mirror of
https://github.com/labmlai/annotated_deep_learning_paper_implementations.git
synced 2025-08-17 03:14:55 +08:00
1322 lines
95 KiB
HTML
1322 lines
95 KiB
HTML
<!DOCTYPE html>
|
|
<html>
|
|
<head>
|
|
<meta http-equiv="content-type" content="text/html;charset=utf-8"/>
|
|
<meta name="viewport" content="width=device-width, initial-scale=1.0"/>
|
|
<meta name="description" content="Annotated implementation to train a PPO agent on Atari Breakout game."/>
|
|
|
|
<meta name="twitter:card" content="summary"/>
|
|
<meta name="twitter:image:src" content="https://avatars1.githubusercontent.com/u/64068543?s=400&v=4"/>
|
|
<meta name="twitter:title" content="PPO Experiment with Atari Breakout"/>
|
|
<meta name="twitter:description" content="Annotated implementation to train a PPO agent on Atari Breakout game."/>
|
|
<meta name="twitter:site" content="@labmlai"/>
|
|
<meta name="twitter:creator" content="@labmlai"/>
|
|
|
|
<meta property="og:url" content="https://nn.labml.ai/rl/ppo/experiment.html"/>
|
|
<meta property="og:title" content="PPO Experiment with Atari Breakout"/>
|
|
<meta property="og:image" content="https://avatars1.githubusercontent.com/u/64068543?s=400&v=4"/>
|
|
<meta property="og:site_name" content="LabML Neural Networks"/>
|
|
<meta property="og:type" content="object"/>
|
|
<meta property="og:title" content="PPO Experiment with Atari Breakout"/>
|
|
<meta property="og:description" content="Annotated implementation to train a PPO agent on Atari Breakout game."/>
|
|
|
|
<title>PPO Experiment with Atari Breakout</title>
|
|
<link rel="shortcut icon" href="/icon.png"/>
|
|
<link rel="stylesheet" href="../../pylit.css">
|
|
<link rel="canonical" href="https://nn.labml.ai/rl/ppo/experiment.html"/>
|
|
<!-- Global site tag (gtag.js) - Google Analytics -->
|
|
<script async src="https://www.googletagmanager.com/gtag/js?id=G-4V3HC8HBLH"></script>
|
|
<script>
|
|
window.dataLayer = window.dataLayer || [];
|
|
|
|
function gtag() {
|
|
dataLayer.push(arguments);
|
|
}
|
|
|
|
gtag('js', new Date());
|
|
|
|
gtag('config', 'G-4V3HC8HBLH');
|
|
</script>
|
|
</head>
|
|
<body>
|
|
<div id='container'>
|
|
<div id="background"></div>
|
|
<div class='section'>
|
|
<div class='docs'>
|
|
<p>
|
|
<a class="parent" href="/">home</a>
|
|
<a class="parent" href="../index.html">rl</a>
|
|
<a class="parent" href="index.html">ppo</a>
|
|
</p>
|
|
<p>
|
|
|
|
<a href="https://github.com/labmlai/annotated_deep_learning_paper_implementations/tree/master/labml_nn/rl/ppo/experiment.py">
|
|
<img alt="Github"
|
|
src="https://img.shields.io/github/stars/labmlai/annotated_deep_learning_paper_implementations?style=social"
|
|
style="max-width:100%;"/></a>
|
|
<a href="https://twitter.com/labmlai"
|
|
rel="nofollow">
|
|
<img alt="Twitter"
|
|
src="https://img.shields.io/twitter/follow/labmlai?style=social"
|
|
style="max-width:100%;"/></a>
|
|
</p>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-0'>
|
|
<div class='docs doc-strings'>
|
|
<div class='section-link'>
|
|
<a href='#section-0'>#</a>
|
|
</div>
|
|
<h1>PPO Experiment with Atari Breakout</h1>
|
|
<p>This experiment trains Proximal Policy Optimization (PPO) agent Atari Breakout game on OpenAI Gym.
|
|
It runs the <a href="../game.html">game environments on multiple processes</a> to sample efficiently.</p>
|
|
<p><a href="https://colab.research.google.com/github/labmlai/annotated_deep_learning_paper_implementations/blob/master/labml_nn/rl/ppo/experiment.ipynb"><img alt="Open In Colab" src="https://colab.research.google.com/assets/colab-badge.svg" /></a>
|
|
<a href="https://app.labml.ai/run/6eff28a0910e11eb9b008db315936e2f"><img alt="View Run" src="https://img.shields.io/badge/labml-experiment-brightgreen" /></a></p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">16</span><span></span><span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Dict</span>
|
|
<span class="lineno">17</span>
|
|
<span class="lineno">18</span><span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
|
|
<span class="lineno">19</span><span class="kn">import</span> <span class="nn">torch</span>
|
|
<span class="lineno">20</span><span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span>
|
|
<span class="lineno">21</span><span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">optim</span>
|
|
<span class="lineno">22</span><span class="kn">from</span> <span class="nn">torch.distributions</span> <span class="kn">import</span> <span class="n">Categorical</span>
|
|
<span class="lineno">23</span>
|
|
<span class="lineno">24</span><span class="kn">from</span> <span class="nn">labml</span> <span class="kn">import</span> <span class="n">monit</span><span class="p">,</span> <span class="n">tracker</span><span class="p">,</span> <span class="n">logger</span><span class="p">,</span> <span class="n">experiment</span>
|
|
<span class="lineno">25</span><span class="kn">from</span> <span class="nn">labml.configs</span> <span class="kn">import</span> <span class="n">FloatDynamicHyperParam</span><span class="p">,</span> <span class="n">IntDynamicHyperParam</span>
|
|
<span class="lineno">26</span><span class="kn">from</span> <span class="nn">labml_helpers.module</span> <span class="kn">import</span> <span class="n">Module</span>
|
|
<span class="lineno">27</span><span class="kn">from</span> <span class="nn">labml_nn.rl.game</span> <span class="kn">import</span> <span class="n">Worker</span>
|
|
<span class="lineno">28</span><span class="kn">from</span> <span class="nn">labml_nn.rl.ppo</span> <span class="kn">import</span> <span class="n">ClippedPPOLoss</span><span class="p">,</span> <span class="n">ClippedValueFunctionLoss</span>
|
|
<span class="lineno">29</span><span class="kn">from</span> <span class="nn">labml_nn.rl.ppo.gae</span> <span class="kn">import</span> <span class="n">GAE</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-1'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-1'>#</a>
|
|
</div>
|
|
<p>Select device</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">32</span><span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">is_available</span><span class="p">():</span>
|
|
<span class="lineno">33</span> <span class="n">device</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">device</span><span class="p">(</span><span class="s2">"cuda:0"</span><span class="p">)</span>
|
|
<span class="lineno">34</span><span class="k">else</span><span class="p">:</span>
|
|
<span class="lineno">35</span> <span class="n">device</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">device</span><span class="p">(</span><span class="s2">"cpu"</span><span class="p">)</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-2'>
|
|
<div class='docs doc-strings'>
|
|
<div class='section-link'>
|
|
<a href='#section-2'>#</a>
|
|
</div>
|
|
<h2>Model</h2>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">38</span><span class="k">class</span> <span class="nc">Model</span><span class="p">(</span><span class="n">Module</span><span class="p">):</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-3'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-3'>#</a>
|
|
</div>
|
|
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">43</span> <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
|
<span class="lineno">44</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-4'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-4'>#</a>
|
|
</div>
|
|
<p>The first convolution layer takes a
|
|
84x84 frame and produces a 20x20 frame</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">48</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span> <span class="n">out_channels</span><span class="o">=</span><span class="mi">32</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">8</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-5'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-5'>#</a>
|
|
</div>
|
|
<p>The second convolution layer takes a
|
|
20x20 frame and produces a 9x9 frame</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">52</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv2</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="o">=</span><span class="mi">32</span><span class="p">,</span> <span class="n">out_channels</span><span class="o">=</span><span class="mi">64</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-6'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-6'>#</a>
|
|
</div>
|
|
<p>The third convolution layer takes a
|
|
9x9 frame and produces a 7x7 frame</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">56</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv3</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="o">=</span><span class="mi">64</span><span class="p">,</span> <span class="n">out_channels</span><span class="o">=</span><span class="mi">64</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-7'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-7'>#</a>
|
|
</div>
|
|
<p>A fully connected layer takes the flattened
|
|
frame from third convolution layer, and outputs
|
|
512 features</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">61</span> <span class="bp">self</span><span class="o">.</span><span class="n">lin</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">in_features</span><span class="o">=</span><span class="mi">7</span> <span class="o">*</span> <span class="mi">7</span> <span class="o">*</span> <span class="mi">64</span><span class="p">,</span> <span class="n">out_features</span><span class="o">=</span><span class="mi">512</span><span class="p">)</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-8'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-8'>#</a>
|
|
</div>
|
|
<p>A fully connected layer to get logits for $\pi$</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">64</span> <span class="bp">self</span><span class="o">.</span><span class="n">pi_logits</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">in_features</span><span class="o">=</span><span class="mi">512</span><span class="p">,</span> <span class="n">out_features</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-9'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-9'>#</a>
|
|
</div>
|
|
<p>A fully connected layer to get value function</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">67</span> <span class="bp">self</span><span class="o">.</span><span class="n">value</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">in_features</span><span class="o">=</span><span class="mi">512</span><span class="p">,</span> <span class="n">out_features</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-10'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-10'>#</a>
|
|
</div>
|
|
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">70</span> <span class="bp">self</span><span class="o">.</span><span class="n">activation</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">()</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-11'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-11'>#</a>
|
|
</div>
|
|
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">72</span> <span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">obs</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
|
|
<span class="lineno">73</span> <span class="n">h</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">activation</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">conv1</span><span class="p">(</span><span class="n">obs</span><span class="p">))</span>
|
|
<span class="lineno">74</span> <span class="n">h</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">activation</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">conv2</span><span class="p">(</span><span class="n">h</span><span class="p">))</span>
|
|
<span class="lineno">75</span> <span class="n">h</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">activation</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">conv3</span><span class="p">(</span><span class="n">h</span><span class="p">))</span>
|
|
<span class="lineno">76</span> <span class="n">h</span> <span class="o">=</span> <span class="n">h</span><span class="o">.</span><span class="n">reshape</span><span class="p">((</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">7</span> <span class="o">*</span> <span class="mi">7</span> <span class="o">*</span> <span class="mi">64</span><span class="p">))</span>
|
|
<span class="lineno">77</span>
|
|
<span class="lineno">78</span> <span class="n">h</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">activation</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">lin</span><span class="p">(</span><span class="n">h</span><span class="p">))</span>
|
|
<span class="lineno">79</span>
|
|
<span class="lineno">80</span> <span class="n">pi</span> <span class="o">=</span> <span class="n">Categorical</span><span class="p">(</span><span class="n">logits</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">pi_logits</span><span class="p">(</span><span class="n">h</span><span class="p">))</span>
|
|
<span class="lineno">81</span> <span class="n">value</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">value</span><span class="p">(</span><span class="n">h</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span>
|
|
<span class="lineno">82</span>
|
|
<span class="lineno">83</span> <span class="k">return</span> <span class="n">pi</span><span class="p">,</span> <span class="n">value</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-12'>
|
|
<div class='docs doc-strings'>
|
|
<div class='section-link'>
|
|
<a href='#section-12'>#</a>
|
|
</div>
|
|
<p>Scale observations from <code>[0, 255]</code> to <code>[0, 1]</code></p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">86</span><span class="k">def</span> <span class="nf">obs_to_torch</span><span class="p">(</span><span class="n">obs</span><span class="p">:</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">)</span> <span class="o">-></span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-13'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-13'>#</a>
|
|
</div>
|
|
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">88</span> <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">obs</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">)</span> <span class="o">/</span> <span class="mf">255.</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-14'>
|
|
<div class='docs doc-strings'>
|
|
<div class='section-link'>
|
|
<a href='#section-14'>#</a>
|
|
</div>
|
|
<h2>Trainer</h2>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">91</span><span class="k">class</span> <span class="nc">Trainer</span><span class="p">:</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-15'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-15'>#</a>
|
|
</div>
|
|
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">96</span> <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span>
|
|
<span class="lineno">97</span> <span class="n">updates</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">epochs</span><span class="p">:</span> <span class="n">IntDynamicHyperParam</span><span class="p">,</span>
|
|
<span class="lineno">98</span> <span class="n">n_workers</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">worker_steps</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">batches</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
|
|
<span class="lineno">99</span> <span class="n">value_loss_coef</span><span class="p">:</span> <span class="n">FloatDynamicHyperParam</span><span class="p">,</span>
|
|
<span class="lineno">100</span> <span class="n">entropy_bonus_coef</span><span class="p">:</span> <span class="n">FloatDynamicHyperParam</span><span class="p">,</span>
|
|
<span class="lineno">101</span> <span class="n">clip_range</span><span class="p">:</span> <span class="n">FloatDynamicHyperParam</span><span class="p">,</span>
|
|
<span class="lineno">102</span> <span class="n">learning_rate</span><span class="p">:</span> <span class="n">FloatDynamicHyperParam</span><span class="p">,</span>
|
|
<span class="lineno">103</span> <span class="p">):</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-16'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-16'>#</a>
|
|
</div>
|
|
<h4>Configurations</h4>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-17'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-17'>#</a>
|
|
</div>
|
|
<p>number of updates</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">107</span> <span class="bp">self</span><span class="o">.</span><span class="n">updates</span> <span class="o">=</span> <span class="n">updates</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-18'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-18'>#</a>
|
|
</div>
|
|
<p>number of epochs to train the model with sampled data</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">109</span> <span class="bp">self</span><span class="o">.</span><span class="n">epochs</span> <span class="o">=</span> <span class="n">epochs</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-19'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-19'>#</a>
|
|
</div>
|
|
<p>number of worker processes</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">111</span> <span class="bp">self</span><span class="o">.</span><span class="n">n_workers</span> <span class="o">=</span> <span class="n">n_workers</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-20'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-20'>#</a>
|
|
</div>
|
|
<p>number of steps to run on each process for a single update</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">113</span> <span class="bp">self</span><span class="o">.</span><span class="n">worker_steps</span> <span class="o">=</span> <span class="n">worker_steps</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-21'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-21'>#</a>
|
|
</div>
|
|
<p>number of mini batches</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">115</span> <span class="bp">self</span><span class="o">.</span><span class="n">batches</span> <span class="o">=</span> <span class="n">batches</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-22'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-22'>#</a>
|
|
</div>
|
|
<p>total number of samples for a single update</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">117</span> <span class="bp">self</span><span class="o">.</span><span class="n">batch_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">n_workers</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">worker_steps</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-23'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-23'>#</a>
|
|
</div>
|
|
<p>size of a mini batch</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">119</span> <span class="bp">self</span><span class="o">.</span><span class="n">mini_batch_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">batch_size</span> <span class="o">//</span> <span class="bp">self</span><span class="o">.</span><span class="n">batches</span>
|
|
<span class="lineno">120</span> <span class="k">assert</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">batch_size</span> <span class="o">%</span> <span class="bp">self</span><span class="o">.</span><span class="n">batches</span> <span class="o">==</span> <span class="mi">0</span><span class="p">)</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-24'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-24'>#</a>
|
|
</div>
|
|
<p>Value loss coefficient</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">123</span> <span class="bp">self</span><span class="o">.</span><span class="n">value_loss_coef</span> <span class="o">=</span> <span class="n">value_loss_coef</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-25'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-25'>#</a>
|
|
</div>
|
|
<p>Entropy bonus coefficient</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">125</span> <span class="bp">self</span><span class="o">.</span><span class="n">entropy_bonus_coef</span> <span class="o">=</span> <span class="n">entropy_bonus_coef</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-26'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-26'>#</a>
|
|
</div>
|
|
<p>Clipping range</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">128</span> <span class="bp">self</span><span class="o">.</span><span class="n">clip_range</span> <span class="o">=</span> <span class="n">clip_range</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-27'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-27'>#</a>
|
|
</div>
|
|
<p>Learning rate</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">130</span> <span class="bp">self</span><span class="o">.</span><span class="n">learning_rate</span> <span class="o">=</span> <span class="n">learning_rate</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-28'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-28'>#</a>
|
|
</div>
|
|
<h4>Initialize</h4>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-29'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-29'>#</a>
|
|
</div>
|
|
<p>create workers</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">135</span> <span class="bp">self</span><span class="o">.</span><span class="n">workers</span> <span class="o">=</span> <span class="p">[</span><span class="n">Worker</span><span class="p">(</span><span class="mi">47</span> <span class="o">+</span> <span class="n">i</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">n_workers</span><span class="p">)]</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-30'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-30'>#</a>
|
|
</div>
|
|
<p>initialize tensors for observations</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">138</span> <span class="bp">self</span><span class="o">.</span><span class="n">obs</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">zeros</span><span class="p">((</span><span class="bp">self</span><span class="o">.</span><span class="n">n_workers</span><span class="p">,</span> <span class="mi">4</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="n">dtype</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">uint8</span><span class="p">)</span>
|
|
<span class="lineno">139</span> <span class="k">for</span> <span class="n">worker</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">workers</span><span class="p">:</span>
|
|
<span class="lineno">140</span> <span class="n">worker</span><span class="o">.</span><span class="n">child</span><span class="o">.</span><span class="n">send</span><span class="p">((</span><span class="s2">"reset"</span><span class="p">,</span> <span class="kc">None</span><span class="p">))</span>
|
|
<span class="lineno">141</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">worker</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">workers</span><span class="p">):</span>
|
|
<span class="lineno">142</span> <span class="bp">self</span><span class="o">.</span><span class="n">obs</span><span class="p">[</span><span class="n">i</span><span class="p">]</span> <span class="o">=</span> <span class="n">worker</span><span class="o">.</span><span class="n">child</span><span class="o">.</span><span class="n">recv</span><span class="p">()</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-31'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-31'>#</a>
|
|
</div>
|
|
<p>model</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">145</span> <span class="bp">self</span><span class="o">.</span><span class="n">model</span> <span class="o">=</span> <span class="n">Model</span><span class="p">()</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="p">)</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-32'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-32'>#</a>
|
|
</div>
|
|
<p>optimizer</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">148</span> <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span> <span class="o">=</span> <span class="n">optim</span><span class="o">.</span><span class="n">Adam</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">parameters</span><span class="p">(),</span> <span class="n">lr</span><span class="o">=</span><span class="mf">2.5e-4</span><span class="p">)</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-33'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-33'>#</a>
|
|
</div>
|
|
<p>GAE with $\gamma = 0.99$ and $\lambda = 0.95$</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">151</span> <span class="bp">self</span><span class="o">.</span><span class="n">gae</span> <span class="o">=</span> <span class="n">GAE</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">n_workers</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">worker_steps</span><span class="p">,</span> <span class="mf">0.99</span><span class="p">,</span> <span class="mf">0.95</span><span class="p">)</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-34'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-34'>#</a>
|
|
</div>
|
|
<p>PPO Loss</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">154</span> <span class="bp">self</span><span class="o">.</span><span class="n">ppo_loss</span> <span class="o">=</span> <span class="n">ClippedPPOLoss</span><span class="p">()</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-35'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-35'>#</a>
|
|
</div>
|
|
<p>Value Loss</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">157</span> <span class="bp">self</span><span class="o">.</span><span class="n">value_loss</span> <span class="o">=</span> <span class="n">ClippedValueFunctionLoss</span><span class="p">()</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-36'>
|
|
<div class='docs doc-strings'>
|
|
<div class='section-link'>
|
|
<a href='#section-36'>#</a>
|
|
</div>
|
|
<h3>Sample data with current policy</h3>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">159</span> <span class="k">def</span> <span class="nf">sample</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-></span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]:</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-37'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-37'>#</a>
|
|
</div>
|
|
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">164</span> <span class="n">rewards</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">zeros</span><span class="p">((</span><span class="bp">self</span><span class="o">.</span><span class="n">n_workers</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">worker_steps</span><span class="p">),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
|
|
<span class="lineno">165</span> <span class="n">actions</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">zeros</span><span class="p">((</span><span class="bp">self</span><span class="o">.</span><span class="n">n_workers</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">worker_steps</span><span class="p">),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">int32</span><span class="p">)</span>
|
|
<span class="lineno">166</span> <span class="n">done</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">zeros</span><span class="p">((</span><span class="bp">self</span><span class="o">.</span><span class="n">n_workers</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">worker_steps</span><span class="p">),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">bool</span><span class="p">)</span>
|
|
<span class="lineno">167</span> <span class="n">obs</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">zeros</span><span class="p">((</span><span class="bp">self</span><span class="o">.</span><span class="n">n_workers</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">worker_steps</span><span class="p">,</span> <span class="mi">4</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="n">dtype</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">uint8</span><span class="p">)</span>
|
|
<span class="lineno">168</span> <span class="n">log_pis</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">zeros</span><span class="p">((</span><span class="bp">self</span><span class="o">.</span><span class="n">n_workers</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">worker_steps</span><span class="p">),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
|
|
<span class="lineno">169</span> <span class="n">values</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">zeros</span><span class="p">((</span><span class="bp">self</span><span class="o">.</span><span class="n">n_workers</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">worker_steps</span> <span class="o">+</span> <span class="mi">1</span><span class="p">),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
|
|
<span class="lineno">170</span>
|
|
<span class="lineno">171</span> <span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">():</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-38'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-38'>#</a>
|
|
</div>
|
|
<p>sample <code>worker_steps</code> from each worker</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">173</span> <span class="k">for</span> <span class="n">t</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">worker_steps</span><span class="p">):</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-39'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-39'>#</a>
|
|
</div>
|
|
<p><code>self.obs</code> keeps track of the last observation from each worker,
|
|
which is the input for the model to sample the next action</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">176</span> <span class="n">obs</span><span class="p">[:,</span> <span class="n">t</span><span class="p">]</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">obs</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-40'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-40'>#</a>
|
|
</div>
|
|
<p>sample actions from $\pi_{\theta_{OLD}}$ for each worker;
|
|
this returns arrays of size <code>n_workers</code></p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">179</span> <span class="n">pi</span><span class="p">,</span> <span class="n">v</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="p">(</span><span class="n">obs_to_torch</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">obs</span><span class="p">))</span>
|
|
<span class="lineno">180</span> <span class="n">values</span><span class="p">[:,</span> <span class="n">t</span><span class="p">]</span> <span class="o">=</span> <span class="n">v</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span>
|
|
<span class="lineno">181</span> <span class="n">a</span> <span class="o">=</span> <span class="n">pi</span><span class="o">.</span><span class="n">sample</span><span class="p">()</span>
|
|
<span class="lineno">182</span> <span class="n">actions</span><span class="p">[:,</span> <span class="n">t</span><span class="p">]</span> <span class="o">=</span> <span class="n">a</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span>
|
|
<span class="lineno">183</span> <span class="n">log_pis</span><span class="p">[:,</span> <span class="n">t</span><span class="p">]</span> <span class="o">=</span> <span class="n">pi</span><span class="o">.</span><span class="n">log_prob</span><span class="p">(</span><span class="n">a</span><span class="p">)</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-41'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-41'>#</a>
|
|
</div>
|
|
<p>run sampled actions on each worker</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">186</span> <span class="k">for</span> <span class="n">w</span><span class="p">,</span> <span class="n">worker</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">workers</span><span class="p">):</span>
|
|
<span class="lineno">187</span> <span class="n">worker</span><span class="o">.</span><span class="n">child</span><span class="o">.</span><span class="n">send</span><span class="p">((</span><span class="s2">"step"</span><span class="p">,</span> <span class="n">actions</span><span class="p">[</span><span class="n">w</span><span class="p">,</span> <span class="n">t</span><span class="p">]))</span>
|
|
<span class="lineno">188</span>
|
|
<span class="lineno">189</span> <span class="k">for</span> <span class="n">w</span><span class="p">,</span> <span class="n">worker</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">workers</span><span class="p">):</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-42'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-42'>#</a>
|
|
</div>
|
|
<p>get results after executing the actions</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">191</span> <span class="bp">self</span><span class="o">.</span><span class="n">obs</span><span class="p">[</span><span class="n">w</span><span class="p">],</span> <span class="n">rewards</span><span class="p">[</span><span class="n">w</span><span class="p">,</span> <span class="n">t</span><span class="p">],</span> <span class="n">done</span><span class="p">[</span><span class="n">w</span><span class="p">,</span> <span class="n">t</span><span class="p">],</span> <span class="n">info</span> <span class="o">=</span> <span class="n">worker</span><span class="o">.</span><span class="n">child</span><span class="o">.</span><span class="n">recv</span><span class="p">()</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-43'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-43'>#</a>
|
|
</div>
|
|
<p>collect episode info, which is available if an episode finished;
|
|
this includes total reward and length of the episode -
|
|
look at <code>Game</code> to see how it works.</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">196</span> <span class="k">if</span> <span class="n">info</span><span class="p">:</span>
|
|
<span class="lineno">197</span> <span class="n">tracker</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="s1">'reward'</span><span class="p">,</span> <span class="n">info</span><span class="p">[</span><span class="s1">'reward'</span><span class="p">])</span>
|
|
<span class="lineno">198</span> <span class="n">tracker</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="s1">'length'</span><span class="p">,</span> <span class="n">info</span><span class="p">[</span><span class="s1">'length'</span><span class="p">])</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-44'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-44'>#</a>
|
|
</div>
|
|
<p>Get value of after the final step</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">201</span> <span class="n">_</span><span class="p">,</span> <span class="n">v</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="p">(</span><span class="n">obs_to_torch</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">obs</span><span class="p">))</span>
|
|
<span class="lineno">202</span> <span class="n">values</span><span class="p">[:,</span> <span class="bp">self</span><span class="o">.</span><span class="n">worker_steps</span><span class="p">]</span> <span class="o">=</span> <span class="n">v</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-45'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-45'>#</a>
|
|
</div>
|
|
<p>calculate advantages</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">205</span> <span class="n">advantages</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">gae</span><span class="p">(</span><span class="n">done</span><span class="p">,</span> <span class="n">rewards</span><span class="p">,</span> <span class="n">values</span><span class="p">)</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-46'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-46'>#</a>
|
|
</div>
|
|
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">208</span> <span class="n">samples</span> <span class="o">=</span> <span class="p">{</span>
|
|
<span class="lineno">209</span> <span class="s1">'obs'</span><span class="p">:</span> <span class="n">obs</span><span class="p">,</span>
|
|
<span class="lineno">210</span> <span class="s1">'actions'</span><span class="p">:</span> <span class="n">actions</span><span class="p">,</span>
|
|
<span class="lineno">211</span> <span class="s1">'values'</span><span class="p">:</span> <span class="n">values</span><span class="p">[:,</span> <span class="p">:</span><span class="o">-</span><span class="mi">1</span><span class="p">],</span>
|
|
<span class="lineno">212</span> <span class="s1">'log_pis'</span><span class="p">:</span> <span class="n">log_pis</span><span class="p">,</span>
|
|
<span class="lineno">213</span> <span class="s1">'advantages'</span><span class="p">:</span> <span class="n">advantages</span>
|
|
<span class="lineno">214</span> <span class="p">}</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-47'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-47'>#</a>
|
|
</div>
|
|
<p>samples are currently in <code>[workers, time_step]</code> table,
|
|
we should flatten it for training</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">218</span> <span class="n">samples_flat</span> <span class="o">=</span> <span class="p">{}</span>
|
|
<span class="lineno">219</span> <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">samples</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
|
|
<span class="lineno">220</span> <span class="n">v</span> <span class="o">=</span> <span class="n">v</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">v</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">v</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">v</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">2</span><span class="p">:])</span>
|
|
<span class="lineno">221</span> <span class="k">if</span> <span class="n">k</span> <span class="o">==</span> <span class="s1">'obs'</span><span class="p">:</span>
|
|
<span class="lineno">222</span> <span class="n">samples_flat</span><span class="p">[</span><span class="n">k</span><span class="p">]</span> <span class="o">=</span> <span class="n">obs_to_torch</span><span class="p">(</span><span class="n">v</span><span class="p">)</span>
|
|
<span class="lineno">223</span> <span class="k">else</span><span class="p">:</span>
|
|
<span class="lineno">224</span> <span class="n">samples_flat</span><span class="p">[</span><span class="n">k</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">v</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">)</span>
|
|
<span class="lineno">225</span>
|
|
<span class="lineno">226</span> <span class="k">return</span> <span class="n">samples_flat</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-48'>
|
|
<div class='docs doc-strings'>
|
|
<div class='section-link'>
|
|
<a href='#section-48'>#</a>
|
|
</div>
|
|
<h3>Train the model based on samples</h3>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">228</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="n">samples</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">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]):</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-49'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-49'>#</a>
|
|
</div>
|
|
<p>It learns faster with a higher number of epochs,
|
|
but becomes a little unstable; that is,
|
|
the average episode reward does not monotonically increase
|
|
over time.
|
|
May be reducing the clipping range might solve it.</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">238</span> <span class="k">for</span> <span class="n">_</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">epochs</span><span class="p">()):</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-50'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-50'>#</a>
|
|
</div>
|
|
<p>shuffle for each epoch</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">240</span> <span class="n">indexes</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randperm</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">batch_size</span><span class="p">)</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-51'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-51'>#</a>
|
|
</div>
|
|
<p>for each mini batch</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">243</span> <span class="k">for</span> <span class="n">start</span> <span class="ow">in</span> <span class="nb">range</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">batch_size</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">mini_batch_size</span><span class="p">):</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-52'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-52'>#</a>
|
|
</div>
|
|
<p>get mini batch</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">245</span> <span class="n">end</span> <span class="o">=</span> <span class="n">start</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">mini_batch_size</span>
|
|
<span class="lineno">246</span> <span class="n">mini_batch_indexes</span> <span class="o">=</span> <span class="n">indexes</span><span class="p">[</span><span class="n">start</span><span class="p">:</span> <span class="n">end</span><span class="p">]</span>
|
|
<span class="lineno">247</span> <span class="n">mini_batch</span> <span class="o">=</span> <span class="p">{}</span>
|
|
<span class="lineno">248</span> <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">samples</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
|
|
<span class="lineno">249</span> <span class="n">mini_batch</span><span class="p">[</span><span class="n">k</span><span class="p">]</span> <span class="o">=</span> <span class="n">v</span><span class="p">[</span><span class="n">mini_batch_indexes</span><span class="p">]</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-53'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-53'>#</a>
|
|
</div>
|
|
<p>train</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">252</span> <span class="n">loss</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_calc_loss</span><span class="p">(</span><span class="n">mini_batch</span><span class="p">)</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-54'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-54'>#</a>
|
|
</div>
|
|
<p>Set learning rate</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">255</span> <span class="k">for</span> <span class="n">pg</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
|
|
<span class="lineno">256</span> <span class="n">pg</span><span class="p">[</span><span class="s1">'lr'</span><span class="p">]</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">learning_rate</span><span class="p">()</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-55'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-55'>#</a>
|
|
</div>
|
|
<p>Zero out the previously calculated gradients</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">258</span> <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="o">.</span><span class="n">zero_grad</span><span class="p">()</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-56'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-56'>#</a>
|
|
</div>
|
|
<p>Calculate gradients</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">260</span> <span class="n">loss</span><span class="o">.</span><span class="n">backward</span><span class="p">()</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-57'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-57'>#</a>
|
|
</div>
|
|
<p>Clip gradients</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">262</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">clip_grad_norm_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">parameters</span><span class="p">(),</span> <span class="n">max_norm</span><span class="o">=</span><span class="mf">0.5</span><span class="p">)</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-58'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-58'>#</a>
|
|
</div>
|
|
<p>Update parameters based on gradients</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">264</span> <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="o">.</span><span class="n">step</span><span class="p">()</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-59'>
|
|
<div class='docs doc-strings'>
|
|
<div class='section-link'>
|
|
<a href='#section-59'>#</a>
|
|
</div>
|
|
<h4>Normalize advantage function</h4>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">266</span> <span class="nd">@staticmethod</span>
|
|
<span class="lineno">267</span> <span class="k">def</span> <span class="nf">_normalize</span><span class="p">(</span><span class="n">adv</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-60'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-60'>#</a>
|
|
</div>
|
|
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">269</span> <span class="k">return</span> <span class="p">(</span><span class="n">adv</span> <span class="o">-</span> <span class="n">adv</span><span class="o">.</span><span class="n">mean</span><span class="p">())</span> <span class="o">/</span> <span class="p">(</span><span class="n">adv</span><span class="o">.</span><span class="n">std</span><span class="p">()</span> <span class="o">+</span> <span class="mf">1e-8</span><span class="p">)</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-61'>
|
|
<div class='docs doc-strings'>
|
|
<div class='section-link'>
|
|
<a href='#section-61'>#</a>
|
|
</div>
|
|
<h3>Calculate total loss</h3>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">271</span> <span class="k">def</span> <span class="nf">_calc_loss</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">samples</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">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">])</span> <span class="o">-></span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-62'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-62'>#</a>
|
|
</div>
|
|
<p>$R_t$ returns sampled from $\pi_{\theta_{OLD}}$</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">277</span> <span class="n">sampled_return</span> <span class="o">=</span> <span class="n">samples</span><span class="p">[</span><span class="s1">'values'</span><span class="p">]</span> <span class="o">+</span> <span class="n">samples</span><span class="p">[</span><span class="s1">'advantages'</span><span class="p">]</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-63'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-63'>#</a>
|
|
</div>
|
|
<p>$\bar{A_t} = \frac{\hat{A_t} - \mu(\hat{A_t})}{\sigma(\hat{A_t})}$,
|
|
where $\hat{A_t}$ is advantages sampled from $\pi_{\theta_{OLD}}$.
|
|
Refer to sampling function in <a href="#main">Main class</a> below
|
|
for the calculation of $\hat{A}_t$.</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">283</span> <span class="n">sampled_normalized_advantage</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_normalize</span><span class="p">(</span><span class="n">samples</span><span class="p">[</span><span class="s1">'advantages'</span><span class="p">])</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-64'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-64'>#</a>
|
|
</div>
|
|
<p>Sampled observations are fed into the model to get $\pi_\theta(a_t|s_t)$ and $V^{\pi_\theta}(s_t)$;
|
|
we are treating observations as state</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">287</span> <span class="n">pi</span><span class="p">,</span> <span class="n">value</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="p">(</span><span class="n">samples</span><span class="p">[</span><span class="s1">'obs'</span><span class="p">])</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-65'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-65'>#</a>
|
|
</div>
|
|
<p>$-\log \pi_\theta (a_t|s_t)$, $a_t$ are actions sampled from $\pi_{\theta_{OLD}}$</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">290</span> <span class="n">log_pi</span> <span class="o">=</span> <span class="n">pi</span><span class="o">.</span><span class="n">log_prob</span><span class="p">(</span><span class="n">samples</span><span class="p">[</span><span class="s1">'actions'</span><span class="p">])</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-66'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-66'>#</a>
|
|
</div>
|
|
<p>Calculate policy loss</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">293</span> <span class="n">policy_loss</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ppo_loss</span><span class="p">(</span><span class="n">log_pi</span><span class="p">,</span> <span class="n">samples</span><span class="p">[</span><span class="s1">'log_pis'</span><span class="p">],</span> <span class="n">sampled_normalized_advantage</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">clip_range</span><span class="p">())</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-67'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-67'>#</a>
|
|
</div>
|
|
<p>Calculate Entropy Bonus</p>
|
|
<p>$\mathcal{L}^{EB}(\theta) =
|
|
\mathbb{E}\Bigl[ S\bigl[\pi_\theta\bigr] (s_t) \Bigr]$</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">299</span> <span class="n">entropy_bonus</span> <span class="o">=</span> <span class="n">pi</span><span class="o">.</span><span class="n">entropy</span><span class="p">()</span>
|
|
<span class="lineno">300</span> <span class="n">entropy_bonus</span> <span class="o">=</span> <span class="n">entropy_bonus</span><span class="o">.</span><span class="n">mean</span><span class="p">()</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-68'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-68'>#</a>
|
|
</div>
|
|
<p>Calculate value function loss</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">303</span> <span class="n">value_loss</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">value_loss</span><span class="p">(</span><span class="n">value</span><span class="p">,</span> <span class="n">samples</span><span class="p">[</span><span class="s1">'values'</span><span class="p">],</span> <span class="n">sampled_return</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">clip_range</span><span class="p">())</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-69'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-69'>#</a>
|
|
</div>
|
|
<p>$\mathcal{L}^{CLIP+VF+EB} (\theta) =
|
|
\mathcal{L}^{CLIP} (\theta) +
|
|
c_1 \mathcal{L}^{VF} (\theta) - c_2 \mathcal{L}^{EB}(\theta)$</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">308</span> <span class="n">loss</span> <span class="o">=</span> <span class="p">(</span><span class="n">policy_loss</span>
|
|
<span class="lineno">309</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">value_loss_coef</span><span class="p">()</span> <span class="o">*</span> <span class="n">value_loss</span>
|
|
<span class="lineno">310</span> <span class="o">-</span> <span class="bp">self</span><span class="o">.</span><span class="n">entropy_bonus_coef</span><span class="p">()</span> <span class="o">*</span> <span class="n">entropy_bonus</span><span class="p">)</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-70'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-70'>#</a>
|
|
</div>
|
|
<p>for monitoring</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">313</span> <span class="n">approx_kl_divergence</span> <span class="o">=</span> <span class="o">.</span><span class="mi">5</span> <span class="o">*</span> <span class="p">((</span><span class="n">samples</span><span class="p">[</span><span class="s1">'log_pis'</span><span class="p">]</span> <span class="o">-</span> <span class="n">log_pi</span><span class="p">)</span> <span class="o">**</span> <span class="mi">2</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">()</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-71'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-71'>#</a>
|
|
</div>
|
|
<p>Add to tracker</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">316</span> <span class="n">tracker</span><span class="o">.</span><span class="n">add</span><span class="p">({</span><span class="s1">'policy_reward'</span><span class="p">:</span> <span class="o">-</span><span class="n">policy_loss</span><span class="p">,</span>
|
|
<span class="lineno">317</span> <span class="s1">'value_loss'</span><span class="p">:</span> <span class="n">value_loss</span><span class="p">,</span>
|
|
<span class="lineno">318</span> <span class="s1">'entropy_bonus'</span><span class="p">:</span> <span class="n">entropy_bonus</span><span class="p">,</span>
|
|
<span class="lineno">319</span> <span class="s1">'kl_div'</span><span class="p">:</span> <span class="n">approx_kl_divergence</span><span class="p">,</span>
|
|
<span class="lineno">320</span> <span class="s1">'clip_fraction'</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">ppo_loss</span><span class="o">.</span><span class="n">clip_fraction</span><span class="p">})</span>
|
|
<span class="lineno">321</span>
|
|
<span class="lineno">322</span> <span class="k">return</span> <span class="n">loss</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-72'>
|
|
<div class='docs doc-strings'>
|
|
<div class='section-link'>
|
|
<a href='#section-72'>#</a>
|
|
</div>
|
|
<h3>Run training loop</h3>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">324</span> <span class="k">def</span> <span class="nf">run_training_loop</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-73'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-73'>#</a>
|
|
</div>
|
|
<p>last 100 episode information</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">330</span> <span class="n">tracker</span><span class="o">.</span><span class="n">set_queue</span><span class="p">(</span><span class="s1">'reward'</span><span class="p">,</span> <span class="mi">100</span><span class="p">,</span> <span class="kc">True</span><span class="p">)</span>
|
|
<span class="lineno">331</span> <span class="n">tracker</span><span class="o">.</span><span class="n">set_queue</span><span class="p">(</span><span class="s1">'length'</span><span class="p">,</span> <span class="mi">100</span><span class="p">,</span> <span class="kc">True</span><span class="p">)</span>
|
|
<span class="lineno">332</span>
|
|
<span class="lineno">333</span> <span class="k">for</span> <span class="n">update</span> <span class="ow">in</span> <span class="n">monit</span><span class="o">.</span><span class="n">loop</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">updates</span><span class="p">):</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-74'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-74'>#</a>
|
|
</div>
|
|
<p>sample with current policy</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">335</span> <span class="n">samples</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">sample</span><span class="p">()</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-75'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-75'>#</a>
|
|
</div>
|
|
<p>train the model</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">338</span> <span class="bp">self</span><span class="o">.</span><span class="n">train</span><span class="p">(</span><span class="n">samples</span><span class="p">)</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-76'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-76'>#</a>
|
|
</div>
|
|
<p>Save tracked indicators.</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">341</span> <span class="n">tracker</span><span class="o">.</span><span class="n">save</span><span class="p">()</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-77'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-77'>#</a>
|
|
</div>
|
|
<p>Add a new line to the screen periodically</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">343</span> <span class="k">if</span> <span class="p">(</span><span class="n">update</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">%</span> <span class="mi">1_000</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
|
|
<span class="lineno">344</span> <span class="n">logger</span><span class="o">.</span><span class="n">log</span><span class="p">()</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-78'>
|
|
<div class='docs doc-strings'>
|
|
<div class='section-link'>
|
|
<a href='#section-78'>#</a>
|
|
</div>
|
|
<h3>Destroy</h3>
|
|
<p>Stop the workers</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">346</span> <span class="k">def</span> <span class="nf">destroy</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-79'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-79'>#</a>
|
|
</div>
|
|
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">351</span> <span class="k">for</span> <span class="n">worker</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">workers</span><span class="p">:</span>
|
|
<span class="lineno">352</span> <span class="n">worker</span><span class="o">.</span><span class="n">child</span><span class="o">.</span><span class="n">send</span><span class="p">((</span><span class="s2">"close"</span><span class="p">,</span> <span class="kc">None</span><span class="p">))</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-80'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-80'>#</a>
|
|
</div>
|
|
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">355</span><span class="k">def</span> <span class="nf">main</span><span class="p">():</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-81'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-81'>#</a>
|
|
</div>
|
|
<p>Create the experiment</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">357</span> <span class="n">experiment</span><span class="o">.</span><span class="n">create</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">'ppo'</span><span class="p">)</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-82'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-82'>#</a>
|
|
</div>
|
|
<p>Configurations</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">359</span> <span class="n">configs</span> <span class="o">=</span> <span class="p">{</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-83'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-83'>#</a>
|
|
</div>
|
|
<p>Number of updates</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">361</span> <span class="s1">'updates'</span><span class="p">:</span> <span class="mi">10000</span><span class="p">,</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-84'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-84'>#</a>
|
|
</div>
|
|
<p>⚙️ Number of epochs to train the model with sampled data.
|
|
You can change this while the experiment is running.
|
|
<a href="https://app.labml.ai/run/6eff28a0910e11eb9b008db315936e2f/hyper_params"><img alt="Example" src="https://img.shields.io/badge/example-hyperparams-brightgreen" /></a></p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">365</span> <span class="s1">'epochs'</span><span class="p">:</span> <span class="n">IntDynamicHyperParam</span><span class="p">(</span><span class="mi">8</span><span class="p">),</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-85'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-85'>#</a>
|
|
</div>
|
|
<p>Number of worker processes</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">367</span> <span class="s1">'n_workers'</span><span class="p">:</span> <span class="mi">8</span><span class="p">,</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-86'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-86'>#</a>
|
|
</div>
|
|
<p>Number of steps to run on each process for a single update</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">369</span> <span class="s1">'worker_steps'</span><span class="p">:</span> <span class="mi">128</span><span class="p">,</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-87'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-87'>#</a>
|
|
</div>
|
|
<p>Number of mini batches</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">371</span> <span class="s1">'batches'</span><span class="p">:</span> <span class="mi">4</span><span class="p">,</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-88'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-88'>#</a>
|
|
</div>
|
|
<p>⚙️ Value loss coefficient.
|
|
You can change this while the experiment is running.
|
|
<a href="https://app.labml.ai/run/6eff28a0910e11eb9b008db315936e2f/hyper_params"><img alt="Example" src="https://img.shields.io/badge/example-hyperparams-brightgreen" /></a></p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">375</span> <span class="s1">'value_loss_coef'</span><span class="p">:</span> <span class="n">FloatDynamicHyperParam</span><span class="p">(</span><span class="mf">0.5</span><span class="p">),</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-89'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-89'>#</a>
|
|
</div>
|
|
<p>⚙️ Entropy bonus coefficient.
|
|
You can change this while the experiment is running.
|
|
<a href="https://app.labml.ai/run/6eff28a0910e11eb9b008db315936e2f/hyper_params"><img alt="Example" src="https://img.shields.io/badge/example-hyperparams-brightgreen" /></a></p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">379</span> <span class="s1">'entropy_bonus_coef'</span><span class="p">:</span> <span class="n">FloatDynamicHyperParam</span><span class="p">(</span><span class="mf">0.01</span><span class="p">),</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-90'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-90'>#</a>
|
|
</div>
|
|
<p>⚙️ Clip range.</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">381</span> <span class="s1">'clip_range'</span><span class="p">:</span> <span class="n">FloatDynamicHyperParam</span><span class="p">(</span><span class="mf">0.1</span><span class="p">),</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-91'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-91'>#</a>
|
|
</div>
|
|
<p>You can change this while the experiment is running.
|
|
<a href="https://app.labml.ai/run/6eff28a0910e11eb9b008db315936e2f/hyper_params"><img alt="Example" src="https://img.shields.io/badge/example-hyperparams-brightgreen" /></a>
|
|
⚙️ Learning rate.</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">385</span> <span class="s1">'learning_rate'</span><span class="p">:</span> <span class="n">FloatDynamicHyperParam</span><span class="p">(</span><span class="mf">1e-3</span><span class="p">,</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mf">1e-3</span><span class="p">)),</span>
|
|
<span class="lineno">386</span> <span class="p">}</span>
|
|
<span class="lineno">387</span>
|
|
<span class="lineno">388</span> <span class="n">experiment</span><span class="o">.</span><span class="n">configs</span><span class="p">(</span><span class="n">configs</span><span class="p">)</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-92'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-92'>#</a>
|
|
</div>
|
|
<p>Initialize the trainer</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">391</span> <span class="n">m</span> <span class="o">=</span> <span class="n">Trainer</span><span class="p">(</span>
|
|
<span class="lineno">392</span> <span class="n">updates</span><span class="o">=</span><span class="n">configs</span><span class="p">[</span><span class="s1">'updates'</span><span class="p">],</span>
|
|
<span class="lineno">393</span> <span class="n">epochs</span><span class="o">=</span><span class="n">configs</span><span class="p">[</span><span class="s1">'epochs'</span><span class="p">],</span>
|
|
<span class="lineno">394</span> <span class="n">n_workers</span><span class="o">=</span><span class="n">configs</span><span class="p">[</span><span class="s1">'n_workers'</span><span class="p">],</span>
|
|
<span class="lineno">395</span> <span class="n">worker_steps</span><span class="o">=</span><span class="n">configs</span><span class="p">[</span><span class="s1">'worker_steps'</span><span class="p">],</span>
|
|
<span class="lineno">396</span> <span class="n">batches</span><span class="o">=</span><span class="n">configs</span><span class="p">[</span><span class="s1">'batches'</span><span class="p">],</span>
|
|
<span class="lineno">397</span> <span class="n">value_loss_coef</span><span class="o">=</span><span class="n">configs</span><span class="p">[</span><span class="s1">'value_loss_coef'</span><span class="p">],</span>
|
|
<span class="lineno">398</span> <span class="n">entropy_bonus_coef</span><span class="o">=</span><span class="n">configs</span><span class="p">[</span><span class="s1">'entropy_bonus_coef'</span><span class="p">],</span>
|
|
<span class="lineno">399</span> <span class="n">clip_range</span><span class="o">=</span><span class="n">configs</span><span class="p">[</span><span class="s1">'clip_range'</span><span class="p">],</span>
|
|
<span class="lineno">400</span> <span class="n">learning_rate</span><span class="o">=</span><span class="n">configs</span><span class="p">[</span><span class="s1">'learning_rate'</span><span class="p">],</span>
|
|
<span class="lineno">401</span> <span class="p">)</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-93'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-93'>#</a>
|
|
</div>
|
|
<p>Run and monitor the experiment</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">404</span> <span class="k">with</span> <span class="n">experiment</span><span class="o">.</span><span class="n">start</span><span class="p">():</span>
|
|
<span class="lineno">405</span> <span class="n">m</span><span class="o">.</span><span class="n">run_training_loop</span><span class="p">()</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-94'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-94'>#</a>
|
|
</div>
|
|
<p>Stop the workers</p>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">407</span> <span class="n">m</span><span class="o">.</span><span class="n">destroy</span><span class="p">()</span></pre></div>
|
|
</div>
|
|
</div>
|
|
<div class='section' id='section-95'>
|
|
<div class='docs'>
|
|
<div class='section-link'>
|
|
<a href='#section-95'>#</a>
|
|
</div>
|
|
<h2>Run it</h2>
|
|
</div>
|
|
<div class='code'>
|
|
<div class="highlight"><pre><span class="lineno">411</span><span class="k">if</span> <span class="vm">__name__</span> <span class="o">==</span> <span class="s2">"__main__"</span><span class="p">:</span>
|
|
<span class="lineno">412</span> <span class="n">main</span><span class="p">()</span></pre></div>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
<script src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.4/MathJax.js?config=TeX-AMS_HTML">
|
|
</script>
|
|
<!-- MathJax configuration -->
|
|
<script type="text/x-mathjax-config">
|
|
MathJax.Hub.Config({
|
|
tex2jax: {
|
|
inlineMath: [ ['$','$'] ],
|
|
displayMath: [ ['$$','$$'] ],
|
|
processEscapes: true,
|
|
processEnvironments: true
|
|
},
|
|
// Center justify equations in code and markdown cells. Elsewhere
|
|
// we use CSS to left justify single line equations in code cells.
|
|
displayAlign: 'center',
|
|
"HTML-CSS": { fonts: ["TeX"] }
|
|
});
|
|
</script>
|
|
<script>
|
|
function handleImages() {
|
|
var images = document.querySelectorAll('p>img')
|
|
|
|
console.log(images);
|
|
for (var i = 0; i < images.length; ++i) {
|
|
handleImage(images[i])
|
|
}
|
|
}
|
|
|
|
function handleImage(img) {
|
|
img.parentElement.style.textAlign = 'center'
|
|
|
|
var modal = document.createElement('div')
|
|
modal.id = 'modal'
|
|
|
|
var modalContent = document.createElement('div')
|
|
modal.appendChild(modalContent)
|
|
|
|
var modalImage = document.createElement('img')
|
|
modalContent.appendChild(modalImage)
|
|
|
|
var span = document.createElement('span')
|
|
span.classList.add('close')
|
|
span.textContent = 'x'
|
|
modal.appendChild(span)
|
|
|
|
img.onclick = function () {
|
|
console.log('clicked')
|
|
document.body.appendChild(modal)
|
|
modalImage.src = img.src
|
|
}
|
|
|
|
span.onclick = function () {
|
|
document.body.removeChild(modal)
|
|
}
|
|
}
|
|
|
|
handleImages()
|
|
</script>
|
|
</body>
|
|
</html> |