mirror of
https://github.com/gryf/coach.git
synced 2025-12-17 19:20:19 +01:00
6d40ad1650
* 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
443 lines
40 KiB
HTML
443 lines
40 KiB
HTML
|
|
|
|
<!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.agents.ddpg_agent — 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.agents.ddpg_agent</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.agents.ddpg_agent</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">copy</span>
|
|
<span class="kn">from</span> <span class="nn">typing</span> <span class="k">import</span> <span class="n">Union</span>
|
|
<span class="kn">from</span> <span class="nn">collections</span> <span class="k">import</span> <span class="n">OrderedDict</span>
|
|
|
|
<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
|
|
|
|
<span class="kn">from</span> <span class="nn">rl_coach.agents.actor_critic_agent</span> <span class="k">import</span> <span class="n">ActorCriticAgent</span>
|
|
<span class="kn">from</span> <span class="nn">rl_coach.agents.agent</span> <span class="k">import</span> <span class="n">Agent</span>
|
|
<span class="kn">from</span> <span class="nn">rl_coach.architectures.embedder_parameters</span> <span class="k">import</span> <span class="n">InputEmbedderParameters</span>
|
|
<span class="kn">from</span> <span class="nn">rl_coach.architectures.head_parameters</span> <span class="k">import</span> <span class="n">DDPGActorHeadParameters</span><span class="p">,</span> <span class="n">VHeadParameters</span>
|
|
<span class="kn">from</span> <span class="nn">rl_coach.architectures.middleware_parameters</span> <span class="k">import</span> <span class="n">FCMiddlewareParameters</span>
|
|
<span class="kn">from</span> <span class="nn">rl_coach.base_parameters</span> <span class="k">import</span> <span class="n">NetworkParameters</span><span class="p">,</span> <span class="n">AlgorithmParameters</span><span class="p">,</span> \
|
|
<span class="n">AgentParameters</span><span class="p">,</span> <span class="n">EmbedderScheme</span>
|
|
<span class="kn">from</span> <span class="nn">rl_coach.core_types</span> <span class="k">import</span> <span class="n">ActionInfo</span><span class="p">,</span> <span class="n">EnvironmentSteps</span>
|
|
<span class="kn">from</span> <span class="nn">rl_coach.exploration_policies.ou_process</span> <span class="k">import</span> <span class="n">OUProcessParameters</span>
|
|
<span class="kn">from</span> <span class="nn">rl_coach.memories.episodic.episodic_experience_replay</span> <span class="k">import</span> <span class="n">EpisodicExperienceReplayParameters</span>
|
|
<span class="kn">from</span> <span class="nn">rl_coach.spaces</span> <span class="k">import</span> <span class="n">BoxActionSpace</span><span class="p">,</span> <span class="n">GoalsSpace</span>
|
|
|
|
|
|
<span class="k">class</span> <span class="nc">DDPGCriticNetworkParameters</span><span class="p">(</span><span class="n">NetworkParameters</span><span class="p">):</span>
|
|
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
|
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">input_embedders_parameters</span> <span class="o">=</span> <span class="p">{</span><span class="s1">'observation'</span><span class="p">:</span> <span class="n">InputEmbedderParameters</span><span class="p">(</span><span class="n">batchnorm</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
|
|
<span class="s1">'action'</span><span class="p">:</span> <span class="n">InputEmbedderParameters</span><span class="p">(</span><span class="n">scheme</span><span class="o">=</span><span class="n">EmbedderScheme</span><span class="o">.</span><span class="n">Shallow</span><span class="p">)}</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">middleware_parameters</span> <span class="o">=</span> <span class="n">FCMiddlewareParameters</span><span class="p">()</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">heads_parameters</span> <span class="o">=</span> <span class="p">[</span><span class="n">VHeadParameters</span><span class="p">()]</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">optimizer_type</span> <span class="o">=</span> <span class="s1">'Adam'</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">batch_size</span> <span class="o">=</span> <span class="mi">64</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">async_training</span> <span class="o">=</span> <span class="kc">False</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">learning_rate</span> <span class="o">=</span> <span class="mf">0.001</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">create_target_network</span> <span class="o">=</span> <span class="kc">True</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">shared_optimizer</span> <span class="o">=</span> <span class="kc">True</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">scale_down_gradients_by_number_of_workers_for_sync_training</span> <span class="o">=</span> <span class="kc">False</span>
|
|
|
|
|
|
<span class="k">class</span> <span class="nc">DDPGActorNetworkParameters</span><span class="p">(</span><span class="n">NetworkParameters</span><span class="p">):</span>
|
|
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
|
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">input_embedders_parameters</span> <span class="o">=</span> <span class="p">{</span><span class="s1">'observation'</span><span class="p">:</span> <span class="n">InputEmbedderParameters</span><span class="p">(</span><span class="n">batchnorm</span><span class="o">=</span><span class="kc">True</span><span class="p">)}</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">middleware_parameters</span> <span class="o">=</span> <span class="n">FCMiddlewareParameters</span><span class="p">(</span><span class="n">batchnorm</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">heads_parameters</span> <span class="o">=</span> <span class="p">[</span><span class="n">DDPGActorHeadParameters</span><span class="p">()]</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">optimizer_type</span> <span class="o">=</span> <span class="s1">'Adam'</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">batch_size</span> <span class="o">=</span> <span class="mi">64</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">async_training</span> <span class="o">=</span> <span class="kc">False</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">learning_rate</span> <span class="o">=</span> <span class="mf">0.0001</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">create_target_network</span> <span class="o">=</span> <span class="kc">True</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">shared_optimizer</span> <span class="o">=</span> <span class="kc">True</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">scale_down_gradients_by_number_of_workers_for_sync_training</span> <span class="o">=</span> <span class="kc">False</span>
|
|
|
|
|
|
<div class="viewcode-block" id="DDPGAlgorithmParameters"><a class="viewcode-back" href="../../../components/agents/policy_optimization/ddpg.html#rl_coach.agents.ddpg_agent.DDPGAlgorithmParameters">[docs]</a><span class="k">class</span> <span class="nc">DDPGAlgorithmParameters</span><span class="p">(</span><span class="n">AlgorithmParameters</span><span class="p">):</span>
|
|
<span class="sd">"""</span>
|
|
<span class="sd"> :param num_steps_between_copying_online_weights_to_target: (StepMethod)</span>
|
|
<span class="sd"> The number of steps between copying the online network weights to the target network weights.</span>
|
|
|
|
<span class="sd"> :param rate_for_copying_weights_to_target: (float)</span>
|
|
<span class="sd"> When copying the online network weights to the target network weights, a soft update will be used, which</span>
|
|
<span class="sd"> weight the new online network weights by rate_for_copying_weights_to_target</span>
|
|
|
|
<span class="sd"> :param num_consecutive_playing_steps: (StepMethod)</span>
|
|
<span class="sd"> The number of consecutive steps to act between every two training iterations</span>
|
|
|
|
<span class="sd"> :param use_target_network_for_evaluation: (bool)</span>
|
|
<span class="sd"> If set to True, the target network will be used for predicting the actions when choosing actions to act.</span>
|
|
<span class="sd"> Since the target network weights change more slowly, the predicted actions will be more consistent.</span>
|
|
|
|
<span class="sd"> :param action_penalty: (float)</span>
|
|
<span class="sd"> The amount by which to penalize the network on high action feature (pre-activation) values.</span>
|
|
<span class="sd"> This can prevent the actions features from saturating the TanH activation function, and therefore prevent the</span>
|
|
<span class="sd"> gradients from becoming very low.</span>
|
|
|
|
<span class="sd"> :param clip_critic_targets: (Tuple[float, float] or None)</span>
|
|
<span class="sd"> The range to clip the critic target to in order to prevent overestimation of the action values.</span>
|
|
|
|
<span class="sd"> :param use_non_zero_discount_for_terminal_states: (bool)</span>
|
|
<span class="sd"> If set to True, the discount factor will be used for terminal states to bootstrap the next predicted state</span>
|
|
<span class="sd"> values. If set to False, the terminal states reward will be taken as the target return for the network.</span>
|
|
<span class="sd"> """</span>
|
|
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
|
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">num_steps_between_copying_online_weights_to_target</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="bp">self</span><span class="o">.</span><span class="n">rate_for_copying_weights_to_target</span> <span class="o">=</span> <span class="mf">0.001</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">num_consecutive_playing_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="bp">self</span><span class="o">.</span><span class="n">use_target_network_for_evaluation</span> <span class="o">=</span> <span class="kc">False</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">action_penalty</span> <span class="o">=</span> <span class="mi">0</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">clip_critic_targets</span> <span class="o">=</span> <span class="kc">None</span> <span class="c1"># expected to be a tuple of the form (min_clip_value, max_clip_value) or None</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">use_non_zero_discount_for_terminal_states</span> <span class="o">=</span> <span class="kc">False</span></div>
|
|
|
|
|
|
<span class="k">class</span> <span class="nc">DDPGAgentParameters</span><span class="p">(</span><span class="n">AgentParameters</span><span class="p">):</span>
|
|
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
|
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">algorithm</span><span class="o">=</span><span class="n">DDPGAlgorithmParameters</span><span class="p">(),</span>
|
|
<span class="n">exploration</span><span class="o">=</span><span class="n">OUProcessParameters</span><span class="p">(),</span>
|
|
<span class="n">memory</span><span class="o">=</span><span class="n">EpisodicExperienceReplayParameters</span><span class="p">(),</span>
|
|
<span class="n">networks</span><span class="o">=</span><span class="n">OrderedDict</span><span class="p">([(</span><span class="s2">"actor"</span><span class="p">,</span> <span class="n">DDPGActorNetworkParameters</span><span class="p">()),</span>
|
|
<span class="p">(</span><span class="s2">"critic"</span><span class="p">,</span> <span class="n">DDPGCriticNetworkParameters</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.agents.ddpg_agent:DDPGAgent'</span>
|
|
|
|
|
|
<span class="c1"># Deep Deterministic Policy Gradients Network - https://arxiv.org/pdf/1509.02971.pdf</span>
|
|
<span class="k">class</span> <span class="nc">DDPGAgent</span><span class="p">(</span><span class="n">ActorCriticAgent</span><span class="p">):</span>
|
|
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">agent_parameters</span><span class="p">,</span> <span class="n">parent</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="s1">'LevelManager'</span><span class="p">,</span> <span class="s1">'CompositeAgent'</span><span class="p">]</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">agent_parameters</span><span class="p">,</span> <span class="n">parent</span><span class="p">)</span>
|
|
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">q_values</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">register_signal</span><span class="p">(</span><span class="s2">"Q"</span><span class="p">)</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">TD_targets_signal</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">register_signal</span><span class="p">(</span><span class="s2">"TD targets"</span><span class="p">)</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">action_signal</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">register_signal</span><span class="p">(</span><span class="s2">"actions"</span><span class="p">)</span>
|
|
|
|
<span class="k">def</span> <span class="nf">learn_from_batch</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">batch</span><span class="p">):</span>
|
|
<span class="n">actor</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">networks</span><span class="p">[</span><span class="s1">'actor'</span><span class="p">]</span>
|
|
<span class="n">critic</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">networks</span><span class="p">[</span><span class="s1">'critic'</span><span class="p">]</span>
|
|
|
|
<span class="n">actor_keys</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ap</span><span class="o">.</span><span class="n">network_wrappers</span><span class="p">[</span><span class="s1">'actor'</span><span class="p">]</span><span class="o">.</span><span class="n">input_embedders_parameters</span><span class="o">.</span><span class="n">keys</span><span class="p">()</span>
|
|
<span class="n">critic_keys</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ap</span><span class="o">.</span><span class="n">network_wrappers</span><span class="p">[</span><span class="s1">'critic'</span><span class="p">]</span><span class="o">.</span><span class="n">input_embedders_parameters</span><span class="o">.</span><span class="n">keys</span><span class="p">()</span>
|
|
|
|
<span class="c1"># TD error = r + discount*max(q_st_plus_1) - q_st</span>
|
|
<span class="n">next_actions</span><span class="p">,</span> <span class="n">actions_mean</span> <span class="o">=</span> <span class="n">actor</span><span class="o">.</span><span class="n">parallel_prediction</span><span class="p">([</span>
|
|
<span class="p">(</span><span class="n">actor</span><span class="o">.</span><span class="n">target_network</span><span class="p">,</span> <span class="n">batch</span><span class="o">.</span><span class="n">next_states</span><span class="p">(</span><span class="n">actor_keys</span><span class="p">)),</span>
|
|
<span class="p">(</span><span class="n">actor</span><span class="o">.</span><span class="n">online_network</span><span class="p">,</span> <span class="n">batch</span><span class="o">.</span><span class="n">states</span><span class="p">(</span><span class="n">actor_keys</span><span class="p">))</span>
|
|
<span class="p">])</span>
|
|
|
|
<span class="n">critic_inputs</span> <span class="o">=</span> <span class="n">copy</span><span class="o">.</span><span class="n">copy</span><span class="p">(</span><span class="n">batch</span><span class="o">.</span><span class="n">next_states</span><span class="p">(</span><span class="n">critic_keys</span><span class="p">))</span>
|
|
<span class="n">critic_inputs</span><span class="p">[</span><span class="s1">'action'</span><span class="p">]</span> <span class="o">=</span> <span class="n">next_actions</span>
|
|
<span class="n">q_st_plus_1</span> <span class="o">=</span> <span class="n">critic</span><span class="o">.</span><span class="n">target_network</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">critic_inputs</span><span class="p">)</span>
|
|
|
|
<span class="c1"># calculate the bootstrapped TD targets while discounting terminal states according to</span>
|
|
<span class="c1"># use_non_zero_discount_for_terminal_states</span>
|
|
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">ap</span><span class="o">.</span><span class="n">algorithm</span><span class="o">.</span><span class="n">use_non_zero_discount_for_terminal_states</span><span class="p">:</span>
|
|
<span class="n">TD_targets</span> <span class="o">=</span> <span class="n">batch</span><span class="o">.</span><span class="n">rewards</span><span class="p">(</span><span class="n">expand_dims</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">ap</span><span class="o">.</span><span class="n">algorithm</span><span class="o">.</span><span class="n">discount</span> <span class="o">*</span> <span class="n">q_st_plus_1</span>
|
|
<span class="k">else</span><span class="p">:</span>
|
|
<span class="n">TD_targets</span> <span class="o">=</span> <span class="n">batch</span><span class="o">.</span><span class="n">rewards</span><span class="p">(</span><span class="n">expand_dims</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span> <span class="o">+</span> \
|
|
<span class="p">(</span><span class="mf">1.0</span> <span class="o">-</span> <span class="n">batch</span><span class="o">.</span><span class="n">game_overs</span><span class="p">(</span><span class="n">expand_dims</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">ap</span><span class="o">.</span><span class="n">algorithm</span><span class="o">.</span><span class="n">discount</span> <span class="o">*</span> <span class="n">q_st_plus_1</span>
|
|
|
|
<span class="c1"># clip the TD targets to prevent overestimation errors</span>
|
|
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">ap</span><span class="o">.</span><span class="n">algorithm</span><span class="o">.</span><span class="n">clip_critic_targets</span><span class="p">:</span>
|
|
<span class="n">TD_targets</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">clip</span><span class="p">(</span><span class="n">TD_targets</span><span class="p">,</span> <span class="o">*</span><span class="bp">self</span><span class="o">.</span><span class="n">ap</span><span class="o">.</span><span class="n">algorithm</span><span class="o">.</span><span class="n">clip_critic_targets</span><span class="p">)</span>
|
|
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">TD_targets_signal</span><span class="o">.</span><span class="n">add_sample</span><span class="p">(</span><span class="n">TD_targets</span><span class="p">)</span>
|
|
|
|
<span class="c1"># get the gradients of the critic output with respect to the action</span>
|
|
<span class="n">critic_inputs</span> <span class="o">=</span> <span class="n">copy</span><span class="o">.</span><span class="n">copy</span><span class="p">(</span><span class="n">batch</span><span class="o">.</span><span class="n">states</span><span class="p">(</span><span class="n">critic_keys</span><span class="p">))</span>
|
|
<span class="n">critic_inputs</span><span class="p">[</span><span class="s1">'action'</span><span class="p">]</span> <span class="o">=</span> <span class="n">actions_mean</span>
|
|
<span class="n">action_gradients</span> <span class="o">=</span> <span class="n">critic</span><span class="o">.</span><span class="n">online_network</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">critic_inputs</span><span class="p">,</span>
|
|
<span class="n">outputs</span><span class="o">=</span><span class="n">critic</span><span class="o">.</span><span class="n">online_network</span><span class="o">.</span><span class="n">gradients_wrt_inputs</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="s1">'action'</span><span class="p">])</span>
|
|
|
|
<span class="c1"># train the critic</span>
|
|
<span class="n">critic_inputs</span> <span class="o">=</span> <span class="n">copy</span><span class="o">.</span><span class="n">copy</span><span class="p">(</span><span class="n">batch</span><span class="o">.</span><span class="n">states</span><span class="p">(</span><span class="n">critic_keys</span><span class="p">))</span>
|
|
<span class="n">critic_inputs</span><span class="p">[</span><span class="s1">'action'</span><span class="p">]</span> <span class="o">=</span> <span class="n">batch</span><span class="o">.</span><span class="n">actions</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">batch</span><span class="o">.</span><span class="n">actions</span><span class="p">()</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">)</span>
|
|
<span class="n">result</span> <span class="o">=</span> <span class="n">critic</span><span class="o">.</span><span class="n">train_and_sync_networks</span><span class="p">(</span><span class="n">critic_inputs</span><span class="p">,</span> <span class="n">TD_targets</span><span class="p">)</span>
|
|
<span class="n">total_loss</span><span class="p">,</span> <span class="n">losses</span><span class="p">,</span> <span class="n">unclipped_grads</span> <span class="o">=</span> <span class="n">result</span><span class="p">[:</span><span class="mi">3</span><span class="p">]</span>
|
|
|
|
<span class="c1"># apply the gradients from the critic to the actor</span>
|
|
<span class="n">initial_feed_dict</span> <span class="o">=</span> <span class="p">{</span><span class="n">actor</span><span class="o">.</span><span class="n">online_network</span><span class="o">.</span><span class="n">gradients_weights_ph</span><span class="p">[</span><span class="mi">0</span><span class="p">]:</span> <span class="o">-</span><span class="n">action_gradients</span><span class="p">}</span>
|
|
<span class="n">gradients</span> <span class="o">=</span> <span class="n">actor</span><span class="o">.</span><span class="n">online_network</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">batch</span><span class="o">.</span><span class="n">states</span><span class="p">(</span><span class="n">actor_keys</span><span class="p">),</span>
|
|
<span class="n">outputs</span><span class="o">=</span><span class="n">actor</span><span class="o">.</span><span class="n">online_network</span><span class="o">.</span><span class="n">weighted_gradients</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span>
|
|
<span class="n">initial_feed_dict</span><span class="o">=</span><span class="n">initial_feed_dict</span><span class="p">)</span>
|
|
|
|
<span class="k">if</span> <span class="n">actor</span><span class="o">.</span><span class="n">has_global</span><span class="p">:</span>
|
|
<span class="n">actor</span><span class="o">.</span><span class="n">apply_gradients_to_global_network</span><span class="p">(</span><span class="n">gradients</span><span class="p">)</span>
|
|
<span class="n">actor</span><span class="o">.</span><span class="n">update_online_network</span><span class="p">()</span>
|
|
<span class="k">else</span><span class="p">:</span>
|
|
<span class="n">actor</span><span class="o">.</span><span class="n">apply_gradients_to_online_network</span><span class="p">(</span><span class="n">gradients</span><span class="p">)</span>
|
|
|
|
<span class="k">return</span> <span class="n">total_loss</span><span class="p">,</span> <span class="n">losses</span><span class="p">,</span> <span class="n">unclipped_grads</span>
|
|
|
|
<span class="k">def</span> <span class="nf">train</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
|
<span class="k">return</span> <span class="n">Agent</span><span class="o">.</span><span class="n">train</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
|
|
|
|
<span class="k">def</span> <span class="nf">choose_action</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">curr_state</span><span class="p">):</span>
|
|
<span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">spaces</span><span class="o">.</span><span class="n">action</span><span class="p">,</span> <span class="n">BoxActionSpace</span><span class="p">)</span> <span class="ow">or</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">spaces</span><span class="o">.</span><span class="n">action</span><span class="p">,</span> <span class="n">GoalsSpace</span><span class="p">)):</span>
|
|
<span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"DDPG works only for continuous control problems"</span><span class="p">)</span>
|
|
<span class="c1"># convert to batch so we can run it through the network</span>
|
|
<span class="n">tf_input_state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">prepare_batch_for_inference</span><span class="p">(</span><span class="n">curr_state</span><span class="p">,</span> <span class="s1">'actor'</span><span class="p">)</span>
|
|
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">ap</span><span class="o">.</span><span class="n">algorithm</span><span class="o">.</span><span class="n">use_target_network_for_evaluation</span><span class="p">:</span>
|
|
<span class="n">actor_network</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">networks</span><span class="p">[</span><span class="s1">'actor'</span><span class="p">]</span><span class="o">.</span><span class="n">target_network</span>
|
|
<span class="k">else</span><span class="p">:</span>
|
|
<span class="n">actor_network</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">networks</span><span class="p">[</span><span class="s1">'actor'</span><span class="p">]</span><span class="o">.</span><span class="n">online_network</span>
|
|
|
|
<span class="n">action_values</span> <span class="o">=</span> <span class="n">actor_network</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">tf_input_state</span><span class="p">)</span><span class="o">.</span><span class="n">squeeze</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">exploration_policy</span><span class="o">.</span><span class="n">get_action</span><span class="p">(</span><span class="n">action_values</span><span class="p">)</span>
|
|
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">action_signal</span><span class="o">.</span><span class="n">add_sample</span><span class="p">(</span><span class="n">action</span><span class="p">)</span>
|
|
|
|
<span class="c1"># get q value</span>
|
|
<span class="n">tf_input_state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">prepare_batch_for_inference</span><span class="p">(</span><span class="n">curr_state</span><span class="p">,</span> <span class="s1">'critic'</span><span class="p">)</span>
|
|
<span class="n">action_batch</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">expand_dims</span><span class="p">(</span><span class="n">action</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
|
|
<span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="n">action</span><span class="p">)</span> <span class="o">!=</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">:</span>
|
|
<span class="n">action_batch</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">action</span><span class="p">]])</span>
|
|
<span class="n">tf_input_state</span><span class="p">[</span><span class="s1">'action'</span><span class="p">]</span> <span class="o">=</span> <span class="n">action_batch</span>
|
|
<span class="n">q_value</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">networks</span><span class="p">[</span><span class="s1">'critic'</span><span class="p">]</span><span class="o">.</span><span class="n">online_network</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">tf_input_state</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">q_values</span><span class="o">.</span><span class="n">add_sample</span><span class="p">(</span><span class="n">q_value</span><span class="p">)</span>
|
|
|
|
<span class="n">action_info</span> <span class="o">=</span> <span class="n">ActionInfo</span><span class="p">(</span><span class="n">action</span><span class="o">=</span><span class="n">action</span><span class="p">,</span>
|
|
<span class="n">action_value</span><span class="o">=</span><span class="n">q_value</span><span class="p">)</span>
|
|
|
|
<span class="k">return</span> <span class="n">action_info</span>
|
|
</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> |