functions_operators_layers.html 21.9 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84


<!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>Design Doc: Functions, Operators, and Layers &mdash; PaddlePaddle  documentation</title>
  

  
  

  

  
  
    

  

  
  
    <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
  

  
  
        <link rel="index" title="Index"
              href="../genindex.html"/>
        <link rel="search" title="Search" href="../search.html"/>
    <link rel="top" title="PaddlePaddle  documentation" href="../index.html"/> 

  <link rel="stylesheet" href="https://cdn.jsdelivr.net/perfect-scrollbar/0.6.14/css/perfect-scrollbar.min.css" type="text/css" />
  <link rel="stylesheet" href="../_static/css/override.css" type="text/css" />
  <script>
  var _hmt = _hmt || [];
  (function() {
    var hm = document.createElement("script");
    hm.src = "//hm.baidu.com/hm.js?b9a314ab40d04d805655aab1deee08ba";
    var s = document.getElementsByTagName("script")[0]; 
    s.parentNode.insertBefore(hm, s);
  })();
  </script>

  

  
  <script src="../_static/js/modernizr.min.js"></script>

</head>

<body class="wy-body-for-nav" role="document">

  
  <header class="site-header">
    <div class="site-logo">
      <a href="/"><img src="../_static/images/PP_w.png"></a>
    </div>
    <div class="site-nav-links">
      <div class="site-menu">
        <a class="fork-on-github" href="https://github.com/PaddlePaddle/Paddle" target="_blank"><i class="fa fa-github"></i>Fork me on Github</a>
        <div class="language-switcher dropdown">
          <a type="button" data-toggle="dropdown">
            <span>English</span>
            <i class="fa fa-angle-up"></i>
            <i class="fa fa-angle-down"></i>
          </a>
          <ul class="dropdown-menu">
            <li><a href="/doc_cn">中文</a></li>
            <li><a href="/doc">English</a></li>
          </ul>
        </div>
        <ul class="site-page-links">
          <li><a href="/">Home</a></li>
        </ul>
      </div>
      <div class="doc-module">
        
        <ul>
<li class="toctree-l1"><a class="reference internal" href="../getstarted/index_en.html">GET STARTED</a></li>
85
<li class="toctree-l1"><a class="reference internal" href="../build_and_install/index_en.html">Install and Build</a></li>
86
<li class="toctree-l1"><a class="reference internal" href="../howto/index_en.html">HOW TO</a></li>
87
<li class="toctree-l1"><a class="reference internal" href="../dev/index_en.html">Development</a></li>
88
<li class="toctree-l1"><a class="reference internal" href="../faq/index_en.html">FAQ</a></li>
89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110
</ul>

        
<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>
  </header>
  
  <div class="main-content-wrap">

    
    <nav class="doc-menu-vertical" role="navigation">
        
          
          <ul>
<li class="toctree-l1"><a class="reference internal" href="../getstarted/index_en.html">GET STARTED</a><ul>
111
<li class="toctree-l2"><a class="reference internal" href="../getstarted/quickstart_en.html">Quick Start</a></li>
112
<li class="toctree-l2"><a class="reference internal" href="../getstarted/concepts/use_concepts_en.html">Basic Concept</a></li>
113 114
</ul>
</li>
115
<li class="toctree-l1"><a class="reference internal" href="../build_and_install/index_en.html">Install and Build</a><ul>
116
<li class="toctree-l2"><a class="reference internal" href="../build_and_install/pip_install_en.html">Install using pip</a></li>
117 118
<li class="toctree-l2"><a class="reference internal" href="../build_and_install/docker_install_en.html">Run in Docker Containers</a></li>
<li class="toctree-l2"><a class="reference internal" href="../build_and_install/build_from_source_en.html">Build from Sources</a></li>
119 120 121
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../howto/index_en.html">HOW TO</a><ul>
122 123 124 125
<li class="toctree-l2"><a class="reference internal" href="../howto/cmd_parameter/index_en.html">Set Command-line Parameters</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../howto/cmd_parameter/use_case_en.html">Use Case</a></li>
<li class="toctree-l3"><a class="reference internal" href="../howto/cmd_parameter/arguments_en.html">Argument Outline</a></li>
<li class="toctree-l3"><a class="reference internal" href="../howto/cmd_parameter/detail_introduction_en.html">Detail Description</a></li>
126 127
</ul>
</li>
128 129 130 131
<li class="toctree-l2"><a class="reference internal" href="../howto/cluster/index_en.html">Distributed Training</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../howto/cluster/preparations_en.html">Preparations</a></li>
<li class="toctree-l3"><a class="reference internal" href="../howto/cluster/cmd_argument_en.html">Command-line arguments</a></li>
<li class="toctree-l3"><a class="reference internal" href="../howto/cluster/multi_cluster/index_en.html">Use different clusters</a><ul>
132 133 134 135
<li class="toctree-l4"><a class="reference internal" href="../howto/cluster/multi_cluster/fabric_en.html">Fabric</a></li>
<li class="toctree-l4"><a class="reference internal" href="../howto/cluster/multi_cluster/openmpi_en.html">OpenMPI</a></li>
<li class="toctree-l4"><a class="reference internal" href="../howto/cluster/multi_cluster/k8s_en.html">Kubernetes</a></li>
<li class="toctree-l4"><a class="reference internal" href="../howto/cluster/multi_cluster/k8s_aws_en.html">Kubernetes on AWS</a></li>
136 137
</ul>
</li>
138 139 140 141
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../howto/rnn/index_en.html">RNN Models</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../howto/rnn/rnn_config_en.html">RNN Configuration</a></li>
142 143 144
<li class="toctree-l3"><a class="reference internal" href="../howto/rnn/recurrent_group_en.html">Recurrent Group Tutorial</a></li>
<li class="toctree-l3"><a class="reference internal" href="../howto/rnn/hierarchical_layer_en.html">Layers supporting hierarchical sequence as input</a></li>
<li class="toctree-l3"><a class="reference internal" href="../howto/rnn/hrnn_rnn_api_compare_en.html">API comparision between RNN and hierarchical RNN</a></li>
145 146 147 148 149
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../howto/optimization/gpu_profiling_en.html">Tune GPU Performance</a></li>
</ul>
</li>
150 151 152 153 154
<li class="toctree-l1"><a class="reference internal" href="../dev/index_en.html">Development</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../dev/contribute_to_paddle_en.html">Contribute Code</a></li>
<li class="toctree-l2"><a class="reference internal" href="../dev/write_docs_en.html">Contribute Documentation</a></li>
</ul>
</li>
155 156 157 158 159 160 161 162
<li class="toctree-l1"><a class="reference internal" href="../faq/index_en.html">FAQ</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../faq/build_and_install/index_en.html">Install, Build and Unit test</a></li>
<li class="toctree-l2"><a class="reference internal" href="../faq/model/index_en.html">Model Configuration</a></li>
<li class="toctree-l2"><a class="reference internal" href="../faq/parameter/index_en.html">Parameter Setting</a></li>
<li class="toctree-l2"><a class="reference internal" href="../faq/local/index_en.html">Local Training and Prediction</a></li>
<li class="toctree-l2"><a class="reference internal" href="../faq/cluster/index_en.html">Cluster Training and Prediction</a></li>
</ul>
</li>
163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231
</ul>

        
    </nav>
    
    <section class="doc-content-wrap">

      

 







<div role="navigation" aria-label="breadcrumbs navigation">
  <ul class="wy-breadcrumbs">
      
    <li>Design Doc: Functions, Operators, and Layers</li>
  </ul>
</div>
      
      <div class="wy-nav-content" id="doc-content">
        <div class="rst-content">
          <div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
           <div itemprop="articleBody">
            
  <div class="section" id="design-doc-functions-operators-and-layers">
<span id="design-doc-functions-operators-and-layers"></span><h1>Design Doc: Functions, Operators, and Layers<a class="headerlink" href="#design-doc-functions-operators-and-layers" title="Permalink to this headline"></a></h1>
<p>In a DL system, we can compose one or more fine grained operators into a coarse grained one.  For example, the FC layer can be composed of a multiplication operator and an add operator.</p>
<p>Historically, some fine grained operations are known as operators, and some coarse level ones are known as layers.  But we need a well-defined separation.</p>
<p>In general, operators are those very fine grained operations, e.g., mul and add. In the implementation, we can write them as C++ functions:</p>
<div class="highlight-c++"><div class="highlight"><pre><span></span><span class="k">template</span> <span class="o">&lt;</span><span class="k">typename</span> <span class="n">T</span><span class="o">&gt;</span> <span class="n">T</span> <span class="n">add</span><span class="p">(</span><span class="n">T</span> <span class="n">x</span><span class="p">,</span> <span class="n">T</span> <span class="n">y</span><span class="p">)</span> <span class="p">{</span> <span class="k">return</span> <span class="n">x</span> <span class="o">+</span> <span class="n">y</span><span class="p">;</span> <span class="p">}</span>
<span class="k">template</span> <span class="o">&lt;</span><span class="k">typename</span> <span class="n">T</span><span class="o">&gt;</span> <span class="n">T</span> <span class="n">mul</span><span class="p">(</span><span class="n">T</span> <span class="n">x</span><span class="p">,</span> <span class="n">T</span> <span class="n">y</span><span class="p">)</span> <span class="p">{</span> <span class="k">return</span> <span class="n">x</span> <span class="o">*</span> <span class="n">y</span><span class="p">;</span> <span class="p">}</span>
</pre></div>
</div>
<p>Then we can wrap them into operators which are C++ classes and can be created from Python bindings by name.  A C macro can do this. For example, the following macro invocation</p>
<div class="highlight-c++"><div class="highlight"><pre><span></span><span class="cp">#define MAKE_FUNCTION_OPERATOR(mul);</span>
</pre></div>
</div>
<p>generates</p>
<div class="highlight-c++"><div class="highlight"><pre><span></span><span class="k">template</span> <span class="o">&lt;</span><span class="k">typename</span> <span class="n">T</span><span class="o">&gt;</span> <span class="k">class</span> <span class="nc">mulOp</span> <span class="o">:</span> <span class="k">public</span> <span class="n">OperatorBase</span> <span class="p">{...};</span>
<span class="n">REGISTER_OP</span><span class="p">(</span><span class="n">mulOp</span><span class="o">&lt;</span><span class="n">float32</span><span class="o">&gt;</span><span class="p">,</span> <span class="s">&quot;mul&quot;</span><span class="p">);</span>
</pre></div>
</div>
<p>so that in Python we can create operator mul by:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">X1</span> <span class="o">=</span> <span class="n">Var</span><span class="p">()</span>
<span class="n">X2</span> <span class="o">=</span> <span class="n">Var</span><span class="p">()</span>
<span class="n">Y</span> <span class="o">=</span> <span class="n">Var</span><span class="p">()</span>
<span class="n">paddle</span><span class="o">.</span><span class="n">cpp</span><span class="o">.</span><span class="n">create_operator</span><span class="p">(</span><span class="s2">&quot;mul&quot;</span><span class="p">,</span> <span class="nb">input</span><span class="o">=</span><span class="p">[</span><span class="n">X1</span><span class="p">,</span> <span class="n">X2</span><span class="p">],</span> <span class="n">output</span><span class="o">=</span><span class="n">Y</span><span class="p">)</span>
</pre></div>
</div>
<p>Also, at the same time, we can compose a coarse level C++ operator class by composing functions <code class="docutils literal"><span class="pre">mul</span></code> and <code class="docutils literal"><span class="pre">add</span></code>:</p>
<div class="highlight-c++"><div class="highlight"><pre><span></span><span class="k">template</span> <span class="o">&lt;</span><span class="k">typename</span> <span class="n">T</span><span class="o">&gt;</span>
<span class="k">class</span> <span class="nc">FCOp</span> <span class="o">:</span> <span class="k">public</span> <span class="n">OperatorBase</span> <span class="p">{</span>
 <span class="k">public</span><span class="o">:</span>
  <span class="kt">void</span> <span class="n">Run</span><span class="p">(...)</span> <span class="p">{</span>
    <span class="n">add</span><span class="p">(</span><span class="n">mul</span><span class="p">(</span><span class="n">Input</span><span class="o">&lt;</span><span class="n">T</span><span class="o">&gt;</span><span class="p">(</span><span class="s">&quot;X&quot;</span><span class="p">),</span> <span class="n">Input</span><span class="o">&lt;</span><span class="n">T</span><span class="o">&gt;</span><span class="p">(</span><span class="s">&quot;W&quot;</span><span class="p">)),</span> <span class="n">Input</span><span class="o">&lt;</span><span class="n">T</span><span class="o">&gt;</span><span class="p">(</span><span class="s">&quot;b&quot;</span><span class="p">);</span>
  <span class="p">}</span>
<span class="p">};</span>
<span class="n">REGISTER_OP</span><span class="p">(</span><span class="n">FCOp</span><span class="p">,</span> <span class="s">&quot;fc&quot;</span><span class="p">);</span>
</pre></div>
</div>
<p>We need to support such composition in Python as well.  To do so, we need a higher level Python wrapping of operator creation than <code class="docutils literal"><span class="pre">paddle.cpp.create_operator</span></code>.  This higher level operator API should be compatible with the layer API.</p>
<p>Let&#8217;s explain using an example.  Suppose that we are going to compose the FC using mul and add in Python, we&#8217;d like to have Python functions <code class="docutils literal"><span class="pre">mul</span></code> and <code class="docutils literal"><span class="pre">add</span></code> defined in module <code class="docutils literal"><span class="pre">operator</span></code>:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="k">def</span> <span class="nf">operator</span><span class="o">.</span><span class="n">mul</span><span class="p">(</span><span class="n">X1</span><span class="p">,</span> <span class="n">X2</span><span class="p">):</span>
    <span class="n">O</span> <span class="o">=</span> <span class="n">Var</span><span class="p">()</span>
232
    <span class="n">paddle</span><span class="o">.</span><span class="n">cpp</span><span class="o">.</span><span class="n">create_operator</span><span class="p">(</span><span class="s2">&quot;mul&quot;</span><span class="p">,</span> <span class="nb">input</span><span class="o">=</span><span class="p">{</span><span class="n">X1</span><span class="p">,</span> <span class="n">Y1</span><span class="p">},</span> <span class="n">output</span><span class="o">=</span><span class="n">O</span><span class="p">)</span>
233 234 235 236
    <span class="k">return</span> <span class="n">O</span>

<span class="k">def</span> <span class="nf">operator</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">X1</span><span class="p">,</span> <span class="n">X2</span><span class="p">):</span>
    <span class="n">O</span> <span class="o">=</span> <span class="n">Var</span><span class="p">()</span>
237
    <span class="n">paddle</span><span class="o">.</span><span class="n">cpp</span><span class="o">.</span><span class="n">create_operator</span><span class="p">(</span><span class="s2">&quot;add&quot;</span><span class="p">,</span> <span class="nb">input</span><span class="o">=</span><span class="p">{</span><span class="n">X1</span><span class="p">,</span> <span class="n">X2</span><span class="p">},</span> <span class="n">output</span><span class="o">=</span><span class="n">O</span><span class="p">)</span>
238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259
    <span class="k">return</span> <span class="n">O</span>
</pre></div>
</div>
<p>Above code snippets are automatically generated.  Given them, users can define</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="k">def</span> <span class="nf">layer</span><span class="o">.</span><span class="n">fc</span><span class="p">(</span><span class="n">X</span><span class="p">):</span>
    <span class="n">W</span> <span class="o">=</span> <span class="n">Var</span><span class="p">()</span>
    <span class="n">b</span> <span class="o">=</span> <span class="n">Var</span><span class="p">()</span>
    <span class="k">return</span> <span class="n">operator</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">operator</span><span class="o">.</span><span class="n">mul</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">W</span><span class="p">),</span> <span class="n">b</span><span class="p">)</span>
</pre></div>
</div>
<p>If we don&#8217;t have <code class="docutils literal"><span class="pre">operator.mul</span></code> and <code class="docutils literal"><span class="pre">operator.add</span></code>, the definiton of <code class="docutils literal"><span class="pre">layer.fc</span></code> would be complicated:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="k">def</span> <span class="nf">layer</span><span class="o">.</span><span class="n">fc</span><span class="p">(</span><span class="n">X</span><span class="p">):</span>
    <span class="n">W</span> <span class="o">=</span> <span class="n">Var</span><span class="p">()</span>
    <span class="n">b</span> <span class="o">=</span> <span class="n">Var</span><span class="p">()</span>
    <span class="n">O1</span> <span class="o">=</span> <span class="n">Var</span><span class="p">()</span>
    <span class="n">paddle</span><span class="o">.</span><span class="n">cpp</span><span class="o">.</span><span class="n">create_operator</span><span class="p">(</span><span class="s2">&quot;mul&quot;</span><span class="p">,</span> <span class="nb">input</span><span class="o">=</span><span class="p">[</span><span class="n">X</span><span class="p">,</span> <span class="n">W</span><span class="p">],</span> <span class="n">output</span><span class="o">=</span><span class="n">O1</span><span class="p">)</span>
    <span class="n">O2</span> <span class="o">=</span> <span class="n">Var</span><span class="p">()</span>
    <span class="n">paddle</span><span class="o">.</span><span class="n">cpp</span><span class="o">.</span><span class="n">create_operator</span><span class="p">(</span><span class="s2">&quot;add&quot;</span><span class="p">,</span> <span class="nb">input</span><span class="o">=</span><span class="p">[</span><span class="n">O1</span><span class="p">,</span> <span class="n">b</span><span class="p">],</span> <span class="n">output</span><span class="o">=</span><span class="n">O2</span><span class="p">)</span>
    <span class="k">return</span> <span class="n">O2</span>
</pre></div>
</div>
<p>We&#8217;d like to have Python bindings to operators in package <code class="docutils literal"><span class="pre">paddle.operator</span></code>, and Python compositions of operators in package <code class="docutils literal"><span class="pre">paddle.layer</span></code>.  So we have the following concepts in above illustrative example:</p>
260 261
<p>| C++ functions/functors | mul          | add          |             |          |
|&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;|&#8212;&#8212;&#8212;&#8212;&#8211;|&#8212;&#8212;&#8212;&#8212;&#8211;|&#8212;&#8212;&#8212;&#8212;-|&#8212;&#8212;&#8212;-|
262 263
| C++ operator class     | mulOp        | addOp        | FCOp        |          |
| Python binding         | operator.mul | operator.add | operator.fc |          |
264
| Python function        |              |              |             | layer.fc |</p>
265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328
<p>This is how we differentiate layer and operators in PaddlePaddle:</p>
<ul class="simple">
<li>those defined in C++ and have a lightweighted Python wrapper in module <code class="docutils literal"><span class="pre">operators</span></code> are operators; whereas</li>
<li>those who don&#8217;t have C++ implementations but a Python implementation that compose C++ operators are known as layers.</li>
</ul>
</div>


           </div>
          </div>
          <footer>
  

  <hr/>

  <div role="contentinfo">
    <p>
        &copy; Copyright 2016, PaddlePaddle developers.

    </p>
  </div>
  Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/snide/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">
        var DOCUMENTATION_OPTIONS = {
            URL_ROOT:'../',
            VERSION:'',
            COLLAPSE_INDEX:false,
            FILE_SUFFIX:'.html',
            HAS_SOURCE:  true,
            SOURCELINK_SUFFIX: ".txt",
        };
    </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 type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.0/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script>
       
  

  
  
    <script type="text/javascript" src="../_static/js/theme.js"></script>
  
  
  <script src="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.7/js/bootstrap.min.js" integrity="sha384-Tc5IQib027qvyjSMfHjOMaLkfuWVxZxUPnCJA7l2mCWNIpG9mGCD8wGNIcPD7Txa" crossorigin="anonymous"></script>
  <script src="https://cdn.jsdelivr.net/perfect-scrollbar/0.6.14/js/perfect-scrollbar.jquery.min.js"></script>
  <script src="../_static/js/paddle_doc_init.js"></script> 

</body>
</html>