var_desc.html 18.4 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>Background &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
</ul>

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

      

 







<div role="navigation" aria-label="breadcrumbs navigation">
  <ul class="wy-breadcrumbs">
      
    <li>Background</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="background">
<span id="background"></span><h1>Background<a class="headerlink" href="#background" title="Permalink to this headline"></a></h1>
194
<p>PaddlePaddle divides the description of neural network computation into two stages: compile time and runtime. At compile time, the neural network computation is described as a <code class="docutils literal"><span class="pre">ProgramDesc</span></code> whereas at runtime an <code class="docutils literal"><span class="pre">Executor</span></code> interprets the <code class="docutils literal"><span class="pre">ProgramDesc</span></code> to compute the operations.</p>
195
<p>PaddlePaddle uses proto message to describe compile time program because :</p>
196
<ol class="simple">
197
<li>The computation program description must be serializable and saved in a file.</li>
198
<li>During distributed training, the serialized program will be sent to multiple workers. It should also be possible to break the program into different components, each of which can be executed on a different worker.</li>
199
</ol>
200
<p>The computation <code class="docutils literal"><span class="pre">Program</span></code> consists of nested <code class="docutils literal"><span class="pre">Blocks</span></code>. Each <code class="docutils literal"><span class="pre">Block</span></code> will consist of data(i.e. <code class="docutils literal"><span class="pre">Variable</span></code>)  and  <code class="docutils literal"><span class="pre">Operations</span></code>. The concept to represent them is in the table below.</p>
201 202 203 204 205
<p>| |compile time|runtime|
|&#8212;|&#8212;|&#8212;|
|Data|VarDesc(proto)|Variable(cpp)|
|Operation|OpDesc(proto)|Operator(cpp)|</p>
</div>
206 207 208
<div class="section" id="definition-of-vartype">
<span id="definition-of-vartype"></span><h1>Definition of VarType<a class="headerlink" href="#definition-of-vartype" title="Permalink to this headline"></a></h1>
<p>A VarDesc should have a name, type and whether or not it is persistable. The are different kinds of variable types supported in PaddlePaddle, apart from the POD_Types like: <code class="docutils literal"><span class="pre">LOD_TENSOR</span></code>, <code class="docutils literal"><span class="pre">SELECTED_ROWS</span></code>, <code class="docutils literal"><span class="pre">FEED_MINIBATCH</span></code>, <code class="docutils literal"><span class="pre">FETCH_LIST</span></code>, <code class="docutils literal"><span class="pre">STEP_SCOPES</span></code>, <code class="docutils literal"><span class="pre">LOD_RANK_TABLE</span></code>, <code class="docutils literal"><span class="pre">LOD_TENSOR_ARRAY</span></code>, <code class="docutils literal"><span class="pre">PLACE_LIST</span></code>, <code class="docutils literal"><span class="pre">READER</span></code> and <code class="docutils literal"><span class="pre">CHANNEL</span></code>. These are declared inside <code class="docutils literal"><span class="pre">VarType</span></code>. A <code class="docutils literal"><span class="pre">VarDesc</span></code> then looks as the following:</p>
209 210
<div class="highlight-proto"><div class="highlight"><pre><span></span><span class="kd">message</span> <span class="nc">VarDesc</span> <span class="p">{</span>
  <span class="k">required</span> <span class="kt">string</span> <span class="na">name</span> <span class="o">=</span> <span class="mi">1</span><span class="p">;</span>
211
  <span class="k">required</span> <span class="n">VarType</span> <span class="na">type</span> <span class="o">=</span> <span class="mi">2</span><span class="p">;</span>
212
  <span class="k">optional</span> <span class="kt">bool</span> <span class="na">persistable</span> <span class="o">=</span> <span class="mi">3</span> <span class="p">[</span> <span class="k">default</span> <span class="o">=</span> <span class="kc">false</span> <span class="p">];</span>
213 214 215 216
<span class="p">}</span>
</pre></div>
</div>
</div>
217 218
<div class="section" id="definition-of-tensordesc">
<span id="definition-of-tensordesc"></span><h1>Definition of TensorDesc<a class="headerlink" href="#definition-of-tensordesc" title="Permalink to this headline"></a></h1>
219 220 221 222 223 224 225 226 227 228
<div class="highlight-proto"><div class="highlight"><pre><span></span><span class="kd">message</span> <span class="nc">TensorDesc</span> <span class="p">{</span>
  <span class="c1">// Should only be PODType. Is enforced in C++</span>
  <span class="k">required</span> <span class="n">Type</span> <span class="na">data_type</span> <span class="o">=</span> <span class="mi">1</span><span class="p">;</span>
  <span class="k">repeated</span> <span class="kt">int64</span> <span class="na">dims</span> <span class="o">=</span> <span class="mi">2</span><span class="p">;</span> <span class="c1">// [UNK, 640, 480] is saved as [-1, 640, 480]</span>
<span class="p">}</span>
</pre></div>
</div>
<p>The <code class="docutils literal"><span class="pre">Type</span></code> here comes from the enum defined inside of <code class="docutils literal"><span class="pre">VarType</span></code> :</p>
<div class="highlight-proto"><div class="highlight"><pre><span></span><span class="kd">enum</span> <span class="n">Type</span> <span class="p">{</span>
  <span class="c1">// Pod Types</span>
229 230 231 232 233 234 235 236
  <span class="na">BOOL</span> <span class="o">=</span> <span class="mi">0</span><span class="p">;</span>
  <span class="na">INT16</span> <span class="o">=</span> <span class="mi">1</span><span class="p">;</span>
  <span class="na">INT32</span> <span class="o">=</span> <span class="mi">2</span><span class="p">;</span>
  <span class="na">INT64</span> <span class="o">=</span> <span class="mi">3</span><span class="p">;</span>
  <span class="na">FP16</span> <span class="o">=</span> <span class="mi">4</span><span class="p">;</span>
  <span class="na">FP32</span> <span class="o">=</span> <span class="mi">5</span><span class="p">;</span>
  <span class="na">FP64</span> <span class="o">=</span> <span class="mi">6</span><span class="p">;</span>

237 238 239 240 241 242 243 244 245 246 247
  <span class="c1">// Other types that may need additional descriptions</span>
  <span class="na">LOD_TENSOR</span> <span class="o">=</span> <span class="mi">7</span><span class="p">;</span>
  <span class="na">SELECTED_ROWS</span> <span class="o">=</span> <span class="mi">8</span><span class="p">;</span>
  <span class="na">FEED_MINIBATCH</span> <span class="o">=</span> <span class="mi">9</span><span class="p">;</span>
  <span class="na">FETCH_LIST</span> <span class="o">=</span> <span class="mi">10</span><span class="p">;</span>
  <span class="na">STEP_SCOPES</span> <span class="o">=</span> <span class="mi">11</span><span class="p">;</span>
  <span class="na">LOD_RANK_TABLE</span> <span class="o">=</span> <span class="mi">12</span><span class="p">;</span>
  <span class="na">LOD_TENSOR_ARRAY</span> <span class="o">=</span> <span class="mi">13</span><span class="p">;</span>
  <span class="na">PLACE_LIST</span> <span class="o">=</span> <span class="mi">14</span><span class="p">;</span>
  <span class="na">READER</span> <span class="o">=</span> <span class="mi">15</span><span class="p">;</span>
  <span class="na">CHANNEL</span> <span class="o">=</span> <span class="mi">16</span><span class="p">;</span>
248 249 250
<span class="p">}</span>
</pre></div>
</div>
251
<p>A TensorDesc describes <code class="docutils literal"><span class="pre">SelectedRows</span></code> and <code class="docutils literal"><span class="pre">LoDTensor</span></code>. For details of <code class="docutils literal"><span class="pre">SelectedRows</span></code>, please reference <a class="reference external" href="./selected_rows.md"><code class="docutils literal"><span class="pre">SelectedRows</span></code></a>.</p>
252
</div>
253 254 255 256
<div class="section" id="definition-of-lodtensordesc">
<span id="definition-of-lodtensordesc"></span><h1>Definition of LodTensorDesc<a class="headerlink" href="#definition-of-lodtensordesc" title="Permalink to this headline"></a></h1>
<div class="highlight-proto"><div class="highlight"><pre><span></span><span class="kd">message</span> <span class="nc">LoDTensorDesc</span> <span class="p">{</span>
  <span class="k">required</span> <span class="n">TensorDesc</span> <span class="na">tensor</span> <span class="o">=</span> <span class="mi">1</span><span class="p">;</span>
257
  <span class="k">optional</span> <span class="kt">int32</span> <span class="na">lod_level</span> <span class="o">=</span> <span class="mi">2</span> <span class="p">[</span> <span class="k">default</span> <span class="o">=</span> <span class="mi">0</span> <span class="p">];</span>
258
<span class="p">}</span>
259 260
</pre></div>
</div>
261
<p>A LoDTensorDesc contains a tensor and a lod_level.</p>
262
</div>
263 264 265
<div class="section" id="definition-of-variable-in-python">
<span id="definition-of-variable-in-python"></span><h1>Definition of Variable in Python<a class="headerlink" href="#definition-of-variable-in-python" title="Permalink to this headline"></a></h1>
<p>For Variable in Python, please reference <a class="reference external" href="./python_api.md"><code class="docutils literal"><span class="pre">Python</span> <span class="pre">API</span></code></a>.</p>
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
</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>