self</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">action_space</span><span class="o">.</span><span class="n">default_action</span><span class="p">)</span><span class="o">.</span><span class="n">next_state</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">state_space</span><span class="p">[</span><span class="s1">'measurements'</span><span class="p">]</span> <span class="o">=</span> <span class="n">VectorObservationSpace</span><span class="p">(</span><span class="n">shape</span><span class="o">=</span><span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">info</span><span class="o">.</span><span class="n">keys</span><span class="p">()))</span>
|
||||
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">spec</span> <span class="ow">and</span> <span class="n">custom_reward_threshold</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">reward_success_threshold</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">spec</span><span class="o">.</span><span class="n">reward_threshold</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">reward_space</span> <span class="o">=</span> <span class="n">RewardSpace</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">reward_success_threshold</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">reward_success_threshold</span><span class="p">)</span>
|
||||
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">target_success_rate</span> <span class="o">=</span> <span class="n">target_success_rate</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">_wrap_state</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">state</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="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">observation_space</span><span class="p">,</span> <span class="n">gym</span><span class="o">.</span><span class="n">spaces</span><span class="o">.</span><span class="n">Dict</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="p">{</span><span class="s1">'observation'</span><span class="p">:</span> <span class="n">state</span><span class="p">}</span>
|
||||
<span class="k">return</span> <span class="n">state</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">_update_state</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">is_atari_env</span> <span class="ow">and</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="s1">'current_ale_lives'</span><span class="p">)</span> \
|
||||
<span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">current_ale_lives</span> <span class="o">!=</span> <span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">unwrapped</span><span class="o">.</span><span class="n">ale</span><span class="o">.</span><span class="n">lives</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="ow">or</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="p">:</span>
|
||||
<span class="c1"># signal termination for life loss</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">done</span> <span class="o">=</span> <span class="kc">True</span>
|
||||
<span class="k">elif</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="ow">and</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">done</span><span class="p">:</span>
|
||||
<span class="c1"># the episode is not terminated in evaluation, but we need to press fire again</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">_press_fire</span><span class="p">()</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">_update_ale_lives</span><span class="p">()</span>
|
||||
<span class="c1"># TODO: update the measurements</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span> <span class="ow">and</span> <span class="s2">"desired_goal"</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">keys</span><span class="p">():</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">goal</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="s1">'desired_goal'</span><span class="p">]</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">_take_action</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">action</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="n">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">clip_action_to_space</span><span class="p">(</span><span class="n">action</span><span class="p">)</span>
|
||||
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">reward</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">done</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">info</span> <span class="o">=</span> <span class="bp">self</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">action</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="bp">self</span><span class="o">.</span><span class="n">_wrap_state</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">)</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">_random_noop</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="c1"># simulate a random initial environment state by stepping for a random number of times between 0 and 30</span>
|
||||
<span class="n">step_count</span> <span class="o">=</span> <span class="mi">0</span>
|
||||
<span class="n">random_initialization_steps</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="mi">0</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">random_initialization_steps</span><span class="p">)</span>
|
||||
<span class="k">while</span> <span class="bp">self</span><span class="o">.</span><span class="n">action_space</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">state</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">step_count</span> <span class="o"><</span> <span class="n">random_initialization_steps</span><span class="p">):</span>
|
||||
<span class="n">step_count</span> <span class="o">+=</span> <span class="mi">1</span>
|
||||
<span class="bp">self</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">action_space</span><span class="o">.</span><span class="n">default_action</span><span class="p">)</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">_press_fire</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="n">fire_action</span> <span class="o">=</span> <span class="mi">1</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_atari_env</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">unwrapped</span><span class="o">.</span><span class="n">get_action_meanings</span><span class="p">()[</span><span class="n">fire_action</span><span class="p">]</span> <span class="o">==</span> <span class="s1">'FIRE'</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">current_ale_lives</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">unwrapped</span><span class="o">.</span><span class="n">ale</span><span class="o">.</span><span class="n">lives</span><span class="p">()</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">step</span><span class="p">(</span><span class="n">fire_action</span><span class="p">)</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">done</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">reset_internal_state</span><span class="p">()</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">_update_ale_lives</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">is_atari_env</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">current_ale_lives</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">unwrapped</span><span class="o">.</span><span class="n">ale</span><span class="o">.</span><span class="n">lives</span><span class="p">()</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">_restart_environment_episode</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">force_environment_reset</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span>
|
||||
<span class="c1"># prevent reset of environment if there are ale lives left</span>
|
||||
<span class="k">if</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">is_atari_env</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">unwrapped</span><span class="o">.</span><span class="n">ale</span><span class="o">.</span><span class="n">lives</span><span class="p">()</span> <span class="o">></span> <span class="mi">0</span><span class="p">)</span> \
|
||||
<span class="ow">and</span> <span class="ow">not</span> <span class="n">force_environment_reset</span> <span class="ow">and</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">timelimit_env_wrapper</span><span class="o">.</span><span class="n">_past_limit</span><span class="p">():</span>
|
||||
<span class="bp">self</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">action_space</span><span class="o">.</span><span class="n">default_action</span><span class="p">)</span>
|
||||
<span class="k">else</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="bp">self</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="bp">self</span><span class="o">.</span><span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_wrap_state</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">_update_ale_lives</span><span class="p">()</span>
|
||||
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_atari_env</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">_random_noop</span><span class="p">()</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">_press_fire</span><span class="p">()</span>
|
||||
|
||||
<span class="c1"># initialize the number of lives</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">_update_ale_lives</span><span class="p">()</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">_set_mujoco_camera</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">camera_idx</span><span class="p">:</span> <span class="nb">int</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> This function can be used to set the camera for rendering the mujoco simulator</span>
|
||||
<span class="sd"> :param camera_idx: The index of the camera to use. Should be defined in the model</span>
|
||||
<span class="sd"> :return: None</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">unwrapped</span><span class="o">.</span><span class="n">viewer</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">unwrapped</span><span class="o">.</span><span class="n">viewer</span><span class="o">.</span><span class="n">cam</span><span class="o">.</span><span class="n">fixedcamid</span> <span class="o">!=</span> <span class="n">camera_idx</span> <span class="ow">and</span>\
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">unwrapped</span><span class="o">.</span><span class="n">viewer</span><span class="o">.</span><span class="n">_ncam</span> <span class="o">></span> <span class="n">camera_idx</span><span class="p">:</span>
|
||||
<span class="kn">from</span> <span class="nn">mujoco_py.generated</span> <span class="k">import</span> <span class="n">const</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">unwrapped</span><span class="o">.</span><span class="n">viewer</span><span class="o">.</span><span class="n">cam</span><span class="o">.</span><span class="n">type</span> <span class="o">=</span> <span class="n">const</span><span class="o">.</span><span class="n">CAMERA_FIXED</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">unwrapped</span><span class="o">.</span><span class="n">viewer</span><span class="o">.</span><span class="n">cam</span><span class="o">.</span><span class="n">fixedcamid</span> <span class="o">=</span> <span class="n">camera_idx</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">_get_robotics_image</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">env</span><span class="o">.</span><span class="n">render</span><span class="p">()</span>
|
||||
<span class="n">image</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">unwrapped</span><span class="o">.</span><span class="n">_get_viewer</span><span class="p">()</span><span class="o">.</span><span class="n">read_pixels</span><span class="p">(</span><span class="mi">1600</span><span class="p">,</span> <span class="mi">900</span><span class="p">,</span> <span class="n">depth</span><span class="o">=</span><span class="kc">False</span><span class="p">)[::</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="p">:,</span> <span class="p">:]</span>
|
||||
<span class="n">image</span> <span class="o">=</span> <span class="n">scipy</span><span class="o">.</span><span class="n">misc</span><span class="o">.</span><span class="n">imresize</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="p">(</span><span class="mi">270</span><span class="p">,</span> <span class="mi">480</span><span class="p">,</span> <span class="mi">3</span><span class="p">))</span>
|
||||
<span class="k">return</span> <span class="n">image</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">_render</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">env</span><span class="o">.</span><span class="n">render</span><span class="p">(</span><span class="n">mode</span><span class="o">=</span><span class="s1">'human'</span><span class="p">)</span>
|
||||
<span class="c1"># required for setting up a fixed camera for mujoco</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_mujoco_env</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">_set_mujoco_camera</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">get_rendered_image</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">is_robotics_env</span><span class="p">:</span>
|
||||
<span class="c1"># necessary for fetch since the rendered image is cropped to an irrelevant part of the simulator</span>
|
||||
<span class="n">image</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_robotics_image</span><span class="p">()</span>
|
||||
<span class="k">else</span><span class="p">:</span>
|
||||
<span class="n">image</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">render</span><span class="p">(</span><span class="n">mode</span><span class="o">=</span><span class="s1">'rgb_array'</span><span class="p">)</span>
|
||||
<span class="c1"># required for setting up a fixed camera for mujoco</span>
|
||||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_mujoco_env</span><span class="p">:</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">_set_mujoco_camera</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
|
||||
<span class="k">return</span> <span class="n">image</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">get_target_success_rate</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-></span> <span class="nb">float</span><span class="p">:</span>
|
||||
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_success_rate</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