layers.html



<!DOCTYPE html>
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
<head>
  <meta charset="utf-8">
  
  <meta name="viewport" content="width=device-width, initial-scale=1.0">
  
  <title>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="up" title="Fluid" href="../fluid.html"/>
        <link rel="next" title="data_feeder" href="data_feeder.html"/>
        <link rel="prev" title="Fluid" href="../fluid.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 class="current">
<li class="toctree-l1"><a class="reference internal" href="../model_configs.html">Model Configuration</a></li>
<li class="toctree-l1"><a class="reference internal" href="../data.html">Data Reader Interface and DataSets</a></li>
<li class="toctree-l1"><a class="reference internal" href="../run_logic.html">Training and Inference</a></li>
<li class="toctree-l1 current"><a class="reference internal" href="../fluid.html">Fluid</a></li>
</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 class="current">
<li class="toctree-l1"><a class="reference internal" href="../model_configs.html">Model Configuration</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../config/activation.html">Activation</a></li>
<li class="toctree-l2"><a class="reference internal" href="../config/layer.html">Layers</a></li>
<li class="toctree-l2"><a class="reference internal" href="../config/evaluators.html">Evaluators</a></li>
<li class="toctree-l2"><a class="reference internal" href="../config/optimizer.html">Optimizer</a></li>
<li class="toctree-l2"><a class="reference internal" href="../config/pooling.html">Pooling</a></li>
<li class="toctree-l2"><a class="reference internal" href="../config/networks.html">Networks</a></li>
<li class="toctree-l2"><a class="reference internal" href="../config/attr.html">Parameter Attribute</a></li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../data.html">Data Reader Interface and DataSets</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../data/data_reader.html">Data Reader Interface</a></li>
<li class="toctree-l2"><a class="reference internal" href="../data/image.html">Image Interface</a></li>
<li class="toctree-l2"><a class="reference internal" href="../data/dataset.html">Dataset</a></li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../run_logic.html">Training and Inference</a></li>
<li class="toctree-l1 current"><a class="reference internal" href="../fluid.html">Fluid</a><ul class="current">
<li class="toctree-l2 current"><a class="current reference internal" href="#">layers</a></li>
<li class="toctree-l2"><a class="reference internal" href="data_feeder.html">data_feeder</a></li>
<li class="toctree-l2"><a class="reference internal" href="executor.html">executor</a></li>
<li class="toctree-l2"><a class="reference internal" href="initializer.html">initializer</a></li>
<li class="toctree-l2"><a class="reference internal" href="evaluator.html">evaluator</a></li>
<li class="toctree-l2"><a class="reference internal" href="nets.html">nets</a></li>
<li class="toctree-l2"><a class="reference internal" href="optimizer.html">optimizer</a></li>
<li class="toctree-l2"><a class="reference internal" href="param_attr.html">param_attr</a></li>
<li class="toctree-l2"><a class="reference internal" href="profiler.html">profiler</a></li>
<li class="toctree-l2"><a class="reference internal" href="regularizer.html">regularizer</a></li>
<li class="toctree-l2"><a class="reference internal" href="io.html">io</a></li>
</ul>
</li>
</ul>

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

      
<div role="navigation" aria-label="breadcrumbs navigation">
  <ul class="wy-breadcrumbs">
      
        <li><a href="../fluid.html">Fluid</a> > </li>
      
    <li>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="layers">
<h1>layers<a class="headerlink" href="#layers" title="Permalink to this headline">¶</a></h1>
<div class="section" id="control-flow">
<h2>control_flow<a class="headerlink" href="#control-flow" title="Permalink to this headline">¶</a></h2>
<div class="section" id="split-lod-tensor">
<h3>split_lod_tensor<a class="headerlink" href="#split-lod-tensor" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">split_lod_tensor</code><span class="sig-paren">(</span><em>input</em>, <em>mask</em>, <em>level=0</em><span class="sig-paren">)</span></dt>
<dd><p><strong>split_lod_tensor</strong></p>
<p>This function takes in an input that contains the complete lod information,
and takes in a mask which is used to mask certain parts of the input.
The output is the true branch and the false branch with the mask applied to
the input at a certain level in the tensor.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>tuple|list|None</em>) &#8211; The input tensor that contains complete
lod information needed to construct the output.</li>
<li><strong>mask</strong> (<em>list</em>) &#8211; A bool column vector which masks the input.</li>
<li><strong>level</strong> (<em>int</em>) &#8211; The specific lod level to rank.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The true branch of tensor as per the mask applied to input.
Variable: The false branch of tensor as per the mask applied to input.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">x</span> <span class="o">=</span> <span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;x&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
<span class="n">x</span><span class="o">.</span><span class="n">persistable</span> <span class="o">=</span> <span class="bp">True</span>

<span class="n">y</span> <span class="o">=</span> <span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;y&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
<span class="n">y</span><span class="o">.</span><span class="n">persistable</span> <span class="o">=</span> <span class="bp">True</span>

<span class="n">out_true</span><span class="p">,</span> <span class="n">out_false</span> <span class="o">=</span> <span class="n">layers</span><span class="o">.</span><span class="n">split_lod_tensor</span><span class="p">(</span>
      <span class="nb">input</span><span class="o">=</span><span class="n">x</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">y</span><span class="p">,</span> <span class="n">level</span><span class="o">=</span><span class="n">level</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="merge-lod-tensor">
<h3>merge_lod_tensor<a class="headerlink" href="#merge-lod-tensor" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">merge_lod_tensor</code><span class="sig-paren">(</span><em>in_true</em>, <em>in_false</em>, <em>x</em>, <em>mask</em>, <em>level=0</em><span class="sig-paren">)</span></dt>
<dd><p><strong>merge_lod_tensor</strong></p>
<p>This function takes in an input <span class="math">\(x\)</span>, the True branch, the False
branch and a binary <span class="math">\(mask\)</span>. Using this information, this function
merges the True and False branches of the tensor into a single Output
at a certain lod level indiacted by <span class="math">\(level\)</span>.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>in_true</strong> (<em>tuple|list|None</em>) &#8211; The True branch to be merged.</li>
<li><strong>in_false</strong> (<em>tuple|list|None</em>) &#8211; The False branch to be merged.</li>
<li><strong>x</strong> (<em>tuple|list|None</em>) &#8211; The input tensor that contains complete
lod information needed to construct the output.</li>
<li><strong>mask</strong> (<em>list</em>) &#8211; A bool column vector which masks the input.</li>
<li><strong>level</strong> (<em>int</em>) &#8211; The specific lod level to rank.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The merged output tensor.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">x</span> <span class="o">=</span> <span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span>
            <span class="n">name</span><span class="o">=</span><span class="s1">&#39;x&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">,</span> <span class="n">stop_gradient</span><span class="o">=</span><span class="bp">False</span><span class="p">)</span>
<span class="n">y</span> <span class="o">=</span> <span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span>
      <span class="n">name</span><span class="o">=</span><span class="s1">&#39;y&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;bool&#39;</span><span class="p">,</span> <span class="n">stop_gradient</span><span class="o">=</span><span class="bp">False</span><span class="p">)</span>

<span class="n">level</span> <span class="o">=</span> <span class="mi">0</span>

<span class="n">out_true</span><span class="p">,</span> <span class="n">out_false</span> <span class="o">=</span> <span class="n">layers</span><span class="o">.</span><span class="n">split_lod_tensor</span><span class="p">(</span>
      <span class="nb">input</span><span class="o">=</span><span class="n">x</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">y</span><span class="p">,</span> <span class="n">level</span><span class="o">=</span><span class="n">level</span><span class="p">)</span>
<span class="n">out</span> <span class="o">=</span> <span class="n">layers</span><span class="o">.</span><span class="n">merge_lod_tensor</span><span class="p">(</span>
      <span class="n">in_true</span><span class="o">=</span><span class="n">out_true</span><span class="p">,</span> <span class="n">in_false</span><span class="o">=</span><span class="n">out_false</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">y</span><span class="p">,</span> <span class="n">x</span><span class="o">=</span><span class="n">x</span><span class="p">,</span> <span class="n">level</span><span class="o">=</span><span class="n">level</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="blockguard">
<h3>BlockGuard<a class="headerlink" href="#blockguard" title="Permalink to this headline">¶</a></h3>
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">BlockGuard</code><span class="sig-paren">(</span><em>main_program</em><span class="sig-paren">)</span></dt>
<dd><p>BlockGuard class.</p>
<p>BlockGuard class is used to create a sub-block in a program by
using the Python <cite>with</cite> keyword.</p>
</dd></dl>

</div>
<div class="section" id="blockguardwithcompletion">
<h3>BlockGuardWithCompletion<a class="headerlink" href="#blockguardwithcompletion" title="Permalink to this headline">¶</a></h3>
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">BlockGuardWithCompletion</code><span class="sig-paren">(</span><em>rnn</em><span class="sig-paren">)</span></dt>
<dd><p>BlockGuardWithCompletion class.</p>
<p>BlockGuardWithCompletion class is used to create an op with a block in a program.</p>
</dd></dl>

</div>
<div class="section" id="staticrnnmemorylink">
<h3>StaticRNNMemoryLink<a class="headerlink" href="#staticrnnmemorylink" title="Permalink to this headline">¶</a></h3>
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">StaticRNNMemoryLink</code><span class="sig-paren">(</span><em>init</em>, <em>pre_mem</em>, <em>mem=None</em><span class="sig-paren">)</span></dt>
<dd><p>StaticRNNMemoryLink class.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first last simple">
<li><strong>init</strong> &#8211; the initial variable for Memory</li>
<li><strong>init</strong> &#8211; Variable</li>
<li><strong>pre_mem</strong> &#8211; the memory variable in previous time step</li>
<li><strong>pre_mem</strong> &#8211; Variable</li>
<li><strong>mem</strong> &#8211; the memory variable in current time step</li>
<li><strong>mem</strong> &#8211; Variable</li>
</ul>
</td>
</tr>
</tbody>
</table>
<p>StaticRNNMemoryLink class is used to create a link between two
memory cells of a StaticRNN.</p>
</dd></dl>

</div>
<div class="section" id="whileguard">
<h3>WhileGuard<a class="headerlink" href="#whileguard" title="Permalink to this headline">¶</a></h3>
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">WhileGuard</code><span class="sig-paren">(</span><em>while_op</em><span class="sig-paren">)</span></dt>
<dd></dd></dl>

</div>
<div class="section" id="while">
<h3>While<a class="headerlink" href="#while" title="Permalink to this headline">¶</a></h3>
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">While</code><span class="sig-paren">(</span><em>cond</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd></dd></dl>

</div>
<div class="section" id="lod-rank-table">
<h3>lod_rank_table<a class="headerlink" href="#lod-rank-table" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">lod_rank_table</code><span class="sig-paren">(</span><em>x</em>, <em>level=0</em><span class="sig-paren">)</span></dt>
<dd><p>LoD Rank Table Operator. Given an input variable <strong>x</strong> and a level number
of LoD, this layer creates a LodRankTable object. A LoDRankTable object
contains a list of bi-element tuples. Each tuple consists of an index and
a length, both of which are int type. Refering to specified level of LoD,
the index is the sequence index number and the length representes the
sequence length. Please note that the list is ranked in descending order by
the length. The following is an example:</p>
<blockquote>
<div><div class="highlight-text"><div class="highlight"><pre><span></span>x is a LoDTensor:
    x.lod = [[0,                2, 3],
             [0,             5, 6, 7]]
    x.data = [a, b, c, d, e, f, g]

1. set level to 0:
    Create lod rank table:
        lod_rank_table_obj = lod_rank_table(x, level=0)

    Get:
        lod_rank_table_obj.items() = [(0, 2), (1, 1)]

2. set level to 1:
    Create lod rank table:
        lod_rank_table_obj = lod_rank_table(x, level=1)

    Get:
        lod_rank_table_obj.items() = [(0, 5), (1, 1), (2, 1)]
</pre></div>
</div>
</div></blockquote>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> (<em>Variable</em>) &#8211; Input variable, a LoDTensor based which to create the lod
rank table.</li>
<li><strong>level</strong> (<em>int</em>) &#8211; Specify the LoD level, on which to create the lod rank
table.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The created LoDRankTable object.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">x</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;x&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">10</span><span class="p">],</span>
                <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">,</span> <span class="n">lod_level</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
<span class="n">out</span> <span class="o">=</span> <span class="n">layers</span><span class="o">.</span><span class="n">lod_rank_table</span><span class="p">(</span><span class="n">x</span><span class="o">=</span><span class="n">x</span><span class="p">,</span> <span class="n">level</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="max-sequence-len">
<h3>max_sequence_len<a class="headerlink" href="#max-sequence-len" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">max_sequence_len</code><span class="sig-paren">(</span><em>rank_table</em><span class="sig-paren">)</span></dt>
<dd><p>Max Sequence Len Operator. Given a LoDRankTable object, this layer
returns the max length of a batch of sequences. In fact, a LoDRankTable
object contains a list of tuples(&lt;sequence index, sequence length&gt;) and
the list is already sorted by sequence length in descending order, so the
operator just returns the sequence length of the first tuple element.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>rank_table</strong> (<em>Variable</em>) &#8211; Input variable which is a LoDRankTable object.</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">The max length of sequence.</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body">Variable</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">x</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;x&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">10</span><span class="p">],</span>
                <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">,</span> <span class="n">lod_level</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
<span class="n">rank_table</span> <span class="o">=</span> <span class="n">layers</span><span class="o">.</span><span class="n">lod_rank_table</span><span class="p">(</span><span class="n">x</span><span class="o">=</span><span class="n">x</span><span class="p">,</span> <span class="n">level</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
<span class="n">max_seq_len</span> <span class="o">=</span> <span class="n">layers</span><span class="o">.</span><span class="n">max_sequence_len</span><span class="p">(</span><span class="n">rank_table</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="topk">
<h3>topk<a class="headerlink" href="#topk" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">topk</code><span class="sig-paren">(</span><em>input</em>, <em>k</em><span class="sig-paren">)</span></dt>
<dd><p><strong>topk</strong></p>
<p>This function performs the operation that selects the k entries in the input
vector and outputs their values and indices as vectors. Thus topk_out[j] is
the j-th largest entry in input, and its index is topk_indices[j]</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable|list</em>) &#8211; The input tensor that has all the data.</li>
<li><strong>k</strong> (<em>int</em>) &#8211; The number of top elements that the function will pick.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first"><dl class="docutils">
<dt>The variable of type array that contains the k largest entries</dt>
<dd><p class="first last">from input.</p>
</dd>
<dt>Variable: The variable of type array that contains the indices of k</dt>
<dd><p class="first last">largest entries from input.</p>
</dd>
</dl>
</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">x</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;x&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">10</span><span class="p">])</span>
<span class="n">k</span> <span class="o">=</span> <span class="mi">5</span>
<span class="n">array</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">topk</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">k</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="lod-tensor-to-array">
<h3>lod_tensor_to_array<a class="headerlink" href="#lod-tensor-to-array" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">lod_tensor_to_array</code><span class="sig-paren">(</span><em>x</em>, <em>table</em><span class="sig-paren">)</span></dt>
<dd><p>Convert a LOD_TENSOR to an LOD_TENSOR_ARRAY.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> (<em>Variable|list</em>) &#8211; The LOD tensor to be converted to a LOD tensor array.</li>
<li><strong>table</strong> (<em>ParamAttr|list</em>) &#8211; The variable that stores the level of lod
which is ordered by sequence length in
descending order.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first"><dl class="docutils">
<dt>The variable of type array that has been converted from a</dt>
<dd><p class="first last">tensor.</p>
</dd>
</dl>
</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">x</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;x&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">10</span><span class="p">])</span>
<span class="n">table</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">lod_rank_table</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">level</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
<span class="n">array</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">lod_tensor_to_array</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">table</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="array-to-lod-tensor">
<h3>array_to_lod_tensor<a class="headerlink" href="#array-to-lod-tensor" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">array_to_lod_tensor</code><span class="sig-paren">(</span><em>x</em>, <em>table</em><span class="sig-paren">)</span></dt>
<dd><p>Convert a LoD_Tensor_Aarry to an LoDTensor.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> (<em>Variable|list</em>) &#8211; The lod tensor array to be converted to a tensor.</li>
<li><strong>table</strong> (<em>ParamAttr|list</em>) &#8211; The variable that stores the level of lod
which is ordered by sequence length in
descending order.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first"><dl class="docutils">
<dt>The variable of type tensor that has been converted</dt>
<dd><p class="first last">from an array.</p>
</dd>
</dl>
</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">x</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;x&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">10</span><span class="p">])</span>
<span class="n">table</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">lod_rank_table</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">level</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
<span class="n">array</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">lod_tensor_to_array</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">table</span><span class="p">)</span>
<span class="n">lod_tensor</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">array_to_lod_tensor</span><span class="p">(</span><span class="n">array</span><span class="p">,</span> <span class="n">table</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="increment">
<h3>increment<a class="headerlink" href="#increment" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">increment</code><span class="sig-paren">(</span><em>x</em>, <em>value=1.0</em>, <em>in_place=True</em><span class="sig-paren">)</span></dt>
<dd><p>This function performs an operation that increments each value in the
input <span class="math">\(x\)</span> by an amount: <span class="math">\(value\)</span> as mentioned in the input
parameter. This operation is performed in-place by default.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> (<em>Variable|list</em>) &#8211; The tensor that has the input values.</li>
<li><strong>value</strong> (<em>float</em>) &#8211; The amount by which the values should be incremented.</li>
<li><strong>in_place</strong> (<em>bool</em>) &#8211; If the increment should be performed in-place.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first"><dl class="docutils">
<dt>The tensor variable storing the transformation of</dt>
<dd><p class="first last">element-wise increment of each value in the input.</p>
</dd>
</dl>
</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">data</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;data&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">32</span><span class="p">,</span> <span class="mi">32</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">)</span>
<span class="n">data</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">increment</span><span class="p">(</span><span class="n">x</span><span class="o">=</span><span class="n">data</span><span class="p">,</span> <span class="n">value</span><span class="o">=</span><span class="mf">3.0</span><span class="p">,</span> <span class="n">in_place</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="array-write">
<h3>array_write<a class="headerlink" href="#array-write" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">array_write</code><span class="sig-paren">(</span><em>x</em>, <em>i</em>, <em>array=None</em><span class="sig-paren">)</span></dt>
<dd><p>This function writes the given input variable to the specified position
indicating by the arrary index to an output LOD_TENSOR_ARRAY. If the
output LOD_TENSOR_ARRAY is not given(None), a new one will be created and
returned.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> (<em>Variable|list</em>) &#8211; The input tensor from which the data will be read.</li>
<li><strong>i</strong> (<em>Variable|list</em>) &#8211; The index of the output LOD_TENSOR_ARRAY, pointing to
the position to which the input tensor will be
written.</li>
<li><strong>array</strong> (<em>Variable|list</em>) &#8211; The output LOD_TENSOR_ARRAY to which the input
tensor will be written. If this parameter is
NONE, a new LOD_TENSOR_ARRAY will be created and
returned.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The output LOD_TENSOR_ARRAY where the input tensor is written.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
</dd></dl>

</div>
<div class="section" id="create-array">
<h3>create_array<a class="headerlink" href="#create-array" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">create_array</code><span class="sig-paren">(</span><em>dtype</em><span class="sig-paren">)</span></dt>
<dd><p>This function creates an array of type <span class="math">\(LOD_TENSOR_ARRAY\)</span> using the
LayerHelper.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>dtype</strong> (<em>int|float</em>) &#8211; The data type of the elements in the array.</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">The tensor variable storing the elements of data type.</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body">Variable</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">data</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">create_array</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="less-than">
<h3>less_than<a class="headerlink" href="#less-than" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">less_than</code><span class="sig-paren">(</span><em>x</em>, <em>y</em>, <em>cond=None</em>, <em>**ignored</em><span class="sig-paren">)</span></dt>
<dd><p><strong>Less than</strong></p>
<p>This layer returns the truth value of <span class="math">\(x &lt; y\)</span> elementwise.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> (<em>Variable</em>) &#8211; First operand of <em>less_than</em></li>
<li><strong>y</strong> (<em>Variable</em>) &#8211; Second operand of <em>less_than</em></li>
<li><strong>cond</strong> (<em>Variable|None</em>) &#8211; Optional output variable to store the result of <em>less_than</em></li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The tensor variable storing the output of <em>less_than</em>.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">less</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">less_than</span><span class="p">(</span><span class="n">x</span><span class="o">=</span><span class="n">label</span><span class="p">,</span> <span class="n">y</span><span class="o">=</span><span class="n">limit</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="array-read">
<h3>array_read<a class="headerlink" href="#array-read" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">array_read</code><span class="sig-paren">(</span><em>array</em>, <em>i</em><span class="sig-paren">)</span></dt>
<dd><p>This function performs the operation to read the data in as an
LOD_TENSOR_ARRAY.
:param array: The input tensor that will be written to an array.
:type array: Variable|list
:param i: The subscript index in tensor array, that points the</p>
<blockquote>
<div>place where data will be written to.</div></blockquote>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Returns:</th><td class="field-body">The tensor type variable that has the data written to it.</td>
</tr>
<tr class="field-even field"><th class="field-name">Return type:</th><td class="field-body">Variable</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
</dd></dl>

</div>
<div class="section" id="shrink-memory">
<h3>shrink_memory<a class="headerlink" href="#shrink-memory" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">shrink_memory</code><span class="sig-paren">(</span><em>x</em>, <em>i</em>, <em>table</em><span class="sig-paren">)</span></dt>
<dd><p>This function creates an operator to shrink_rnn_memory using the RankTable
as mentioned in the input parameter.</p>
</dd></dl>

</div>
<div class="section" id="array-length">
<h3>array_length<a class="headerlink" href="#array-length" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">array_length</code><span class="sig-paren">(</span><em>array</em><span class="sig-paren">)</span></dt>
<dd><p>This function performs the operation to find the length of the input
LOD_TENSOR_ARRAY.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>array</strong> (<em>LOD_TENSOR_ARRAY</em>) &#8211; The input array that will be used
to compute the length.</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">The length of the input LoDTensorArray.</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body">Variable</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
</dd></dl>

</div>
<div class="section" id="ifelse">
<h3>IfElse<a class="headerlink" href="#ifelse" title="Permalink to this headline">¶</a></h3>
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">IfElse</code><span class="sig-paren">(</span><em>cond</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd></dd></dl>

</div>
<div class="section" id="dynamicrnn">
<h3>DynamicRNN<a class="headerlink" href="#dynamicrnn" title="Permalink to this headline">¶</a></h3>
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">DynamicRNN</code><span class="sig-paren">(</span><em>name=None</em><span class="sig-paren">)</span></dt>
<dd></dd></dl>

</div>
<div class="section" id="conditionalblock">
<h3>ConditionalBlock<a class="headerlink" href="#conditionalblock" title="Permalink to this headline">¶</a></h3>
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">ConditionalBlock</code><span class="sig-paren">(</span><em>inputs</em>, <em>is_scalar_condition=False</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd></dd></dl>

</div>
<div class="section" id="staticrnn">
<h3>StaticRNN<a class="headerlink" href="#staticrnn" title="Permalink to this headline">¶</a></h3>
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">StaticRNN</code><span class="sig-paren">(</span><em>name=None</em><span class="sig-paren">)</span></dt>
<dd><p>StaticRNN class.</p>
<p>StaticRNN class is used to create a StaticRNN. The RNN will have its
own parameters like inputs, outputs, memories, status and length.</p>
<dl class="method">
<dt>
<code class="descname">memory</code><span class="sig-paren">(</span><em>init=None</em>, <em>shape=None</em>, <em>batch_ref=None</em>, <em>init_value=0.0</em>, <em>init_batch_dim_idx=0</em>, <em>ref_batch_dim_idx=1</em><span class="sig-paren">)</span></dt>
<dd><table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first last simple">
<li><strong>init</strong> &#8211; boot memory, if not set, a shape, batch_ref must be provided</li>
<li><strong>shape</strong> &#8211; shape of the boot memory</li>
<li><strong>batch_ref</strong> &#8211; batch size reference variable</li>
<li><strong>init_value</strong> &#8211; the init value of boot memory</li>
<li><strong>init_batch_dim_idx</strong> &#8211; the index of batch size in init&#8217;s dimension</li>
<li><strong>ref_batch_dim_idx</strong> &#8211; the index of batch size in batch_ref&#8217;s dimension</li>
</ul>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</dd></dl>

</div>
<div class="section" id="reorder-lod-tensor-by-rank">
<h3>reorder_lod_tensor_by_rank<a class="headerlink" href="#reorder-lod-tensor-by-rank" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">reorder_lod_tensor_by_rank</code><span class="sig-paren">(</span><em>x</em>, <em>rank_table</em><span class="sig-paren">)</span></dt>
<dd><p>ReorderLoDTensorByRankTable operator.</p>
<p>Input(X) is a batch of sequences. Input(RankTable) stores new orders of the
input sequence batch. The reorder_lod_tensor_by_rank operator reorders the
Input(X) according to the information provided by Input(RankTable).</p>
<p>For example:</p>
<p>If the indices stored in the Input(RankTable) are [3, 0, 2, 1], the
Input(X) will be reordered that the fourth sequence in Input(X) will become the
first one, and then followed by the original first, third, and the second one.</p>
<p>This is:
X = [Seq0, Seq1, Seq2, Seq3]. The indices in RankTable are [3, 0, 2, 1].
Out =  [Seq3, Seq0, Seq2, Seq1] with a new LoD information.</p>
<p>If the LoD information of Input(X) is empty, this means Input(X) is not sequence
data. This is also identical to a batch of sequences where each sequence has a
fixed length 1. In this case, the reorder_lod_tensor_by_rank operator reorders
each slice of Input(X) along the first axis according to Input(RankTable).</p>
<p>This is:
X = [Slice0, Slice1, Slice2, Slice3] and its LoD information is empty. The
indices in RankTable are [3, 0, 2, 1].
Out = [Slice3, Slice0, Slice2, Slice1] with no LoD information is appended.</p>
<p>NOTE: This operator sorts Input(X) according to a given LoDRankTable which does
not need to be calculated according to Input(X). It can be calculated according
to another different sequence, and then this operator sorts Input(X) according
to the given LoDRankTable.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; (LoDTensor), the input lod tensor to be reordered according to Input(RankTable).
Duplicable: False  Optional: False</li>
<li><strong>rank_table</strong> &#8211; (LoDRankTable), the rank table according to which Input(X) is reordered.
Duplicable: False  Optional: False</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">(LoDTensor), the reordered lod tensor.</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="paralleldo">
<h3>ParallelDo<a class="headerlink" href="#paralleldo" title="Permalink to this headline">¶</a></h3>
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">ParallelDo</code><span class="sig-paren">(</span><em>places</em>, <em>use_nccl=False</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd><p>ParallelDo class.</p>
<p>ParallelDo class is used to create a ParallelDo.</p>
</dd></dl>

</div>
<div class="section" id="print">
<h3>Print<a class="headerlink" href="#print" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">Print</code><span class="sig-paren">(</span><em>input</em>, <em>first_n=-1</em>, <em>message=None</em>, <em>summarize=-1</em>, <em>print_tensor_name=True</em>, <em>print_tensor_type=True</em>, <em>print_tensor_shape=True</em>, <em>print_tensor_lod=True</em>, <em>print_phase='both'</em><span class="sig-paren">)</span></dt>
<dd><p><strong>Print operator</strong></p>
<p>This creates a print op that will print when a tensor is accessed.</p>
<p>Wraps the tensor passed in so that whenever that a tensor is accessed,
the message <cite>message</cite> is printed, along with the current value of the
tensor <cite>t</cite>.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable</em>) &#8211; A Tensor to print.</li>
<li><strong>summarize</strong> (<em>int</em>) &#8211; Print this number of elements in the tensor, will print
all if left is negative.</li>
<li><strong>message</strong> (<em>str</em>) &#8211; A string message to print as a prefix.</li>
<li><strong>first_n</strong> (<em>int</em>) &#8211; Only log <cite>first_n</cite> number of times.</li>
<li><strong>print_tensor_name</strong> (<em>bool</em>) &#8211; Print the tensor name.</li>
<li><strong>print_tensor_type</strong> (<em>bool</em>) &#8211; Print the tensor type.</li>
<li><strong>print_tensor_shape</strong> (<em>bool</em>) &#8211; Print the tensor shape.</li>
<li><strong>print_tensor_lod</strong> (<em>bool</em>) &#8211; Print the tensor lod.</li>
<li><strong>print_phase</strong> (<em>str</em>) &#8211; Which phase to displace, including &#8216;forward&#8217;,
&#8216;backward&#8217; and &#8216;both&#8217;. If set to &#8216;backward&#8217; or &#8216;both&#8217;, will
print the gradients of input tensor.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">Output tensor, same data with input tensor.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span>
</pre></div>
</div>
<p>value = some_layer(...)
Print(value, summarize=10,</p>
<blockquote>
<div>message=&#8221;The content of some_layer: &#8221;)</div></blockquote>
</dd></dl>

</div>
</div>
<div class="section" id="device">
<h2>device<a class="headerlink" href="#device" title="Permalink to this headline">¶</a></h2>
<div class="section" id="get-places">
<h3>get_places<a class="headerlink" href="#get-places" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">get_places</code><span class="sig-paren">(</span><em>device_count=None</em>, <em>device_type=None</em><span class="sig-paren">)</span></dt>
<dd><p>Returns a list of places based on flags. The list will be used for parallel
execution.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>device_count</strong> (<em>INT</em>) &#8211; device count</li>
<li><strong>device_type</strong> (<em>STRING</em>) &#8211; device type</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">vector of Place</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
</div>
<div class="section" id="io">
<h2>io<a class="headerlink" href="#io" title="Permalink to this headline">¶</a></h2>
<div class="section" id="data">
<h3>data<a class="headerlink" href="#data" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">data</code><span class="sig-paren">(</span><em>name</em>, <em>shape</em>, <em>append_batch_size=True</em>, <em>dtype='float32'</em>, <em>lod_level=0</em>, <em>type=VarType.LOD_TENSOR</em>, <em>stop_gradient=True</em><span class="sig-paren">)</span></dt>
<dd><p><strong>Data Layer</strong></p>
<p>This function takes in the input and based on whether data has
to be returned back as a minibatch, it creates the global variable by using
the helper functions. The global variables can be accessed by all the
following operators in the graph.</p>
<p>All the input variables of this function are passed in as local variables
to the LayerHelper constructor.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>name</strong> (<em>str</em>) &#8211; The name/alias of the function</li>
<li><strong>shape</strong> (<em>list</em>) &#8211; Tuple declaring the shape.</li>
<li><strong>append_batch_size</strong> (<em>bool</em>) &#8211; Whether or not to append the data as a batch.</li>
<li><strong>dtype</strong> (<em>int|float</em>) &#8211; The type of data : float32, float_16, int etc</li>
<li><strong>type</strong> (<em>VarType</em>) &#8211; The output type. By default it is LOD_TENSOR.</li>
<li><strong>lod_level</strong> (<em>int</em>) &#8211; The LoD Level. 0 means the input data is not a sequence.</li>
<li><strong>main_program</strong> (<em>Program</em>) &#8211; Name of the main program that calls this</li>
<li><strong>startup_program</strong> (<em>Program</em>) &#8211; Name of the startup program</li>
<li><strong>stop_gradient</strong> (<em>bool</em>) &#8211; A boolean that mentions whether gradient should flow.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The global variable that gives access to the data.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">data</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;x&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">784</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="blockguardserv">
<h3>BlockGuardServ<a class="headerlink" href="#blockguardserv" title="Permalink to this headline">¶</a></h3>
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">BlockGuardServ</code><span class="sig-paren">(</span><em>server</em><span class="sig-paren">)</span></dt>
<dd><p>BlockGuardServ class.</p>
<p>BlockGuardServ class is used to create an op with a block in a program.</p>
</dd></dl>

</div>
<div class="section" id="listenandserv">
<h3>ListenAndServ<a class="headerlink" href="#listenandserv" title="Permalink to this headline">¶</a></h3>
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">ListenAndServ</code><span class="sig-paren">(</span><em>endpoint</em>, <em>fan_in=1</em>, <em>optimizer_mode=True</em><span class="sig-paren">)</span></dt>
<dd><p>ListenAndServ class.</p>
<p>ListenAndServ class is used to wrap listen_and_serv op to create a server
which can receive variables from clients and run a block.</p>
</dd></dl>

</div>
<div class="section" id="send">
<h3>Send<a class="headerlink" href="#send" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">Send</code><span class="sig-paren">(</span><em>endpoints</em>, <em>send_vars</em>, <em>get_vars</em><span class="sig-paren">)</span></dt>
<dd><p>Send layer</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first last simple">
<li><strong>endpoints</strong> &#8211; comma seperated IP:PORT pairs in the order
of send_vars to send</li>
<li><strong>send_vars</strong> &#8211; vars to send</li>
<li><strong>get_vars</strong> &#8211; vars to get from server after send completes.</li>
</ul>
</td>
</tr>
</tbody>
</table>
<p>Send variables to the server side, and get vars from server
side when server have finished running server side program.</p>
</dd></dl>

</div>
</div>
<div class="section" id="nn">
<h2>nn<a class="headerlink" href="#nn" title="Permalink to this headline">¶</a></h2>
<div class="section" id="fc">
<h3>fc<a class="headerlink" href="#fc" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">fc</code><span class="sig-paren">(</span><em>input</em>, <em>size</em>, <em>num_flatten_dims=1</em>, <em>param_attr=None</em>, <em>bias_attr=None</em>, <em>act=None</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd><p><strong>Fully Connected Layer</strong></p>
<p>The fully connected layer can take multiple tensors as its inputs. It
creates a variable (one for each input tensor) called weights for each
input tensor, which represents a fully connected weight matrix from
each input unit to each output unit. The fully connected layer
multiplies each input tensor with its coresponding weight to produce
an output Tensor. If multiple input tensors are given, the results of
multiple multiplications will be sumed up. If bias_attr is not None,
a biases variable will be created and added to the output. Finally,
if activation is not None, it will be applied to the output as well.</p>
<p>This process can be formulated as follows:</p>
<div class="math">
\[Out = Act({\sum_{i=0}^{N-1}X_iW_i + b})\]</div>
<p>In the above equation:</p>
<ul class="simple">
<li><span class="math">\(N\)</span>: Number of the input.</li>
<li><span class="math">\(X_i\)</span>: The input tensor.</li>
<li><span class="math">\(W\)</span>: The weights created by this layer.</li>
<li><span class="math">\(b\)</span>: The bias parameter created by this layer (if needed).</li>
<li><span class="math">\(Act\)</span>: The activation funtion.</li>
<li><span class="math">\(Out\)</span>: The output tensor.</li>
</ul>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable|list</em>) &#8211; The input tensor(s) to the fully connected layer.</li>
<li><strong>size</strong> (<em>int</em>) &#8211; The number of output units in the fully connected layer.</li>
<li><strong>num_flatten_dims</strong> (<em>int</em>) &#8211; The fc layer can accept an input tensor with more
than two dimensions. If this happens, the
multidimensional tensor will first be flattened
into a 2-dimensional matrix. The parameter
<cite>num_flatten_dims</cite> determines how the input tensor
is flattened: the first <cite>num_flatten_dims</cite>
(inclusive, index starts from 1) dimensions will
be flatten to form the first dimension of the
final matrix (height of the matrix), and the rest
<cite>rank(X) - num_flatten_dims</cite> dimensions are
flattened to form the second dimension of the
final matrix (width of the matrix). For example,
suppose <cite>X</cite> is a 6-dimensional tensor with a shape
[2, 3, 4, 5, 6], and <cite>num_flatten_dims</cite> = 3. Then,
the flattened matrix will have a shape
[2 x 3 x 4, 5 x 6] = [24, 30]. By default,
<cite>num_flatten_dims</cite> is set to 1.</li>
<li><strong>param_attr</strong> (<em>ParamAttr|list</em>) &#8211; The parameter attribute for learnable
parameters/weights of the fully connected
layer.</li>
<li><strong>param_initializer</strong> (<em>ParamAttr|list</em>) &#8211; The initializer used for the
weight/parameter. If set None,
XavierInitializer() will be used.</li>
<li><strong>bias_attr</strong> (<em>ParamAttr|list</em>) &#8211; The parameter attribute for the bias parameter
for this layer. If set None, no bias will be
added to the output units.</li>
<li><strong>bias_initializer</strong> (<em>ParamAttr|list</em>) &#8211; The initializer used for the bias.
If set None, then ConstantInitializer()
will be used.</li>
<li><strong>act</strong> (<em>str</em>) &#8211; Activation to be applied to the output of the fully connected
layer.</li>
<li><strong>name</strong> (<em>str</em>) &#8211; Name/alias of the fully connected layer.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The output tensor variable.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first">Variable</p>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Raises:</th><td class="field-body"><p class="first last"><code class="xref py py-exc docutils literal"><span class="pre">ValueError</span></code> &#8211; If rank of the input tensor is less than 2.</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">data</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s2">&quot;data&quot;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">32</span><span class="p">,</span> <span class="mi">32</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s2">&quot;float32&quot;</span><span class="p">)</span>
<span class="n">fc</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">fc</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">data</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="mi">1000</span><span class="p">,</span> <span class="n">act</span><span class="o">=</span><span class="s2">&quot;tanh&quot;</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="embedding">
<h3>embedding<a class="headerlink" href="#embedding" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">embedding</code><span class="sig-paren">(</span><em>input</em>, <em>size</em>, <em>is_sparse=False</em>, <em>padding_idx=None</em>, <em>param_attr=None</em>, <em>dtype='float32'</em><span class="sig-paren">)</span></dt>
<dd><p><strong>Embedding Layer</strong></p>
<p>This layer is used to lookup embeddings of IDs, provided by <code class="xref py py-attr docutils literal"><span class="pre">input</span></code>, in
a lookup table. The result of this lookup is the embedding of each ID in the
<code class="xref py py-attr docutils literal"><span class="pre">input</span></code>.</p>
<p>All the input variables are passed in as local variables to the LayerHelper
constructor.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable</em>) &#8211; The tensor variable containing the IDs.</li>
<li><strong>size</strong> (<em>tuple|list</em>) &#8211; The shape of the look up table parameter. It should
have two elements which indicate the size of the dictionary of
embeddings and the size of each embedding vector respectively.</li>
<li><strong>is_sparse</strong> (<em>bool</em>) &#8211; The flag indicating whether to use sparse update.</li>
<li><strong>padding_idx</strong> (<em>int|long|None</em>) &#8211; If <code class="xref py py-attr docutils literal"><span class="pre">None</span></code>, it makes no effect to lookup.
Otherwise the given <code class="xref py py-attr docutils literal"><span class="pre">padding_idx</span></code> indicates padding the output
with zeros whenever lookup encounters it in <code class="xref py py-attr docutils literal"><span class="pre">input</span></code>. If
<span class="math">\(padding_idx &lt; 0\)</span>, the padding_idx to use in lookup is
<span class="math">\(size[0] + dim\)</span>.</li>
<li><strong>param_attr</strong> (<em>ParamAttr</em>) &#8211; Parameters for this layer</li>
<li><strong>dtype</strong> (<em>np.dtype|core.VarDesc.VarType|str</em>) &#8211; The type of data : float32, float_16, int etc</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The tensor variable storing the embeddings of the                   supplied inputs.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">dict_size</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">dataset</span><span class="o">.</span><span class="n">ids</span><span class="p">)</span>
<span class="n">data</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;ids&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">32</span><span class="p">,</span> <span class="mi">32</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">)</span>
<span class="n">fc</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">embedding</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">data</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">[</span><span class="n">dict_size</span><span class="p">,</span> <span class="mi">16</span><span class="p">])</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="dynamic-lstm">
<h3>dynamic_lstm<a class="headerlink" href="#dynamic-lstm" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">dynamic_lstm</code><span class="sig-paren">(</span><em>input</em>, <em>size</em>, <em>param_attr=None</em>, <em>bias_attr=None</em>, <em>use_peepholes=True</em>, <em>is_reverse=False</em>, <em>gate_activation='sigmoid'</em>, <em>cell_activation='tanh'</em>, <em>candidate_activation='tanh'</em>, <em>dtype='float32'</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd><p><strong>Dynamic LSTM Layer</strong></p>
<p>The defalut implementation is diagonal/peephole connection
(<a class="reference external" href="https://arxiv.org/pdf/1402.1128.pdf">https://arxiv.org/pdf/1402.1128.pdf</a>), the formula is as follows:</p>
<div class="math">
\[ \begin{align}\begin{aligned}i_t &amp; = \sigma(W_{ix}x_{t} + W_{ih}h_{t-1} + W_{ic}c_{t-1} + b_i)\\f_t &amp; = \sigma(W_{fx}x_{t} + W_{fh}h_{t-1} + W_{fc}c_{t-1} + b_f)\\\tilde{c_t} &amp; = act_g(W_{cx}x_t + W_{ch}h_{t-1} + b_c)\\o_t &amp; = \sigma(W_{ox}x_{t} + W_{oh}h_{t-1} + W_{oc}c_t + b_o)\\c_t &amp; = f_t \odot c_{t-1} + i_t \odot \tilde{c_t}\\h_t &amp; = o_t \odot act_h(c_t)\end{aligned}\end{align} \]</div>
<p>where the <span class="math">\(W\)</span> terms denote weight matrices (e.g. <span class="math">\(W_{xi}\)</span> is
the matrix of weights from the input gate to the input), <span class="math">\(W_{ic},     W_{fc}, W_{oc}\)</span> are diagonal weight matrices for peephole connections. In
our implementation, we use vectors to reprenset these diagonal weight
matrices. The <span class="math">\(b\)</span> terms denote bias vectors (<span class="math">\(b_i\)</span> is the input
gate bias vector), <span class="math">\(\sigma\)</span> is the non-linear activations, such as
logistic sigmoid function, and <span class="math">\(i, f, o\)</span> and <span class="math">\(c\)</span> are the input
gate, forget gate, output gate, and cell activation vectors, respectively,
all of which have the same size as the cell output activation vector <span class="math">\(h\)</span>.</p>
<p>The <span class="math">\(\odot\)</span> is the element-wise product of the vectors. <span class="math">\(act_g\)</span>
and <span class="math">\(act_h\)</span> are the cell input and cell output activation functions
and <cite>tanh</cite> is usually used for them. <span class="math">\(\tilde{c_t}\)</span> is also called
candidate hidden state, which is computed based on the current input and
the previous hidden state.</p>
<p>Set <cite>use_peepholes</cite> to <cite>False</cite> to disable peephole connection. The formula
is omitted here, please refer to the paper
<a class="reference external" href="http://www.bioinf.jku.at/publications/older/2604.pdf">http://www.bioinf.jku.at/publications/older/2604.pdf</a> for details.</p>
<p>Note that these <span class="math">\(W_{xi}x_{t}, W_{xf}x_{t}, W_{xc}x_{t}, W_{xo}x_{t}\)</span>
operations on the input <span class="math">\(x_{t}\)</span> are NOT included in this operator.
Users can choose to use fully-connect layer before LSTM layer.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable</em>) &#8211; The input of dynamic_lstm layer, which supports
variable-time length input sequence. The underlying
tensor in this Variable is a matrix with shape
(T X 4D), where T is the total time steps in this
mini-batch, D is the hidden size.</li>
<li><strong>size</strong> (<em>int</em>) &#8211; 4 * hidden size.</li>
<li><strong>param_attr</strong> (<em>ParamAttr|None</em>) &#8211; <p>The parameter attribute for the learnable
hidden-hidden weights.</p>
<ul>
<li>Weights = {<span class="math">\(W_{ch}, W_{ih},                                                 W_{fh}, W_{oh}\)</span>}</li>
<li>The shape is (D x 4D), where D is the hidden
size.</li>
</ul>
</li>
<li><strong>bias_attr</strong> (<em>ParamAttr|None</em>) &#8211; <p>The bias attribute for the learnable bias
weights, which contains two parts, input-hidden
bias weights and peephole connections weights if
setting <cite>use_peepholes</cite> to <cite>True</cite>.</p>
<ol class="arabic">
<li><cite>use_peepholes = False</cite></li>
</ol>
<blockquote>
<div><ul>
<li>Biases = {<span class="math">\(b_c, b_i, b_f, b_o\)</span>}.</li>
<li>The shape is (1 x 4D).</li>
</ul>
</div></blockquote>
<ol class="arabic" start="2">
<li><cite>use_peepholes = True</cite></li>
</ol>
<blockquote>
<div><ul>
<li>Biases = { <span class="math">\(b_c, b_i, b_f, b_o, W_{ic},                                                  W_{fc}, W_{oc}\)</span>}.</li>
<li>The shape is (1 x 7D).</li>
</ul>
</div></blockquote>
</li>
<li><strong>use_peepholes</strong> (<em>bool</em>) &#8211; Whether to enable diagonal/peephole connections,
default <cite>True</cite>.</li>
<li><strong>is_reverse</strong> (<em>bool</em>) &#8211; Whether to compute reversed LSTM, default <cite>False</cite>.</li>
<li><strong>gate_activation</strong> (<em>str</em>) &#8211; The activation for input gate, forget gate and
output gate. Choices = [&#8220;sigmoid&#8221;, &#8220;tanh&#8221;, &#8220;relu&#8221;,
&#8220;identity&#8221;], default &#8220;sigmoid&#8221;.</li>
<li><strong>cell_activation</strong> (<em>str</em>) &#8211; The activation for cell output. Choices = [&#8220;sigmoid&#8221;,
&#8220;tanh&#8221;, &#8220;relu&#8221;, &#8220;identity&#8221;], default &#8220;tanh&#8221;.</li>
<li><strong>candidate_activation</strong> (<em>str</em>) &#8211; The activation for candidate hidden state.
Choices = [&#8220;sigmoid&#8221;, &#8220;tanh&#8221;, &#8220;relu&#8221;, &#8220;identity&#8221;],
default &#8220;tanh&#8221;.</li>
<li><strong>dtype</strong> (<em>str</em>) &#8211; Data type. Choices = [&#8220;float32&#8221;, &#8220;float64&#8221;], default &#8220;float32&#8221;.</li>
<li><strong>name</strong> (<em>str|None</em>) &#8211; A name for this layer(optional). If set None, the layer
will be named automatically.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The hidden state, and cell state of LSTM. The shape of both         is (T x D), and lod is the same with the <cite>input</cite>.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">tuple</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">hidden_dim</span> <span class="o">=</span> <span class="mi">512</span>
<span class="n">forward_proj</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">fc</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">input_seq</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">hidden_dim</span> <span class="o">*</span> <span class="mi">4</span><span class="p">,</span>
                               <span class="n">act</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">bias_attr</span><span class="o">=</span><span class="bp">None</span><span class="p">)</span>
<span class="n">forward</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">dynamic_lstm</span><span class="p">(</span>
    <span class="nb">input</span><span class="o">=</span><span class="n">forward_proj</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">hidden_dim</span> <span class="o">*</span> <span class="mi">4</span><span class="p">,</span> <span class="n">use_peepholes</span><span class="o">=</span><span class="bp">False</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="dynamic-lstmp">
<h3>dynamic_lstmp<a class="headerlink" href="#dynamic-lstmp" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">dynamic_lstmp</code><span class="sig-paren">(</span><em>input</em>, <em>size</em>, <em>proj_size</em>, <em>param_attr=None</em>, <em>bias_attr=None</em>, <em>use_peepholes=True</em>, <em>is_reverse=False</em>, <em>gate_activation='sigmoid'</em>, <em>cell_activation='tanh'</em>, <em>candidate_activation='tanh'</em>, <em>proj_activation='tanh'</em>, <em>dtype='float32'</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd><p><strong>Dynamic LSTMP Layer</strong></p>
<p>LSTMP (LSTM with recurrent projection) layer has a separate projection
layer after the LSTM layer, projecting the original hidden state to a
lower-dimensional one, which is proposed to reduce the number of total
parameters and furthermore computational complexity for the LSTM,
espeacially for the case that the size of output units is relative
large (<a class="reference external" href="https://research.google.com/pubs/archive/43905.pdf">https://research.google.com/pubs/archive/43905.pdf</a>).</p>
<p>The formula is as follows:</p>
<div class="math">
\[ \begin{align}\begin{aligned}i_t &amp; = \sigma(W_{ix}x_{t} + W_{ir}r_{t-1} + W_{ic}c_{t-1} + b_i)\\f_t &amp; = \sigma(W_{fx}x_{t} + W_{fr}r_{t-1} + W_{fc}c_{t-1} + b_f)\\\tilde{c_t} &amp; = act_g(W_{cx}x_t + W_{cr}r_{t-1} + b_c)\\o_t &amp; = \sigma(W_{ox}x_{t} + W_{or}r_{t-1} + W_{oc}c_t + b_o)\\c_t &amp; = f_t \odot c_{t-1} + i_t \odot \tilde{c_t}\\h_t &amp; = o_t \odot act_h(c_t)\\r_t &amp; = \overline{act_h}(W_{rh}h_t)\end{aligned}\end{align} \]</div>
<p>In the above formula:</p>
<ul class="simple">
<li><span class="math">\(W\)</span>: Denotes weight matrices (e.g. <span class="math">\(W_{xi}\)</span> is           the matrix of weights from the input gate to the input).</li>
<li><span class="math">\(W_{ic}\)</span>, <span class="math">\(W_{fc}\)</span>, <span class="math">\(W_{oc}\)</span>: Diagonal weight           matrices for peephole connections. In our implementation,           we use vectors to reprenset these diagonal weight matrices.</li>
<li><span class="math">\(b\)</span>: Denotes bias vectors (e.g. <span class="math">\(b_i\)</span> is the input gate           bias vector).</li>
<li><span class="math">\(\sigma\)</span>: The activation, such as logistic sigmoid function.</li>
<li><span class="math">\(i, f, o\)</span> and <span class="math">\(c\)</span>: The input gate, forget gate, output           gate, and cell activation vectors, respectively, all of which have           the same size as the cell output activation vector <span class="math">\(h\)</span>.</li>
<li><span class="math">\(h\)</span>: The hidden state.</li>
<li><span class="math">\(r\)</span>: The recurrent projection of the hidden state.</li>
<li><span class="math">\(\tilde{c_t}\)</span>: The candidate hidden state, whose           computation is based on the current input and previous hidden state.</li>
<li><span class="math">\(\odot\)</span>: The element-wise product of the vectors.</li>
<li><span class="math">\(act_g\)</span> and <span class="math">\(act_h\)</span>: The cell input and cell output           activation functions and <cite>tanh</cite> is usually used for them.</li>
<li><span class="math">\(\overline{act_h}\)</span>: The activation function for the projection           output, usually using <cite>identity</cite> or same as <span class="math">\(act_h\)</span>.</li>
</ul>
<p>Set <cite>use_peepholes</cite> to <cite>False</cite> to disable peephole connection. The formula
is omitted here, please refer to the paper
<a class="reference external" href="http://www.bioinf.jku.at/publications/older/2604.pdf">http://www.bioinf.jku.at/publications/older/2604.pdf</a> for details.</p>
<p>Note that these <span class="math">\(W_{xi}x_{t}, W_{xf}x_{t}, W_{xc}x_{t}, W_{xo}x_{t}\)</span>
operations on the input <span class="math">\(x_{t}\)</span> are NOT included in this operator.
Users can choose to use fully-connected layer before LSTMP layer.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable</em>) &#8211; The input of dynamic_lstmp layer, which supports
variable-time length input sequence. The underlying
tensor in this Variable is a matrix with shape
(T X 4D), where T is the total time steps in this
mini-batch, D is the hidden size.</li>
<li><strong>size</strong> (<em>int</em>) &#8211; 4 * hidden size.</li>
<li><strong>proj_size</strong> (<em>int</em>) &#8211; The size of projection output.</li>
<li><strong>param_attr</strong> (<em>ParamAttr|None</em>) &#8211; <p>The parameter attribute for the learnable
hidden-hidden weight and projection weight.</p>
<ul>
<li>Hidden-hidden weight = {<span class="math">\(W_{ch}, W_{ih},                                                 W_{fh}, W_{oh}\)</span>}.</li>
<li>The shape of hidden-hidden weight is (P x 4D),
where P is the projection size and D the hidden
size.</li>
<li>Projection weight = {<span class="math">\(W_{rh}\)</span>}.</li>
<li>The shape of projection weight is (D x P).</li>
</ul>
</li>
<li><strong>bias_attr</strong> (<em>ParamAttr|None</em>) &#8211; <p>The bias attribute for the learnable bias
weights, which contains two parts, input-hidden
bias weights and peephole connections weights if
setting <cite>use_peepholes</cite> to <cite>True</cite>.</p>
<ol class="arabic">
<li><cite>use_peepholes = False</cite></li>
</ol>
<blockquote>
<div><ul>
<li>Biases = {<span class="math">\(b_c, b_i, b_f, b_o\)</span>}.</li>
<li>The shape is (1 x 4D).</li>
</ul>
</div></blockquote>
<ol class="arabic" start="2">
<li><cite>use_peepholes = True</cite></li>
</ol>
<blockquote>
<div><ul>
<li>Biases = { <span class="math">\(b_c, b_i, b_f, b_o, W_{ic},                                                  W_{fc}, W_{oc}\)</span>}.</li>
<li>The shape is (1 x 7D).</li>
</ul>
</div></blockquote>
</li>
<li><strong>use_peepholes</strong> (<em>bool</em>) &#8211; Whether to enable diagonal/peephole connections,
default <cite>True</cite>.</li>
<li><strong>is_reverse</strong> (<em>bool</em>) &#8211; Whether to compute reversed LSTM, default <cite>False</cite>.</li>
<li><strong>gate_activation</strong> (<em>str</em>) &#8211; The activation for input gate, forget gate and
output gate. Choices = [&#8220;sigmoid&#8221;, &#8220;tanh&#8221;, &#8220;relu&#8221;,
&#8220;identity&#8221;], default &#8220;sigmoid&#8221;.</li>
<li><strong>cell_activation</strong> (<em>str</em>) &#8211; The activation for cell output. Choices = [&#8220;sigmoid&#8221;,
&#8220;tanh&#8221;, &#8220;relu&#8221;, &#8220;identity&#8221;], default &#8220;tanh&#8221;.</li>
<li><strong>candidate_activation</strong> (<em>str</em>) &#8211; The activation for candidate hidden state.
Choices = [&#8220;sigmoid&#8221;, &#8220;tanh&#8221;, &#8220;relu&#8221;, &#8220;identity&#8221;],
default &#8220;tanh&#8221;.</li>
<li><strong>proj_activation</strong> (<em>str</em>) &#8211; The activation for projection output.
Choices = [&#8220;sigmoid&#8221;, &#8220;tanh&#8221;, &#8220;relu&#8221;, &#8220;identity&#8221;],
default &#8220;tanh&#8221;.</li>
<li><strong>dtype</strong> (<em>str</em>) &#8211; Data type. Choices = [&#8220;float32&#8221;, &#8220;float64&#8221;], default &#8220;float32&#8221;.</li>
<li><strong>name</strong> (<em>str|None</em>) &#8211; A name for this layer(optional). If set None, the layer
will be named automatically.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The projection of hidden state, and cell state of LSTMP. The                shape of projection is (T x P), for the cell state which is                (T x D), and both LoD is the same with the <cite>input</cite>.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">tuple</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">hidden_dim</span><span class="p">,</span> <span class="n">proj_dim</span> <span class="o">=</span> <span class="mi">512</span><span class="p">,</span> <span class="mi">256</span>
<span class="n">fc_out</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">fc</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">input_seq</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">hidden_dim</span> <span class="o">*</span> <span class="mi">4</span><span class="p">,</span>
                         <span class="n">act</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">bias_attr</span><span class="o">=</span><span class="bp">None</span><span class="p">)</span>
<span class="n">proj_out</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">dynamic_lstmp</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">fc_out</span><span class="p">,</span>
                                         <span class="n">size</span><span class="o">=</span><span class="n">hidden_dim</span> <span class="o">*</span> <span class="mi">4</span><span class="p">,</span>
                                         <span class="n">proj_size</span><span class="o">=</span><span class="n">proj_dim</span><span class="p">,</span>
                                         <span class="n">use_peepholes</span><span class="o">=</span><span class="bp">False</span><span class="p">,</span>
                                         <span class="n">is_reverse</span><span class="o">=</span><span class="bp">True</span><span class="p">,</span>
                                         <span class="n">cell_activation</span><span class="o">=</span><span class="s2">&quot;tanh&quot;</span><span class="p">,</span>
                                         <span class="n">proj_activation</span><span class="o">=</span><span class="s2">&quot;tanh&quot;</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="dynamic-gru">
<h3>dynamic_gru<a class="headerlink" href="#dynamic-gru" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">dynamic_gru</code><span class="sig-paren">(</span><em>input</em>, <em>size</em>, <em>param_attr=None</em>, <em>bias_attr=None</em>, <em>is_reverse=False</em>, <em>gate_activation='sigmoid'</em>, <em>candidate_activation='tanh'</em>, <em>h_0=None</em><span class="sig-paren">)</span></dt>
<dd><p><strong>Dynamic GRU Layer</strong></p>
<p>Refer to <a class="reference external" href="https://arxiv.org/abs/1412.3555">Empirical Evaluation of Gated Recurrent Neural Networks on
Sequence Modeling</a></p>
<p>The formula is as follows:</p>
<div class="math">
\[ \begin{align}\begin{aligned}u_t &amp; = act_g(W_{ux}x_{t} + W_{uh}h_{t-1} + b_u)\\r_t &amp; = act_g(W_{rx}x_{t} + W_{rh}h_{t-1} + b_r)\\\tilde{h_t} &amp; = act_c(W_{cx}x_{t} + W_{ch}(r_t \odot h_{t-1}) + b_c)\\h_t &amp; = (1-u_t) \odot h_{t-1} + u_t \odot \tilde{h_t}\end{aligned}\end{align} \]</div>
<p>The <span class="math">\(\odot\)</span> is the element-wise product of the vectors. <span class="math">\(act_g\)</span>
is the update gate and reset gate activation function and <span class="math">\(sigmoid\)</span>
is usually used for it. <span class="math">\(act_c\)</span> is the activation function for
candidate hidden state and <span class="math">\(tanh\)</span> is usually used for it.</p>
<p>Note that these <span class="math">\(W_{ux}x_{t}, W_{rx}x_{t}, W_{cx}x_{t}\)</span> operations on
the input <span class="math">\(x_{t}\)</span> are NOT included in this operator. Users can choose
to use fully-connect layer before GRU layer.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable</em>) &#8211; The input of dynamic_gru layer, which supports
variable-time length input sequence. The underlying tensor in this
Variable is a matrix with shape <span class="math">\((T \times 3D)\)</span>, where
<span class="math">\(T\)</span> is the total time steps in this mini-batch, <span class="math">\(D\)</span>
is the hidden size.</li>
<li><strong>size</strong> (<em>int</em>) &#8211; The dimension of the gru cell.</li>
<li><strong>param_attr</strong> (<em>ParamAttr|None</em>) &#8211; <p>The parameter attribute for the learnable
hidden-hidden weight matrix. Note:</p>
<ul>
<li>The shape of the weight matrix is <span class="math">\((T \times 3D)\)</span>, where
<span class="math">\(D\)</span> is the hidden size.</li>
<li>All elements in the weight matrix can be divided into two parts.
The first part are weights of the update gate and reset gate with
shape <span class="math">\((D \times 2D)\)</span>, and the second part are weights for
candidate hidden state with shape <span class="math">\((D \times D)\)</span>.</li>
</ul>
</li>
<li><strong>bias_attr</strong> (<em>ParamAttr</em>) &#8211; The parameter attribute for learnable the
hidden-hidden bias.</li>
<li><strong>is_reverse</strong> (<em>bool</em>) &#8211; Whether to compute reversed GRU, default
<code class="xref py py-attr docutils literal"><span class="pre">False</span></code>.</li>
<li><strong>gate_activation</strong> (<em>str</em>) &#8211; The activation for update gate and reset gate.
Choices = [&#8220;sigmoid&#8221;, &#8220;tanh&#8221;, &#8220;relu&#8221;, &#8220;identity&#8221;], default &#8220;sigmoid&#8221;.</li>
<li><strong>activation</strong> (<em>str</em>) &#8211; The activation for candidate hidden state.
Choices = [&#8220;sigmoid&#8221;, &#8220;tanh&#8221;, &#8220;relu&#8221;, &#8220;identity&#8221;], default &#8220;tanh&#8221;.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The hidden state of GRU. The shape is <span class="math">\((T \times D)\)</span>,             and lod is the same with the input.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">hidden_dim</span> <span class="o">=</span> <span class="mi">512</span>
<span class="n">x</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">fc</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">data</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">hidden_dim</span> <span class="o">*</span> <span class="mi">3</span><span class="p">)</span>
<span class="n">hidden</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">dynamic_gru</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">x</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="n">hidden_dim</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="gru-unit">
<h3>gru_unit<a class="headerlink" href="#gru-unit" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">gru_unit</code><span class="sig-paren">(</span><em>input</em>, <em>hidden</em>, <em>size</em>, <em>weight=None</em>, <em>bias=None</em>, <em>activation='tanh'</em>, <em>gate_activation='sigmoid'</em><span class="sig-paren">)</span></dt>
<dd><p>GRU unit layer. The equation of a gru step is:</p>
<blockquote>
<div><div class="math">
\[ \begin{align}\begin{aligned}u_t &amp; = actGate(xu_{t} + W_u h_{t-1} + b_u)\\r_t &amp; = actGate(xr_{t} + W_r h_{t-1} + b_r)\\m_t &amp; = actNode(xm_t + W_c dot(r_t, h_{t-1}) + b_m)\\h_t &amp; = dot((1-u_t), m_t) + dot(u_t, h_{t-1})\end{aligned}\end{align} \]</div>
</div></blockquote>
<p>The inputs of gru unit includes <span class="math">\(z_t\)</span>, <span class="math">\(h_{t-1}\)</span>. In terms
of the equation above, the <span class="math">\(z_t\)</span> is split into 3 parts -
<span class="math">\(xu_t\)</span>, <span class="math">\(xr_t\)</span> and <span class="math">\(xm_t\)</span>. This means that in order to
implement a full GRU unit operator for an input, a fully
connected layer has to be applied, such that <span class="math">\(z_t = W_{fc}x_t\)</span>.</p>
<p>The terms <span class="math">\(u_t\)</span> and <span class="math">\(r_t\)</span> represent the update and reset gates
of the GRU cell. Unlike LSTM, GRU has one lesser gate. However, there is
an intermediate candidate hidden output, which is denoted by <span class="math">\(m_t\)</span>.
This layer has three outputs <span class="math">\(h_t\)</span>, <span class="math">\(dot(r_t, h_{t-1})\)</span>
and concatenation of <span class="math">\(u_t\)</span>, <span class="math">\(r_t\)</span> and <span class="math">\(m_t\)</span>.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable</em>) &#8211; The fc transformed input value of current step.</li>
<li><strong>hidden</strong> (<em>Variable</em>) &#8211; The hidden value of lstm unit from previous step.</li>
<li><strong>size</strong> (<em>integer</em>) &#8211; The input dimension value.</li>
<li><strong>weight</strong> (<em>ParamAttr</em>) &#8211; The weight parameters for gru unit. Default: None</li>
<li><strong>bias</strong> (<em>ParamAttr</em>) &#8211; The bias parameters for gru unit. Default: None</li>
<li><strong>activation</strong> (<em>string</em>) &#8211; The activation type for cell (actNode).
Default: &#8216;tanh&#8217;</li>
<li><strong>gate_activation</strong> (<em>string</em>) &#8211; The activation type for gates (actGate).
Default: &#8216;sigmoid&#8217;</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The hidden value, reset-hidden value and gate values.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">tuple</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="c1"># assuming we have x_t_data and prev_hidden of size=10</span>
<span class="n">x_t</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">fc</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">x_t_data</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="mi">30</span><span class="p">)</span>
<span class="n">hidden_val</span><span class="p">,</span> <span class="n">r_h_val</span><span class="p">,</span> <span class="n">gate_val</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">gru_unit</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">x_t</span><span class="p">,</span>
                                       <span class="n">hidden</span> <span class="o">=</span> <span class="n">prev_hidden</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="linear-chain-crf">
<h3>linear_chain_crf<a class="headerlink" href="#linear-chain-crf" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">linear_chain_crf</code><span class="sig-paren">(</span><em>input</em>, <em>label</em>, <em>param_attr=None</em><span class="sig-paren">)</span></dt>
<dd></dd></dl>

</div>
<div class="section" id="crf-decoding">
<h3>crf_decoding<a class="headerlink" href="#crf-decoding" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">crf_decoding</code><span class="sig-paren">(</span><em>input</em>, <em>param_attr</em>, <em>label=None</em><span class="sig-paren">)</span></dt>
<dd></dd></dl>

</div>
<div class="section" id="cos-sim">
<h3>cos_sim<a class="headerlink" href="#cos-sim" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">cos_sim</code><span class="sig-paren">(</span><em>X</em>, <em>Y</em><span class="sig-paren">)</span></dt>
<dd><p>This function performs the cosine similarity between two tensors
X and Y and returns that as the output.</p>
</dd></dl>

</div>
<div class="section" id="cross-entropy">
<h3>cross_entropy<a class="headerlink" href="#cross-entropy" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">cross_entropy</code><span class="sig-paren">(</span><em>input</em>, <em>label</em>, <em>soft_label=False</em><span class="sig-paren">)</span></dt>
<dd><p><strong>Cross Entropy Layer</strong></p>
<p>This layer computes the cross entropy between <cite>input</cite> and <cite>label</cite>. It
supports both standard cross-entropy and soft-label cross-entropy loss
computation.</p>
<ol class="arabic">
<li><dl class="first docutils">
<dt>One-hot cross-entropy:</dt>
<dd><p class="first"><cite>soft_label = False</cite>, <cite>Label[i, 0]</cite> indicates the class index for sample i:</p>
<div class="last math">
\[Y[i] = -\log(X[i, Label[i]])\]</div>
</dd>
</dl>
</li>
<li><dl class="first docutils">
<dt>Soft-label cross-entropy:</dt>
<dd><p class="first"><cite>soft_label = True</cite>, <cite>Label[i, j]</cite> indicates the soft label of class j
for sample i:</p>
<div class="last math">
\[Y[i] = \sum_j{-Label[i, j] * log(X[i, j])}\]</div>
</dd>
</dl>
<p>Please make sure that in this case the summation of each row of <cite>label</cite>
equals one.</p>
</li>
<li><dl class="first docutils">
<dt>One-hot cross-entropy with vecterized <cite>label</cite>:</dt>
<dd><p class="first last">As a special case of 2), when each row of &#8216;label&#8217; has only one
non-zero element which is equal to 1, soft-label cross-entropy degenerates
to a one-hot cross-entropy with one-hot label representation.</p>
</dd>
</dl>
</li>
</ol>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable|list</em>) &#8211; a 2-D tensor with shape [N x D], where N is the
batch size and D is the number of classes. This
input is a probability computed by the previous
operator, which is almost always the result of
a softmax operator.</li>
<li><strong>label</strong> (<em>Variable|list</em>) &#8211; the ground truth which is a 2-D tensor. When
<cite>soft_label</cite> is set to <cite>False</cite>, <cite>label</cite> is a
tensor&lt;int64&gt; with shape [N x 1]. When
<cite>soft_label</cite> is set to <cite>True</cite>, <cite>label</cite> is a
tensor&lt;float/double&gt; with shape [N x D].</li>
<li><strong>soft_label</strong> (<em>bool</em>) &#8211; a flag indicating whether to
interpretate the given labels as soft
labels, default <cite>False</cite>.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">A 2-D tensor with shape [N x 1], the cross entropy loss.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Raises:</th><td class="field-body"><p class="first"><cite>ValueError</cite> &#8211; 1) the 1st dimension of <cite>input</cite> and <cite>label</cite> are not equal.
2) when <cite>soft_label == True</cite>, and the 2nd dimension of</p>
<blockquote>
<div><p><cite>input</cite> and <cite>label</cite> are not equal.</p>
</div></blockquote>
<ol class="last arabic simple" start="3">
<li>when <cite>soft_label == False</cite>, and the 2nd dimension of
<cite>label</cite> is not 1.</li>
</ol>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">predict</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">fc</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">net</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">classdim</span><span class="p">,</span> <span class="n">act</span><span class="o">=</span><span class="s1">&#39;softmax&#39;</span><span class="p">)</span>
<span class="n">cost</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">cross_entropy</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">predict</span><span class="p">,</span> <span class="n">label</span><span class="o">=</span><span class="n">label</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="square-error-cost">
<h3>square_error_cost<a class="headerlink" href="#square-error-cost" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">square_error_cost</code><span class="sig-paren">(</span><em>input</em>, <em>label</em><span class="sig-paren">)</span></dt>
<dd><p><strong>Square error cost layer</strong></p>
<p>This layer accepts input predictions and target label and returns the
squared error cost.</p>
<p>For predictions, <span class="math">\(X\)</span>, and target labels, <span class="math">\(Y\)</span>, the equation is:</p>
<div class="math">
\[Out = (X - Y)^2\]</div>
<p>In the above equation:</p>
<blockquote>
<div><ul class="simple">
<li><span class="math">\(X\)</span>: Input predictions, a tensor.</li>
<li><span class="math">\(Y\)</span>: Input labels, a tensor.</li>
<li><span class="math">\(Out\)</span>: Output value, same shape with <span class="math">\(X\)</span>.</li>
</ul>
</div></blockquote>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable</em>) &#8211; Input tensor, has predictions.</li>
<li><strong>label</strong> (<em>Variable</em>) &#8211; Label tensor, has target labels.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The tensor variable storing the element-wise squared error                   difference of input and label.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">y</span> <span class="o">=</span> <span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;y&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">)</span>
<span class="n">y_predict</span> <span class="o">=</span> <span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;y_predict&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">)</span>
<span class="n">cost</span> <span class="o">=</span> <span class="n">layers</span><span class="o">.</span><span class="n">square_error_cost</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">y_predict</span><span class="p">,</span> <span class="n">label</span><span class="o">=</span><span class="n">y</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="accuracy">
<h3>accuracy<a class="headerlink" href="#accuracy" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">accuracy</code><span class="sig-paren">(</span><em>input</em>, <em>label</em>, <em>k=1</em>, <em>correct=None</em>, <em>total=None</em><span class="sig-paren">)</span></dt>
<dd><p>This function computes the accuracy using the input and label.
The output is the top_k inputs and their indices.</p>
</dd></dl>

</div>
<div class="section" id="chunk-eval">
<h3>chunk_eval<a class="headerlink" href="#chunk-eval" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">chunk_eval</code><span class="sig-paren">(</span><em>input</em>, <em>label</em>, <em>chunk_scheme</em>, <em>num_chunk_types</em>, <em>excluded_chunk_types=None</em><span class="sig-paren">)</span></dt>
<dd><p>This function computes and outputs the precision, recall and
F1-score of chunk detection.</p>
</dd></dl>

</div>
<div class="section" id="sequence-conv">
<h3>sequence_conv<a class="headerlink" href="#sequence-conv" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">sequence_conv</code><span class="sig-paren">(</span><em>input</em>, <em>num_filters</em>, <em>filter_size=3</em>, <em>filter_stride=1</em>, <em>padding=None</em>, <em>bias_attr=None</em>, <em>param_attr=None</em>, <em>act=None</em><span class="sig-paren">)</span></dt>
<dd><p>This function creates the op for sequence_conv, using the inputs and
other convolutional configurations for the filters and stride as given
in the input parameters to the function.</p>
</dd></dl>

</div>
<div class="section" id="conv2d">
<h3>conv2d<a class="headerlink" href="#conv2d" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">conv2d</code><span class="sig-paren">(</span><em>input</em>, <em>num_filters</em>, <em>filter_size</em>, <em>stride=None</em>, <em>padding=None</em>, <em>groups=None</em>, <em>param_attr=None</em>, <em>bias_attr=None</em>, <em>use_cudnn=True</em>, <em>act=None</em><span class="sig-paren">)</span></dt>
<dd><p><strong>Convlution2D Layer</strong></p>
<p>The convolution2D layer calculates the output based on the input, filter
and strides, paddings, dilations, groups parameters. Input(Input) and
Output(Output) are in NCHW format. Where N is batch size, C is the number of
channels, H is the height of the feature, and W is the width of the feature.
The details of convolution layer, please refer UFLDL&#8217;s <a class="reference external" href="http://ufldl.stanford.edu/tutorial/supervised/FeatureExtractionUsingConvolution/">convolution,</a> .
If bias attribution and activation type are provided, bias is added to the
output of the convolution, and the corresponding activation function is
applied to the final result.</p>
<p>For each input <span class="math">\(X\)</span>, the equation is:</p>
<div class="math">
\[Out = \sigma (W \ast X + b)\]</div>
<p>In the above equation:</p>
<ul class="simple">
<li><span class="math">\(X\)</span>: Input value, a tensor with NCHW format.</li>
<li><span class="math">\(W\)</span>: Filter value, a tensor with MCHW format.</li>
<li><span class="math">\(\ast\)</span>: Convolution operation.</li>
<li><span class="math">\(b\)</span>: Bias value, a 2-D tensor with shape [M, 1].</li>
<li><span class="math">\(\sigma\)</span>: Activation function.</li>
<li><dl class="first docutils">
<dt><span class="math">\(Out\)</span>: Output value, the shape of <span class="math">\(Out\)</span> and <span class="math">\(X\)</span> may be</dt>
<dd>different.</dd>
</dl>
</li>
</ul>
<p class="rubric">Example</p>
<ul>
<li><p class="first">Input:</p>
<p>Input shape: $(N, C_{in}, H_{in}, W_{in})$</p>
<p>Filter shape: $(C_{out}, C_{in}, H_f, W_f)$</p>
</li>
<li><p class="first">Output:
Output shape: $(N, C_{out}, H_{out}, W_{out})$</p>
</li>
</ul>
<p>Where</p>
<div class="math">
\[\]</div>
<p>H_{out}&amp;= frac{(H_{in} + 2 * paddings[0] - (dilations[0] * (H_f - 1) + 1))}{strides[0]} + 1 \
W_{out}&amp;= frac{(W_{in} + 2 * paddings[1] - (dilations[1] * (W_f - 1) + 1))}{strides[1]} + 1</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable</em>) &#8211; The input image with [N, C, H, W] format.</li>
<li><strong>num_filters</strong> (<em>int</em>) &#8211; The number of filter. It is as same as the output
image channel.</li>
<li><strong>filter_size</strong> (<em>int|tuple|None</em>) &#8211; The filter size. If filter_size is a tuple,
it must contain two integers, (filter_size_H, filter_size_W).
Otherwise, the filter will be a square.</li>
<li><strong>stride</strong> (<em>int|tuple</em>) &#8211; The stride size. If stride is a tuple, it must
contain two integers, (stride_H, stride_W). Otherwise, the
stride_H = stride_W = stride. Default: stride = 1.</li>
<li><strong>padding</strong> (<em>int|tuple</em>) &#8211; The padding size. If padding is a tuple, it must
contain two integers, (padding_H, padding_W). Otherwise, the
padding_H = padding_W = padding. Default: padding = 0.</li>
<li><strong>groups</strong> (<em>int</em>) &#8211; The groups number of the Conv2d Layer. According to grouped
convolution in Alex Krizhevsky&#8217;s Deep CNN paper: when group=2,
the first half of the filters is only connected to the first half
of the input channels, while the second half of the filters is only
connected to the second half of the input channels. Default: groups=1</li>
<li><strong>param_attr</strong> (<em>ParamAttr</em>) &#8211; The parameters to the Conv2d Layer. Default: None</li>
<li><strong>bias_attr</strong> (<em>ParamAttr</em>) &#8211; Bias parameter for the Conv2d layer. Default: None</li>
<li><strong>use_cudnn</strong> (<em>bool</em>) &#8211; Use cudnn kernel or not, it is valid only when the cudnn
library is installed. Default: True</li>
<li><strong>act</strong> (<em>str</em>) &#8211; Activation type. Default: None</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The tensor variable storing the convolution and                   non-linearity activation result.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first">Variable</p>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Raises:</th><td class="field-body"><p class="first last"><code class="xref py py-exc docutils literal"><span class="pre">ValueError</span></code> &#8211; If the shapes of input, filter_size, stride, padding and
groups mismatch.</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">data</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span>
    <span class="n">name</span><span class="o">=</span><span class="s1">&#39;data&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">32</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">)</span>
<span class="n">conv2d</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">conv2d</span><span class="p">(</span>
    <span class="nb">input</span><span class="o">=</span><span class="n">data</span><span class="p">,</span> <span class="n">num_filters</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">filter_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">act</span><span class="o">=</span><span class="s2">&quot;relu&quot;</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="sequence-pool">
<h3>sequence_pool<a class="headerlink" href="#sequence-pool" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">sequence_pool</code><span class="sig-paren">(</span><em>input</em>, <em>pool_type</em><span class="sig-paren">)</span></dt>
<dd><p>This function add the operator for sequence pooling.
It pools features of all time-steps of each instance, and is applied
on top of the input using pool_type mentioned in the parameters.</p>
<p>It supports four pool_type:</p>
<ul class="simple">
<li>average: <span class="math">\(Out[i] = \frac{\sum_i X_i}{N}\)</span></li>
<li>sum:     <span class="math">\(Out[i] = \sum_jX_{ij}\)</span></li>
<li>sqrt:    <span class="math">\(Out[i] = \frac{\sum_jX_{ij}}{\sqrt{len(X_i)}}\)</span></li>
<li>max:     <span class="math">\(Out[i] = max(X_i)\)</span></li>
</ul>
<div class="highlight-text"><div class="highlight"><pre><span></span>x is a 1-level LoDTensor:
  x.lod = [[0, 2, 5, 7]]
  x.data = [1, 3, 2, 4, 6, 5, 1]
  x.dims = [7, 1]

then output is a Tensor:
  out.dim = [3, 1]
  with condition len(x.lod[-1]) - 1 == out.dims[0]

for different pool_type:
  average: out.data = [2, 4, 3], where 2=(1+3)/2, 4=(2+4+6)/3, 3=(5+1)/2
  sum    : out.data = [4, 12, 6], where 4=1+3, 12=2+4+6, 6=5+1
  sqrt   : out.data = [2.82, 6.93, 4.24], where 2.82=(1+3)/sqrt(2),
             6.93=(2+4+6)/sqrt(3), 4.24=(5+1)/sqrt(2)
  max    : out.data = [3, 6, 5], where 3=max(1,3), 6=max(2,4,6), 5=max(5,1)
</pre></div>
</div>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>variable</em>) &#8211; The input variable which is a LoDTensor.</li>
<li><strong>pool_type</strong> (<em>string</em>) &#8211; The pooling type of sequence_pool.
It supports average, sum, sqrt and max.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">The sequence pooling variable which is a Tensor.</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">x</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;x&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">7</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
                 <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">,</span> <span class="n">lod_level</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
<span class="n">avg_x</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">sequence_pool</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">x</span><span class="p">,</span> <span class="n">pool_type</span><span class="o">=</span><span class="s1">&#39;average&#39;</span><span class="p">)</span>
<span class="n">sum_x</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">sequence_pool</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">x</span><span class="p">,</span> <span class="n">pool_type</span><span class="o">=</span><span class="s1">&#39;sum&#39;</span><span class="p">)</span>
<span class="n">sqrt_x</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">sequence_pool</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">x</span><span class="p">,</span> <span class="n">pool_type</span><span class="o">=</span><span class="s1">&#39;sqrt&#39;</span><span class="p">)</span>
<span class="n">max_x</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">sequence_pool</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">x</span><span class="p">,</span> <span class="n">pool_type</span><span class="o">=</span><span class="s1">&#39;max&#39;</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="pool2d">
<h3>pool2d<a class="headerlink" href="#pool2d" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">pool2d</code><span class="sig-paren">(</span><em>input</em>, <em>pool_size</em>, <em>pool_type</em>, <em>pool_stride=None</em>, <em>pool_padding=None</em>, <em>global_pooling=False</em>, <em>use_cudnn=True</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd><p>This function adds the operator for pooling in 2 dimensions, using the
pooling configurations mentioned in input parameters.</p>
</dd></dl>

</div>
<div class="section" id="batch-norm">
<h3>batch_norm<a class="headerlink" href="#batch-norm" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">batch_norm</code><span class="sig-paren">(</span><em>input</em>, <em>act=None</em>, <em>is_test=False</em>, <em>momentum=0.9</em>, <em>epsilon=1e-05</em>, <em>param_attr=None</em>, <em>bias_attr=None</em>, <em>data_layout='NCHW'</em>, <em>name=None</em>, <em>moving_mean_name=None</em>, <em>moving_variance_name=None</em><span class="sig-paren">)</span></dt>
<dd><p>This function helps create an operator to implement
the BatchNorm layer using the configurations from the input parameters.</p>
</dd></dl>

</div>
<div class="section" id="layer-norm">
<h3>layer_norm<a class="headerlink" href="#layer-norm" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">layer_norm</code><span class="sig-paren">(</span><em>input</em>, <em>scale=True</em>, <em>shift=True</em>, <em>begin_norm_axis=1</em>, <em>epsilon=1e-05</em>, <em>param_attr=None</em>, <em>bias_attr=None</em>, <em>act=None</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd><p><strong>Layer Normalization</strong></p>
<p>Assume feature vectors exist on dimensions
<code class="xref py py-attr docutils literal"><span class="pre">begin_norm_axis</span> <span class="pre">...</span> <span class="pre">rank(input)</span></code> and calculate the moment statistics
along these dimensions for each feature vector <span class="math">\(a\)</span> with size
<span class="math">\(H\)</span>, then normalize each feature vector using the corresponding
statistics. After that, apply learnable gain and bias on the normalized
tensor to scale and shift if <code class="xref py py-attr docutils literal"><span class="pre">scale</span></code> and <code class="xref py py-attr docutils literal"><span class="pre">shift</span></code> are set.</p>
<p>Refer to <a class="reference external" href="https://arxiv.org/pdf/1607.06450v1.pdf">Layer Normalization</a></p>
<p>The formula is as follows:</p>
<div class="math">
\[ \begin{align}\begin{aligned}\mu &amp; = \frac{1}{H}\sum_{i=1}^{H} a_i\\\sigma &amp; = \sqrt{\frac{1}{H}\sum_{i=1}^{H}(a_i - \mu)^2}\\h &amp; = f(\frac{g}{\sigma}(a - \mu) + b)\end{aligned}\end{align} \]</div>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable</em>) &#8211; The input tensor variable.</li>
<li><strong>scale</strong> (<em>bool</em>) &#8211; Whether to learn the adaptive gain <span class="math">\(g\)</span> after
normalization.</li>
<li><strong>shift</strong> (<em>bool</em>) &#8211; Whether to learn the adaptive bias <span class="math">\(b\)</span> after
normalization.</li>
<li><strong>begin_norm_axis</strong> (<em>bool</em>) &#8211; The normalization will be performed along
dimensions from <code class="xref py py-attr docutils literal"><span class="pre">begin_norm_axis</span></code> to <code class="xref py py-attr docutils literal"><span class="pre">rank(input)</span></code>.</li>
<li><strong>epsilon</strong> (<em>float</em>) &#8211; The small value added to the variance to prevent
division by zero.</li>
<li><strong>param_attr</strong> (<em>ParamAttr|None</em>) &#8211; The parameter attribute for the learnable
gain <span class="math">\(g\)</span>.</li>
<li><strong>bias_attr</strong> (<em>ParamAttr|None</em>) &#8211; The parameter attribute for the learnable
bias <span class="math">\(b\)</span>.</li>
<li><strong>act</strong> (<em>str</em>) &#8211; Activation to be applied to the output of layer normalizaiton.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">A tensor variable with the same shape as the input.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">data</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span>
  <span class="n">name</span><span class="o">=</span><span class="s1">&#39;data&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">32</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">)</span>
<span class="n">x</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">layer_norm</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">data</span><span class="p">,</span> <span class="n">begin_norm_axis</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="beam-search-decode">
<h3>beam_search_decode<a class="headerlink" href="#beam-search-decode" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">beam_search_decode</code><span class="sig-paren">(</span><em>ids</em>, <em>scores</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd></dd></dl>

</div>
<div class="section" id="conv2d-transpose">
<h3>conv2d_transpose<a class="headerlink" href="#conv2d-transpose" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">conv2d_transpose</code><span class="sig-paren">(</span><em>input</em>, <em>num_filters</em>, <em>output_size=None</em>, <em>filter_size=None</em>, <em>padding=None</em>, <em>stride=None</em>, <em>dilation=None</em>, <em>param_attr=None</em>, <em>use_cudnn=True</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd><p><strong>Convlution2D transpose layer</strong></p>
<p>The convolution2D transpose layer calculates the output based on the input,
filter, and dilations, strides, paddings. Input(Input) and output(Output)
are in NCHW format. Where N is batch size, C is the number of channels,
H is the height of the feature, and W is the width of the feature.
Parameters(dilations, strides, paddings) are two elements. These two elements
represent height and width, respectively. The details of convolution transpose
layer, please refer to the following explanation and references
<a class="reference external" href="http://www.matthewzeiler.com/wp-content/uploads/2017/07/cvpr2010.pdf">therein</a>.</p>
<p>For each input <span class="math">\(X\)</span>, the equation is:</p>
<div class="math">
\[Out = W \ast X\]</div>
<p>In the above equation:</p>
<ul class="simple">
<li><span class="math">\(X\)</span>: Input value, a tensor with NCHW format.</li>
<li><span class="math">\(W\)</span>: Filter value, a tensor with MCHW format.</li>
<li><span class="math">\(\ast\)</span> : Convolution transpose operation.</li>
<li><dl class="first docutils">
<dt><span class="math">\(Out\)</span>: Output value, the shape of <span class="math">\(Out\)</span> and <span class="math">\(X\)</span> may be</dt>
<dd>different.</dd>
</dl>
</li>
</ul>
<p class="rubric">Example</p>
<ul>
<li><p class="first">Input:</p>
<p>Input shape: $(N, C_{in}, H_{in}, W_{in})$</p>
<p>Filter shape: $(C_{in}, C_{out}, H_f, W_f)$</p>
</li>
<li><p class="first">Output:</p>
<p>Output shape: $(N, C_{out}, H_{out}, W_{out})$</p>
</li>
</ul>
<p>Where</p>
<div class="math">
\[\begin{split}H_{out} &amp;= (H_{in} - 1) * strides[0] - 2 * paddings[0] + dilations[0] * (H_f - 1) + 1 \\
W_{out} &amp;= (W_{in} - 1) * strides[1] - 2 * paddings[1] + dilations[1] * (W_f - 1) + 1\end{split}\]</div>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable</em>) &#8211; The input image with [N, C, H, W] format.</li>
<li><strong>num_filters</strong> (<em>int</em>) &#8211; The number of the filter. It is as same as the output
image channel.</li>
<li><strong>output_size</strong> (<em>int|tuple|None</em>) &#8211; The output image size. If output size is a
tuple, it must contain two integers, (image_H, image_W). This
parameter only works when filter_size is None.</li>
<li><strong>filter_size</strong> (<em>int|tuple|None</em>) &#8211; The filter size. If filter_size is a tuple,
it must contain two integers, (filter_size_H, filter_size_W).
Otherwise, the filter will be a square. None if use output size to
calculate filter_size.</li>
<li><strong>padding</strong> (<em>int|tuple</em>) &#8211; The padding size. If padding is a tuple, it must
contain two integers, (padding_H, padding_W). Otherwise, the
padding_H = padding_W = padding. Default: padding = 0.</li>
<li><strong>stride</strong> (<em>int|tuple</em>) &#8211; The stride size. If stride is a tuple, it must
contain two integers, (stride_H, stride_W). Otherwise, the
stride_H = stride_W = stride. Default: stride = 1.</li>
<li><strong>dilation</strong> (<em>int|tuple</em>) &#8211; The dilation size. If dilation is a tuple, it must
contain two integers, (dilation_H, dilation_W). Otherwise, the
dilation_H = dilation_W = dilation. Default: dilation = 1.</li>
<li><strong>param_attr</strong> (<em>ParamAttr</em>) &#8211; The parameters to the Conv2d_transpose Layer.
Default: None</li>
<li><strong>use_cudnn</strong> (<em>bool</em>) &#8211; Use cudnn kernel or not, it is valid only when the cudnn
library is installed. Default: True</li>
<li><strong>name</strong> (<em>str|None</em>) &#8211; A name for this layer(optional). If set None, the layer
will be named automatically.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The tensor variable storing the convolution transpose result.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first">Variable</p>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Raises:</th><td class="field-body"><p class="first last"><code class="xref py py-exc docutils literal"><span class="pre">ValueError</span></code> &#8211; If the shapes of input, filter_size, stride, padding and
groups mismatch.</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">data</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span>
    <span class="n">name</span><span class="o">=</span><span class="s1">&#39;data&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">32</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">)</span>
<span class="n">conv2d_transpose</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">conv2d_transpose</span><span class="p">(</span>
    <span class="nb">input</span><span class="o">=</span><span class="n">data</span><span class="p">,</span> <span class="n">num_filters</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">filter_size</span><span class="o">=</span><span class="mi">3</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="sequence-expand">
<h3>sequence_expand<a class="headerlink" href="#sequence-expand" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">sequence_expand</code><span class="sig-paren">(</span><em>x</em>, <em>y</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd><p>Sequence Expand Layer. This layer will expand the input variable <strong>x</strong>
according to LoD information of <strong>y</strong>. And the following examples will
explain how sequence_expand works:</p>
<div class="highlight-text"><div class="highlight"><pre><span></span>* Case 1
    x is a LoDTensor:
        x.lod = [[0,       2, 3],
                 [0, 1,    3, 4]]
        x.data = [a, b, c, d]
        x.dims = [4, 1]

    y is a LoDTensor:
        y.lod = [[0,    2,    4],
                 [0, 3, 6, 7, 8]]

    with condition len(y.lod[-1]) - 1 == x.dims[0]

    then output is a 2-level LoDTensor:
        out.lod = [[0,                2,    4],
                   [0,       3,       6, 7, 8]]
        out.data = [a, a, a, b, b, b, c, d]
        out.dims = [8, 1]

* Case 2
    x is a Tensor:
        x.data = [a, b, c]
        x.dims = [3, 1]

    y is a LoDTensor:
        y.lod = [[0, 2, 3, 6]]

    with condition len(y.lod[-1]) - 1 == x.dims[0]

    then output is a 1-level LoDTensor:
        out.lod = [[0,    2, 3,      6]]
        out.data = [a, a, b, c, c, c]
        out.dims = [6, 1]
</pre></div>
</div>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> (<em>Variable</em>) &#8211; The input variable which is a Tensor or LoDTensor.</li>
<li><strong>y</strong> (<em>Variable</em>) &#8211; The input variable which is a LoDTensor.</li>
<li><strong>name</strong> (<em>str|None</em>) &#8211; A name for this layer(optional). If set None, the layer
will be named automatically.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The expanded variable which is a LoDTensor.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">x</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;x&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">10</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">)</span>
<span class="n">y</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;y&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">10</span><span class="p">,</span> <span class="mi">20</span><span class="p">],</span>
                 <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">,</span> <span class="n">lod_level</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
<span class="n">out</span> <span class="o">=</span> <span class="n">layers</span><span class="o">.</span><span class="n">sequence_expand</span><span class="p">(</span><span class="n">x</span><span class="o">=</span><span class="n">x</span><span class="p">,</span> <span class="n">y</span><span class="o">=</span><span class="n">y</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="lstm-unit">
<h3>lstm_unit<a class="headerlink" href="#lstm-unit" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">lstm_unit</code><span class="sig-paren">(</span><em>x_t</em>, <em>hidden_t_prev</em>, <em>cell_t_prev</em>, <em>forget_bias=0.0</em>, <em>param_attr=None</em>, <em>bias_attr=None</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd><p>Lstm unit layer. The equation of a lstm step is:</p>
<blockquote>
<div><div class="math">
\[ \begin{align}\begin{aligned}i_t &amp; = \sigma(W_{x_i}x_{t} + W_{h_i}h_{t-1} + b_i)\\f_t &amp; = \sigma(W_{x_f}x_{t} + W_{h_f}h_{t-1} + b_f)\\c_t &amp; = f_tc_{t-1} + i_t tanh (W_{x_c}x_t + W_{h_c}h_{t-1} + b_c)\\o_t &amp; = \sigma(W_{x_o}x_{t} + W_{h_o}h_{t-1} + b_o)\\h_t &amp; = o_t tanh(c_t)\end{aligned}\end{align} \]</div>
</div></blockquote>
<p>The inputs of lstm unit include <span class="math">\(x_t\)</span>, <span class="math">\(h_{t-1}\)</span> and
<span class="math">\(c_{t-1}\)</span>. The 2nd dimensions of <span class="math">\(h_{t-1}\)</span> and <span class="math">\(c_{t-1}\)</span>
should be same. The implementation separates the linear transformation and
non-linear transformation apart. Here, we take <span class="math">\(i_t\)</span> as an example.
The linear transformation is applied by calling a <cite>fc</cite> layer and the
equation is:</p>
<blockquote>
<div><div class="math">
\[L_{i_t} = W_{x_i}x_{t} + W_{h_i}h_{t-1} + b_i\]</div>
</div></blockquote>
<p>The non-linear transformation is applied by calling <cite>lstm_unit_op</cite> and the
equation is:</p>
<blockquote>
<div><div class="math">
\[i_t = \sigma(L_{i_t})\]</div>
</div></blockquote>
<p>This layer has two outputs including <span class="math">\(h_t\)</span> and <span class="math">\(o_t\)</span>.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x_t</strong> (<em>Variable</em>) &#8211; The input value of current step, a 2-D tensor with shape
M x N, M for batch size and N for input size.</li>
<li><strong>hidden_t_prev</strong> (<em>Variable</em>) &#8211; The hidden value of lstm unit, a 2-D tensor
with shape M x S, M for batch size and S for size of lstm unit.</li>
<li><strong>cell_t_prev</strong> (<em>Variable</em>) &#8211; The cell value of lstm unit, a 2-D tensor with
shape M x S, M for batch size and S for size of lstm unit.</li>
<li><strong>forget_bias</strong> (<em>float</em>) &#8211; The forget bias of lstm unit.</li>
<li><strong>param_attr</strong> (<em>ParamAttr</em>) &#8211; The attributes of parameter weights, used to set
initializer, name etc.</li>
<li><strong>bias_attr</strong> (<em>ParamAttr</em>) &#8211; The attributes of bias weights, if not False,
bias weights will be created and be set to default value.</li>
<li><strong>name</strong> (<em>str|None</em>) &#8211; A name for this layer(optional). If set None, the layer
will be named automatically.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The hidden value and cell value of lstm unit.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first">tuple</p>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Raises:</th><td class="field-body"><p class="first last"><code class="xref py py-exc docutils literal"><span class="pre">ValueError</span></code> &#8211; The ranks of <strong>x_t</strong>, <strong>hidden_t_prev</strong> and <strong>cell_t_prev</strong>
not be 2 or the 1st dimensions of <strong>x_t</strong>, <strong>hidden_t_prev</strong>
and <strong>cell_t_prev</strong> not be the same or the 2nd dimensions of
<strong>hidden_t_prev</strong> and <strong>cell_t_prev</strong> not be the same.</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">x_t</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">fc</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">x_t_data</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="mi">10</span><span class="p">)</span>
<span class="n">prev_hidden</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">fc</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">prev_hidden_data</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="mi">30</span><span class="p">)</span>
<span class="n">prev_cell</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">fc</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">prev_cell_data</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="mi">30</span><span class="p">)</span>
<span class="n">hidden_value</span><span class="p">,</span> <span class="n">cell_value</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">lstm_unit</span><span class="p">(</span><span class="n">x_t</span><span class="o">=</span><span class="n">x_t</span><span class="p">,</span>
                                       <span class="n">hidden_t_prev</span><span class="o">=</span><span class="n">prev_hidden</span><span class="p">,</span>
                                       <span class="n">cell_t_prev</span><span class="o">=</span><span class="n">prev_cell</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="reduce-sum">
<h3>reduce_sum<a class="headerlink" href="#reduce-sum" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">reduce_sum</code><span class="sig-paren">(</span><em>input</em>, <em>dim=None</em>, <em>keep_dim=False</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd><p>Computes the sum of tensor elements over the given dimension.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable</em>) &#8211; The input variable which is a Tensor or LoDTensor.</li>
<li><strong>dim</strong> (<em>int|None</em>) &#8211; The dimension along which the sum is performed. If
<code class="xref py py-attr docutils literal"><span class="pre">None</span></code>, sum all elements of <code class="xref py py-attr docutils literal"><span class="pre">input</span></code> and return a
Tensor variable with a single element, otherwise must be in the
range <span class="math">\([-rank(input), rank(input))\)</span>. If <span class="math">\(dim &lt; 0\)</span>,
the dimension to reduce is <span class="math">\(rank + dim\)</span>.</li>
<li><strong>keep_dim</strong> (<em>bool|False</em>) &#8211; Whether to reserve the reduced dimension in the
output Tensor. The result tensor will have one fewer dimension
than the <code class="xref py py-attr docutils literal"><span class="pre">input</span></code> unless <code class="xref py py-attr docutils literal"><span class="pre">keep_dim</span></code> is true.</li>
<li><strong>name</strong> (<em>str|None</em>) &#8211; A name for this layer(optional). If set None, the layer
will be named automatically.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The reduced Tensor variable.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="c1"># x is a Tensor variable with following elements:</span>
<span class="c1">#    [[0.2, 0.3, 0.5, 0.9]</span>
<span class="c1">#     [0.1, 0.2, 0.6, 0.7]]</span>
<span class="c1"># Each example is followed by the correspending output tensor.</span>
<span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">reduce_sum</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>  <span class="c1"># [3.5]</span>
<span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">reduce_sum</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>  <span class="c1"># [0.3, 0.5, 1.1, 1.6]</span>
<span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">reduce_sum</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>  <span class="c1"># [1.9, 1.6]</span>
<span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">reduce_sum</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">keep_dim</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span>  <span class="c1"># [[1.9], [1.6]]</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="reduce-mean">
<h3>reduce_mean<a class="headerlink" href="#reduce-mean" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">reduce_mean</code><span class="sig-paren">(</span><em>input</em>, <em>dim=None</em>, <em>keep_dim=False</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd><p>Computes the mean of tensor elements over the given dimension.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable</em>) &#8211; The input variable which is a Tensor or LoDTensor.</li>
<li><strong>dim</strong> (<em>int|None</em>) &#8211; The dimension along which the mean is computed. If
<code class="xref py py-attr docutils literal"><span class="pre">None</span></code>, compute the mean over all elements of <code class="xref py py-attr docutils literal"><span class="pre">input</span></code>
and return a Tensor variable with a single element, otherwise
must be in the range <span class="math">\([-rank(input), rank(input))\)</span>. If
<span class="math">\(dim &lt; 0\)</span>, the dimension to reduce is <span class="math">\(rank + dim\)</span>.</li>
<li><strong>keep_dim</strong> (<em>bool</em>) &#8211; Whether to reserve the reduced dimension in the
output Tensor. The result tensor will have one fewer dimension
than the <code class="xref py py-attr docutils literal"><span class="pre">input</span></code> unless <code class="xref py py-attr docutils literal"><span class="pre">keep_dim</span></code> is true.</li>
<li><strong>name</strong> (<em>str|None</em>) &#8211; A name for this layer(optional). If set None, the layer
will be named automatically.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The reduced Tensor variable.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="c1"># x is a Tensor variable with following elements:</span>
<span class="c1">#    [[0.2, 0.3, 0.5, 0.9]</span>
<span class="c1">#     [0.1, 0.2, 0.6, 0.7]]</span>
<span class="c1"># Each example is followed by the correspending output tensor.</span>
<span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">reduce_mean</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>  <span class="c1"># [0.4375]</span>
<span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">reduce_mean</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>  <span class="c1"># [0.15, 0.25, 0.55, 0.8]</span>
<span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">reduce_mean</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>  <span class="c1"># [0.475, 0.4]</span>
<span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">reduce_mean</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">keep_dim</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span>  <span class="c1"># [[0.475], [0.4]]</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="reduce-max">
<h3>reduce_max<a class="headerlink" href="#reduce-max" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">reduce_max</code><span class="sig-paren">(</span><em>input</em>, <em>dim=None</em>, <em>keep_dim=False</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd><p>Computes the maximum of tensor elements over the given dimension.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable</em>) &#8211; The input variable which is a Tensor or LoDTensor.</li>
<li><strong>dim</strong> (<em>int|None</em>) &#8211; The dimension along which the maximum is computed.
If <code class="xref py py-attr docutils literal"><span class="pre">None</span></code>, compute the maximum over all elements of
<code class="xref py py-attr docutils literal"><span class="pre">input</span></code> and return a Tensor variable with a single element,
otherwise must be in the range <span class="math">\([-rank(input), rank(input))\)</span>.
If <span class="math">\(dim &lt; 0\)</span>, the dimension to reduce is <span class="math">\(rank + dim\)</span>.</li>
<li><strong>keep_dim</strong> (<em>bool</em>) &#8211; Whether to reserve the reduced dimension in the
output Tensor. The result tensor will have one fewer dimension
than the <code class="xref py py-attr docutils literal"><span class="pre">input</span></code> unless <code class="xref py py-attr docutils literal"><span class="pre">keep_dim</span></code> is true.</li>
<li><strong>name</strong> (<em>str|None</em>) &#8211; A name for this layer(optional). If set None, the layer
will be named automatically.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The reduced Tensor variable.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="c1"># x is a Tensor variable with following elements:</span>
<span class="c1">#    [[0.2, 0.3, 0.5, 0.9]</span>
<span class="c1">#     [0.1, 0.2, 0.6, 0.7]]</span>
<span class="c1"># Each example is followed by the correspending output tensor.</span>
<span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">reduce_max</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>  <span class="c1"># [0.9]</span>
<span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">reduce_max</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>  <span class="c1"># [0.2, 0.3, 0.6, 0.9]</span>
<span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">reduce_max</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>  <span class="c1"># [0.9, 0.7]</span>
<span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">reduce_max</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">keep_dim</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span>  <span class="c1"># [[0.9], [0.7]]</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="reduce-min">
<h3>reduce_min<a class="headerlink" href="#reduce-min" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">reduce_min</code><span class="sig-paren">(</span><em>input</em>, <em>dim=None</em>, <em>keep_dim=False</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd><p>Computes the minimum of tensor elements over the given dimension.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable</em>) &#8211; The input variable which is a Tensor or LoDTensor.</li>
<li><strong>dim</strong> (<em>int|None</em>) &#8211; The dimension along which the minimum is computed.
If <code class="xref py py-attr docutils literal"><span class="pre">None</span></code>, compute the minimum over all elements of
<code class="xref py py-attr docutils literal"><span class="pre">input</span></code> and return a Tensor variable with a single element,
otherwise must be in the range <span class="math">\([-rank(input), rank(input))\)</span>.
If <span class="math">\(dim &lt; 0\)</span>, the dimension to reduce is <span class="math">\(rank + dim\)</span>.</li>
<li><strong>keep_dim</strong> (<em>bool</em>) &#8211; Whether to reserve the reduced dimension in the
output Tensor. The result tensor will have one fewer dimension
than the <code class="xref py py-attr docutils literal"><span class="pre">input</span></code> unless <code class="xref py py-attr docutils literal"><span class="pre">keep_dim</span></code> is true.</li>
<li><strong>name</strong> (<em>str|None</em>) &#8211; A name for this layer(optional). If set None, the layer
will be named automatically.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The reduced Tensor variable.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="c1"># x is a Tensor variable with following elements:</span>
<span class="c1">#    [[0.2, 0.3, 0.5, 0.9]</span>
<span class="c1">#     [0.1, 0.2, 0.6, 0.7]]</span>
<span class="c1"># Each example is followed by the correspending output tensor.</span>
<span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">reduce_min</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>  <span class="c1"># [0.1]</span>
<span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">reduce_min</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>  <span class="c1"># [0.1, 0.2, 0.5, 0.7]</span>
<span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">reduce_min</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>  <span class="c1"># [0.2, 0.1]</span>
<span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">reduce_min</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">keep_dim</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span>  <span class="c1"># [[0.2], [0.1]]</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="sequence-first-step">
<h3>sequence_first_step<a class="headerlink" href="#sequence-first-step" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">sequence_first_step</code><span class="sig-paren">(</span><em>input</em><span class="sig-paren">)</span></dt>
<dd><p>This funciton get the first step of sequence.</p>
<div class="highlight-text"><div class="highlight"><pre><span></span>x is a 1-level LoDTensor:
  x.lod = [[0, 2, 5, 7]]
  x.data = [1, 3, 2, 4, 6, 5, 1]
  x.dims = [7, 1]

then output is a Tensor:
  out.dim = [3, 1]
  with condition len(x.lod[-1]) - 1 == out.dims[0]
  out.data = [1, 2, 5], where 1=first(1,3), 2=first(2,4,6), 5=first(5,1)
</pre></div>
</div>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>input</strong> (<em>variable</em>) &#8211; The input variable which is a LoDTensor.</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">The sequence&#8217;s first step variable which is a Tensor.</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">x</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;x&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">7</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
                 <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">,</span> <span class="n">lod_level</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
<span class="n">x_first_step</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">sequence_first_step</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">x</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="sequence-last-step">
<h3>sequence_last_step<a class="headerlink" href="#sequence-last-step" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">sequence_last_step</code><span class="sig-paren">(</span><em>input</em><span class="sig-paren">)</span></dt>
<dd><p>This funciton get the last step of sequence.</p>
<div class="highlight-text"><div class="highlight"><pre><span></span>x is a 1-level LoDTensor:
  x.lod = [[0, 2, 5, 7]]
  x.data = [1, 3, 2, 4, 6, 5, 1]
  x.dims = [7, 1]

then output is a Tensor:
  out.dim = [3, 1]
  with condition len(x.lod[-1]) - 1 == out.dims[0]
  out.data = [3, 6, 1], where 3=last(1,3), 6=last(2,4,6), 1=last(5,1)
</pre></div>
</div>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>input</strong> (<em>variable</em>) &#8211; The input variable which is a LoDTensor.</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">The sequence&#8217;s last step variable which is a Tensor.</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">x</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;x&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">7</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
                 <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">,</span> <span class="n">lod_level</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
<span class="n">x_last_step</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">sequence_last_step</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">x</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="dropout">
<h3>dropout<a class="headerlink" href="#dropout" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">dropout</code><span class="sig-paren">(</span><em>x</em>, <em>dropout_prob</em>, <em>is_test=False</em>, <em>seed=None</em><span class="sig-paren">)</span></dt>
<dd><p>Computes dropout.</p>
<p>Drop or keep each element of <cite>x</cite> independently. Dropout is a regularization
technique for reducing overfitting by preventing neuron co-adaption during
training. The dropout operator randomly set (according to the given dropout
probability) the outputs of some units to zero, while others are remain
unchanged.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> (<em>variable</em>) &#8211; The input tensor.</li>
<li><strong>dropout_prob</strong> (<em>float</em>) &#8211; Probability of setting units to zero.</li>
<li><strong>is_test</strong> (<em>bool</em>) &#8211; A flag indicating whether it is in test phrase or not.</li>
<li><strong>seed</strong> (<em>int</em>) &#8211; A Python integer used to create random seeds. If this
parameter is set to None, a random seed is used.
NOTE: If an integer seed is given, always the same output
units will be dropped. DO NOT use a fixed seed in training.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">A tensor variable.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">x</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s2">&quot;data&quot;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">32</span><span class="p">,</span> <span class="mi">32</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s2">&quot;float32&quot;</span><span class="p">)</span>
<span class="n">droped</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">dropout</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">x</span><span class="p">,</span> <span class="n">dropout_rate</span><span class="o">=</span><span class="mf">0.5</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="split">
<h3>split<a class="headerlink" href="#split" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">split</code><span class="sig-paren">(</span><em>input</em>, <em>num_or_sections</em>, <em>dim=-1</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd><p>Split the input tensor into multiple sub-tensors.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable</em>) &#8211; The input variable which is a Tensor or LoDTensor.</li>
<li><strong>num_or_sections</strong> (<em>int|list</em>) &#8211; If <code class="xref py py-attr docutils literal"><span class="pre">num_or_sections</span></code> is an integer,
then the integer indicates the number of equal sized sub-tensors
that the tensor will be divided into. If <code class="xref py py-attr docutils literal"><span class="pre">num_or_sections</span></code>
is a list of integers, the length of list indicates the number of
sub-tensors and the integers indicate the sizes of sub-tensors&#8217;
<code class="xref py py-attr docutils literal"><span class="pre">dim</span></code> dimension orderly.</li>
<li><strong>dim</strong> (<em>int</em>) &#8211; The dimension along which to split. If <span class="math">\(dim &lt; 0\)</span>, the
dimension to split along is <span class="math">\(rank(input) + dim\)</span>.</li>
<li><strong>name</strong> (<em>str|None</em>) &#8211; A name for this layer(optional). If set None, the layer
will be named automatically.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The list of segmented tensor variables.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">List</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="c1"># x is a Tensor variable with shape [3, 9, 5]:</span>
<span class="n">x0</span><span class="p">,</span> <span class="n">x1</span><span class="p">,</span> <span class="n">x2</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">num_or_sections</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
<span class="n">x0</span><span class="o">.</span><span class="n">shape</span>  <span class="c1"># [3, 3, 5]</span>
<span class="n">x1</span><span class="o">.</span><span class="n">shape</span>  <span class="c1"># [3, 3, 5]</span>
<span class="n">x2</span><span class="o">.</span><span class="n">shape</span>  <span class="c1"># [3, 3, 5]</span>
<span class="n">x0</span><span class="p">,</span> <span class="n">x1</span><span class="p">,</span> <span class="n">x2</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">num_or_sections</span><span class="o">=</span><span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
<span class="n">x0</span><span class="o">.</span><span class="n">shape</span>  <span class="c1"># [3, 2, 5]</span>
<span class="n">x1</span><span class="o">.</span><span class="n">shape</span>  <span class="c1"># [3, 3, 5]</span>
<span class="n">x2</span><span class="o">.</span><span class="n">shape</span>  <span class="c1"># [3, 4, 5]</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="ctc-greedy-decoder">
<h3>ctc_greedy_decoder<a class="headerlink" href="#ctc-greedy-decoder" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">ctc_greedy_decoder</code><span class="sig-paren">(</span><em>input</em>, <em>blank</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd><p>This op is used to decode sequences by greedy policy by below steps:
1. Get the indexes of max value for each row in input. a.k.a.</p>
<blockquote>
<div>numpy.argmax(input, axis=0).</div></blockquote>
<ol class="arabic simple" start="2">
<li>For each sequence in result of step1, merge repeated tokens between two
blanks and delete all blanks.</li>
</ol>
<p>A simple example as below:</p>
<div class="highlight-text"><div class="highlight"><pre><span></span>Given:

input.data = [[0.6, 0.1, 0.3, 0.1],
              [0.3, 0.2, 0.4, 0.1],
              [0.1, 0.5, 0.1, 0.3],
              [0.5, 0.1, 0.3, 0.1],

              [0.5, 0.1, 0.3, 0.1],
              [0.2, 0.2, 0.2, 0.4],
              [0.2, 0.2, 0.1, 0.5],
              [0.5, 0.1, 0.3, 0.1]]

input.lod = [[0, 4, 8]]

Then:

output.data = [[2],
               [1],
               [3]]

output.lod = [[0, 2, 3]]
</pre></div>
</div>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable</em>) &#8211; (LoDTensor&lt;float&gt;), the probabilities of
variable-length sequences, which is a 2-D Tensor with
LoD information. It&#8217;s shape is [Lp, num_classes + 1],
where Lp is the sum of all input sequences&#8217; length and
num_classes is the true number of classes. (not
including the blank label).</li>
<li><strong>blank</strong> (<em>int</em>) &#8211; the blank label index of Connectionist Temporal
Classification (CTC) loss, which is in thehalf-opened
interval [0, num_classes + 1).</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">CTC greedy decode result. If all the sequences in result were
empty, the result LoDTensor will be [-1] with LoD [[0]] and dims [1, 1].</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">x</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;x&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">8</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">)</span>

<span class="n">cost</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">ctc_greedy_decoder</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">x</span><span class="p">,</span> <span class="n">blank</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="edit-distance">
<h3>edit_distance<a class="headerlink" href="#edit-distance" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">edit_distance</code><span class="sig-paren">(</span><em>input</em>, <em>label</em>, <em>normalized=False</em>, <em>ignored_tokens=None</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd><p>EditDistance operator computes the edit distances between a batch of
hypothesis strings and their references. Edit distance, also called
Levenshtein distance, measures how dissimilar two strings are by counting
the minimum number of operations to transform one string into anthor.
Here the operations include insertion, deletion, and substitution.</p>
<p>For example, given hypothesis string A = &#8220;kitten&#8221; and reference
B = &#8220;sitting&#8221;, the edit distance is 3 for A will be transformed into B
at least after two substitutions and one insertion:</p>
<p>&#8220;kitten&#8221; -&gt; &#8220;sitten&#8221; -&gt; &#8220;sittin&#8221; -&gt; &#8220;sitting&#8221;</p>
<p>Input(Hyps) is a LoDTensor consisting of all the hypothesis strings with
the total number denoted by <cite>batch_size</cite>, and the separation is specified
by the LoD information. And the <cite>batch_size</cite> reference strings are arranged
in order in the same way in the LoDTensor Input(Refs).</p>
<p>Output(Out) contains the <cite>batch_size</cite> results and each stands for the edit
distance for a pair of strings respectively. If Attr(normalized) is true,
the edit distance will be divided by the length of reference string.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable</em>) &#8211; The indices for hypothesis strings.</li>
<li><strong>label</strong> (<em>Variable</em>) &#8211; The indices for reference strings.</li>
<li><strong>normalized</strong> (<em>bool</em>) &#8211; Indicated whether to normalize the edit distance by
the length of reference string.</li>
<li><strong>ignored_tokens</strong> (<em>list of int</em>) &#8211; Tokens that should be removed before
calculating edit distance.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">sequence-to-sequence edit distance in shape [batch_size, 1].</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">x</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;x&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">8</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">)</span>
<span class="n">y</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;y&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">7</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">)</span>

<span class="n">cost</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">edit_distance</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">x</span><span class="p">,</span><span class="n">label</span><span class="o">=</span><span class="n">y</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="l2-normalize">
<h3>l2_normalize<a class="headerlink" href="#l2-normalize" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">l2_normalize</code><span class="sig-paren">(</span><em>x</em>, <em>axis</em>, <em>epsilon=1e-12</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd><p><strong>L2 normalize Layer</strong></p>
<p>The l2 normalize layer normalizes <cite>x</cite> along dimension <cite>axis</cite> using an L2
norm. For a 1-D tensor (<cite>dim</cite> is fixed to 0), this layer computes</p>
<p>output = x / sqrt(max(sum(x**2), epsilon))</p>
<p>For <cite>x</cite> with more dimensions, this layer independently normalizes each 1-D
slice along dimension <cite>axis</cite>.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> (<em>Variable|list</em>) &#8211; The input tensor to l2_normalize layer.</li>
<li><strong>axis</strong> (<em>int</em>) &#8211; Dimension along which to normalize the input.</li>
<li><strong>epsilon</strong> (<em>float</em>) &#8211; A lower bound value for <cite>x</cite>&#8216;s l2 norm. sqrt(epsilon) will
be used as the divisor if the l2 norm of <cite>x</cite> is less than
sqrt(epsilon).</li>
<li><strong>name</strong> (<em>str|None</em>) &#8211; A name for this layer(optional). If set None, the layer
will be named automatically.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The output tensor variable.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">data</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s2">&quot;data&quot;</span><span class="p">,</span>
                         <span class="n">shape</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">17</span><span class="p">,</span> <span class="mi">13</span><span class="p">),</span>
                         <span class="n">dtype</span><span class="o">=</span><span class="s2">&quot;float32&quot;</span><span class="p">)</span>
<span class="n">normed</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">l2_normalize</span><span class="p">(</span><span class="n">x</span><span class="o">=</span><span class="n">data</span><span class="p">,</span> <span class="n">axis</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="matmul">
<h3>matmul<a class="headerlink" href="#matmul" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">matmul</code><span class="sig-paren">(</span><em>x</em>, <em>y</em>, <em>transpose_x=False</em>, <em>transpose_y=False</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd><p>Applies matrix multiplication to two tensors.</p>
<p>Currently, the input tensors&#8217; rank can be any, but when the rank of any
inputs is bigger than 3, this two inputs&#8217; rank should be equal.</p>
<p>The actual behavior depends on the shapes of <span class="math">\(x\)</span>, <span class="math">\(y\)</span> and the
flag values of <code class="xref py py-attr docutils literal"><span class="pre">transpose_x</span></code>, <code class="xref py py-attr docutils literal"><span class="pre">transpose_y</span></code>. Specifically:</p>
<ul class="simple">
<li>If a transpose flag is specified, the last two dimensions of the tensor
are transposed. If the tensor is rank-1 of shape <span class="math">\([D]\)</span>, then for
<span class="math">\(x\)</span> it is treated as <span class="math">\([1, D]\)</span> in nontransposed form and as
<span class="math">\([D, 1]\)</span> in transposed form, whereas for <span class="math">\(y\)</span> it is the
opposite: It is treated as <span class="math">\([D, 1]\)</span> in nontransposed form and as
<span class="math">\([1, D]\)</span> in transposed form.</li>
<li>After transpose, the two tensors are 2-D or n-D and matrix multiplication
performs in the following way.<ul>
<li>If both are 2-D, they are multiplied like conventional matrices.</li>
<li>If either is n-D, it is treated as a stack of matrices residing in the
last two dimensions and a batched matrix multiply supporting broadcast
applies on the two tensors.</li>
</ul>
</li>
</ul>
<p>Also note that if the raw tensor <span class="math">\(x\)</span> or <span class="math">\(y\)</span> is rank-1 and
nontransposed, the prepended or appended dimension <span class="math">\(1\)</span> will be
removed after matrix multiplication.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> (<em>Variable</em>) &#8211; The input variable which is a Tensor or LoDTensor.</li>
<li><strong>y</strong> (<em>Variable</em>) &#8211; The input variable which is a Tensor or LoDTensor.</li>
<li><strong>transpose_x</strong> (<em>bool</em>) &#8211; Whether to transpose <span class="math">\(x\)</span> before multiplication.</li>
<li><strong>transpose_y</strong> (<em>bool</em>) &#8211; Whether to transpose <span class="math">\(y\)</span> before multiplication.</li>
<li><strong>name</strong> (<em>str|None</em>) &#8211; A name for this layer(optional). If set None, the layer
will be named automatically.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The product Tensor variable.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="c1"># Examples to clarify shapes of the inputs and output</span>
<span class="c1"># x: [B, ..., M, K], y: [B, ..., K, N]</span>
<span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">matmul</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">y</span><span class="p">)</span>  <span class="c1"># out: [B, ..., M, N]</span>

<span class="c1"># x: [B, M, K], y: [B, K, N]</span>
<span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">matmul</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">y</span><span class="p">)</span>  <span class="c1"># out: [B, M, N]</span>

<span class="c1"># x: [B, M, K], y: [K, N]</span>
<span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">matmul</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">y</span><span class="p">)</span>  <span class="c1"># out: [B, M, N]</span>

<span class="c1"># x: [M, K], y: [K, N]</span>
<span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">matmul</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">y</span><span class="p">)</span>  <span class="c1"># out: [M, N]</span>

<span class="c1"># x: [B, M, K], y: [K]</span>
<span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">matmul</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">y</span><span class="p">)</span>  <span class="c1"># out: [B, M]</span>

<span class="c1"># x: [K], y: [K]</span>
<span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">matmul</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">y</span><span class="p">)</span>  <span class="c1"># out: [1]</span>

<span class="c1"># x: [M], y: [N]</span>
<span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">matmul</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="bp">True</span><span class="p">,</span> <span class="bp">True</span><span class="p">)</span>  <span class="c1"># out: [M, N]</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="warpctc">
<h3>warpctc<a class="headerlink" href="#warpctc" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">warpctc</code><span class="sig-paren">(</span><em>input</em>, <em>label</em>, <em>blank=0</em>, <em>norm_by_times=False</em><span class="sig-paren">)</span></dt>
<dd><p>An operator integrating the open source Warp-CTC library
(<a class="reference external" href="https://github.com/baidu-research/warp-ctc">https://github.com/baidu-research/warp-ctc</a>)
to compute Connectionist Temporal Classification (CTC) loss.
It can be aliased as softmax with CTC, since a native softmax activation is
interated to the Warp-CTC library, to to normlize values for each row of the
input tensor.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable</em>) &#8211; (LodTensor, default: LoDTensor&lt;float&gt;),
the unscaled probabilities of variable-length sequences,
which is a 2-D Tensor with LoD information.
It&#8217;s shape is [Lp, num_classes + 1], where Lp is the sum of all input
sequences&#8217; length and num_classes is the true number of classes.
(not including the blank label).</li>
<li><strong>label</strong> (<em>Variable</em>) &#8211; (LodTensor, default: LoDTensor&lt;int&gt;), the ground truth
of variable-length sequence, which is a 2-D Tensor with LoD
information. It is of the shape [Lg, 1], where Lg is th sum of
all labels&#8217; length.</li>
<li><strong>blank</strong> &#8211; (int, default: 0), the blank label index of Connectionist
Temporal Classification (CTC) loss, which is in the
half-opened interval [0, num_classes + 1).</li>
<li><strong>norm_by_times</strong> &#8211; (bool, default: false), whether to normalize</li>
<li><strong>gradients by the number of time-step</strong><strong>, </strong><strong>which is also the</strong> (<em>the</em>) &#8211; </li>
<li><strong>length. There is no need to normalize the gradients</strong> (<em>sequence's</em>) &#8211; </li>
<li><strong>warpctc layer was follewed by a mean_op.</strong> (<em>if</em>) &#8211; </li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The Connectionist Temporal Classification (CTC) loss,
which is a 2-D Tensor of the shape [batch_size, 1].</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
</dd></dl>

</div>
<div class="section" id="sequence-reshape">
<h3>sequence_reshape<a class="headerlink" href="#sequence-reshape" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">sequence_reshape</code><span class="sig-paren">(</span><em>input</em>, <em>new_dim</em><span class="sig-paren">)</span></dt>
<dd><p><strong>Sequence Reshape Layer</strong></p>
<p>This layer will rearrange the input sequences. The new dimension is set by
user. Length of each sequence is computed according to original length,
original dimension and new dimension. The following example will help to
illustrate the function of this layer:</p>
<div class="highlight-text"><div class="highlight"><pre><span></span>x is a LoDTensor:
    x.lod  = [[0, 2, 6]]
    x.data = [[1, 2], [3, 4],
              [5, 6], [7, 8], [9, 10], [11, 12]]
    x.dims = [6, 2]

set new_dim = 4

then out is a LoDTensor:
    out.lod  = [[0, 1, 3]]
    out.data = [[1, 2, 3, 4],
                [5, 6, 7, 8], [9, 10, 11, 12]]
    out.dims = [3, 4]
</pre></div>
</div>
<p>Currently, only 1-level LoDTensor is supported and please make sure
(original length * original dimension) can be divided by new dimension with
no remainder for each sequence.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable</em>) &#8211; (LodTensor, default: LoDTensor&lt;float&gt;), a 2-D LoDTensor
with shape being [N, M] where M for dimension.</li>
<li><strong>new_dim</strong> (<em>int</em>) &#8211; New dimension which the input LoDTensor is reshaped to.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">Reshaped LoDTensor according to new dimension.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">x</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;x&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">5</span><span class="p">,</span> <span class="mi">20</span><span class="p">],</span>
                  <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">,</span> <span class="n">lod_level</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
<span class="n">x_reshaped</span> <span class="o">=</span> <span class="n">layers</span><span class="o">.</span><span class="n">sequence_reshape</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">x</span><span class="p">,</span> <span class="n">new_dim</span><span class="o">=</span><span class="mi">10</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="transpose">
<h3>transpose<a class="headerlink" href="#transpose" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">transpose</code><span class="sig-paren">(</span><em>x</em>, <em>perm</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd><p><strong>transpose Layer</strong></p>
<p>Permute the dimensions of <cite>input</cite> according to <cite>perm</cite>.</p>
<p>The <cite>i</cite>-th dimension  of the returned tensor will correspond to the
perm[i]-th dimension of <cite>input</cite>.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable</em>) &#8211; (Tensor), A Tensor.</li>
<li><strong>perm</strong> (<em>list</em>) &#8211; A permutation of the dimensions of <cite>input</cite>.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">A transposed Tensor.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">x</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;x&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">5</span><span class="p">,</span> <span class="mi">10</span><span class="p">,</span> <span class="mi">15</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">)</span>
<span class="n">x_transposed</span> <span class="o">=</span> <span class="n">layers</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">perm</span><span class="o">=</span><span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">2</span><span class="p">])</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="im2sequence">
<h3>im2sequence<a class="headerlink" href="#im2sequence" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">im2sequence</code><span class="sig-paren">(</span><em>input</em>, <em>filter_size=1</em>, <em>stride=1</em>, <em>padding=0</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd><p>Extracts image patches from the input tensor to form a tensor of shape
{input.batch_size * output_height * output_width, filter_size_H *
filter_size_W * input.channels} which is similar with im2col.
This op use filter / kernel to scan images and convert these images to
sequences. After expanding, the number of time step are
output_height * output_width for an image, in which output_height and
output_width are calculated by below equation:</p>
<div class="math">
\[output\_size = 1 +             (2 * padding + img\_size - block\_size + stride - 1) / stride\]</div>
<p>And the dimension of each time step is block_y * block_x * input.channels.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable</em>) &#8211; The input should be a tensor in NCHW format.</li>
<li><strong>filter_size</strong> (<em>int|tuple|None</em>) &#8211; The filter size. If filter_size is a tuple,
it must contain two integers, (filter_size_H, filter_size_W).
Otherwise, the filter will be a square.</li>
<li><strong>stride</strong> (<em>int|tuple</em>) &#8211; The stride size. If stride is a tuple, it must
contain two integers, (stride_H, stride_W). Otherwise, the
stride_H = stride_W = stride. Default: stride = 1.</li>
<li><strong>padding</strong> (<em>int|tuple</em>) &#8211; The padding size. If padding is a tuple, it can
contain two integers like (padding_H, padding_W) which means
padding_up = padding_down = padding_H and
padding_left = padding_right = padding_W. Or it can use
(padding_up, padding_left, padding_down, padding_right) to indicate
paddings of four direction. Otherwise, a scalar padding means
padding_up = padding_down = padding_left = padding_right = padding
Default: padding = 0.</li>
<li><strong>name</strong> (<em>int</em>) &#8211; The name of this layer. It is optional.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The output is a LoDTensor with shape
{input.batch_size * output_height * output_width,
filter_size_H * filter_size_W * input.channels}.
If we regard output as a matrix, each row of this matrix is
a step of a sequence.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">output</p>
</td>
</tr>
</tbody>
</table>
<p>Examples:</p>
<p>As an example:</p>
<blockquote>
<div><div class="highlight-text"><div class="highlight"><pre><span></span>Given:

x = [[[[ 6.  2.  1.]
       [ 8.  3.  5.]
       [ 0.  2.  6.]]

      [[ 2.  4.  4.]
       [ 6.  3.  0.]
       [ 6.  4.  7.]]]

     [[[ 6.  7.  1.]
       [ 5.  7.  9.]
       [ 2.  4.  8.]]

      [[ 1.  2.  1.]
       [ 1.  3.  5.]
       [ 9.  0.  8.]]]]

x.dims = {2, 2, 3, 3}

And:

filter = [2, 2]
stride = [1, 1]
padding = [0, 0]

Then:

output.data = [[ 6.  2.  8.  3.  2.  4.  6.  3.]
               [ 2.  1.  3.  5.  4.  4.  3.  0.]
               [ 8.  3.  0.  2.  6.  3.  6.  4.]
               [ 3.  5.  2.  6.  3.  0.  4.  7.]
               [ 6.  7.  5.  7.  1.  2.  1.  3.]
               [ 7.  1.  7.  9.  2.  1.  3.  5.]
               [ 5.  7.  2.  4.  1.  3.  9.  0.]
               [ 7.  9.  4.  8.  3.  5.  0.  8.]]

output.dims = {8, 9}

output.lod = [[0, 4, 8]]
</pre></div>
</div>
<p>The simple usage is:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">output</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">im2sequence</span><span class="p">(</span>
    <span class="nb">input</span><span class="o">=</span><span class="n">layer</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span> <span class="n">filter_size</span><span class="o">=</span><span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">])</span>
</pre></div>
</div>
</div></blockquote>
</dd></dl>

</div>
<div class="section" id="nce">
<h3>nce<a class="headerlink" href="#nce" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">nce</code><span class="sig-paren">(</span><em>input</em>, <em>label</em>, <em>num_total_classes</em>, <em>sample_weight=None</em>, <em>param_attr=None</em>, <em>bias_attr=None</em>, <em>num_neg_samples=None</em><span class="sig-paren">)</span></dt>
<dd><p>Compute and return the noise-contrastive estimation training loss.
See [Noise-contrastive estimation: A new estimation principle for unnormalized statistical models](<a class="reference external" href="http://www.jmlr.org/proceedings/papers/v9/gutmann10a/gutmann10a.pdf">http://www.jmlr.org/proceedings/papers/v9/gutmann10a/gutmann10a.pdf</a>).
By default this operator uses a uniform distribution for sampling.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> &#8211; (Tensor) A tensor of shape [batch_size, dim].
Duplicable: False  Optional: False</li>
<li><strong>label</strong> &#8211; (Tensor) A tensor of shape [batch_size, num_true_class]. &#8216;num_true_class&#8217; is the number of target classes in each sample.The number of target classes per sample should be same. If you have a variable number of target classes, you can pad them out to a constant number by either repeating them or by padding with an otherwise unused class.)
Duplicable: False  Optional: False</li>
<li><strong>weight</strong> &#8211; (Tensor) A tensor of shape [num_class, dim]. &#8216;num_class&#8217; is the total number of class.
Duplicable: False  Optional: False</li>
<li><strong>bias</strong> &#8211; (Tensor) A tensor of shape [num_class, 1]. &#8216;num_class&#8217; is the total number of class. It is a dispensable input.
Duplicable: False  Optional: True</li>
<li><strong>sample_weight</strong> &#8211; (Tensor) A tensor of shape [batch_size, 1] storing a weight for each sample. And it is a dispensable input. The default value of sample is 1.
Duplicable: False  Optional: True</li>
<li><strong>num_total_classes</strong> (<em>INT</em>) &#8211; Total number of classes in all samples.</li>
<li><strong>num_neg_samples</strong> (<em>INT</em>) &#8211; The number of negative classes. The default value is 10.</li>
<li><strong>custom_neg_classes</strong> (<em>INTS</em>) &#8211; This attribute only be used in unitest. Classes in this list wiil be used as negative classes for every samples. Under normal conditions, user should avoid setting this attribute.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">(Tensor) A tensor of shape [batch_size, 1]. Cost of samples.</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="beam-search">
<h3>beam_search<a class="headerlink" href="#beam-search" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">beam_search</code><span class="sig-paren">(</span><em>pre_ids</em>, <em>ids</em>, <em>scores</em>, <em>beam_size</em>, <em>end_id</em>, <em>level=0</em><span class="sig-paren">)</span></dt>
<dd><p>This function implements the beam search algorithm.</p>
</dd></dl>

</div>
<div class="section" id="row-conv">
<h3>row_conv<a class="headerlink" href="#row-conv" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">row_conv</code><span class="sig-paren">(</span><em>input</em>, <em>future_context_size</em>, <em>param_attr=None</em>, <em>act=None</em><span class="sig-paren">)</span></dt>
<dd><p>Row Conv Operator. This layer will apply lookahead convolution to
<strong>input</strong>. The input variable should be a 2D LoDTensor with shape [T, D].
Parameters with shape [future_context_size + 1, D] will be created. The math
equation of row convolution is as follows:</p>
<div class="math">
\[Out_{i} = \sum_{j = i} ^ {i + \tau} X_{j} \odot W_{i - j}\]</div>
<p>In the above equation:</p>
<ul class="simple">
<li><span class="math">\(Out_{i}\)</span>: The i-th row of output variable with shape [1, D].</li>
<li><span class="math">\(\tau\)</span>: Future context size.</li>
<li><span class="math">\(X_{j}\)</span>: The j-th row of input variable with shape [1, D].</li>
<li><span class="math">\(W_{i-j}\)</span>: The (i-j)-th row of parameters with shape [1, D].</li>
</ul>
<p>More details about row_conv please refer to the paper     (<a class="reference external" href="http://www.cs.cmu.edu/~dyogatam/papers/wang+etal.iclrworkshop2016.pdf">http://www.cs.cmu.edu/~dyogatam/papers/wang+etal.iclrworkshop2016.pdf</a>) and
the design document     (<a class="reference external" href="https://github.com/PaddlePaddle/Paddle/issues/2228#issuecomment-303903645">https://github.com/PaddlePaddle/Paddle/issues/2228#issuecomment-303903645</a>).</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable</em>) &#8211; Input variable, a 2D LoDTensor with shape [T, D].</li>
<li><strong>future_context_size</strong> (<em>int</em>) &#8211; Future context size. Please note, the shape
of convolution kernel is [future_context_size + 1, D].</li>
<li><strong>param_attr</strong> (<em>ParamAttr</em>) &#8211; Attributes of parameters, including
name, initializer etc.</li>
<li><strong>act</strong> (<em>str</em>) &#8211; Non-linear activation to be applied to output variable.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The output tensor with same shape as input tensor.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">x</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;x&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">16</span><span class="p">],</span>
                <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">,</span> <span class="n">lod_level</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
<span class="n">out</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">row_conv</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">x</span><span class="p">,</span> <span class="n">future_context_size</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="multiplex">
<h3>multiplex<a class="headerlink" href="#multiplex" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">multiplex</code><span class="sig-paren">(</span><em>inputs</em>, <em>index</em><span class="sig-paren">)</span></dt>
<dd><p><strong>Multiplex Layer</strong></p>
<p>Referring to the given index variable, this layer selects rows from the
input variables to construct a multiplex variable. Assuming that there are
<span class="math">\(m\)</span> input variables and <span class="math">\(I_i\)</span> represents the i-th input
variable and <span class="math">\(i\)</span> is in [0, <span class="math">\(m\)</span>). All input variables are
tensors with same shape [<span class="math">\(d_0\)</span>, <span class="math">\(d_1\)</span>, ..., <span class="math">\(d_R\)</span>].
Please note that rank of the input tensor should be at least 2. Each input
variable will be treated as a 2-D matrix with shape [<span class="math">\(M\)</span>, <span class="math">\(N\)</span>]
where <span class="math">\(M\)</span> for <span class="math">\(d_0\)</span> and <span class="math">\(N\)</span> for <span class="math">\(d_1\)</span> * <span class="math">\(d_2\)</span>
* ... * <span class="math">\(d_R\)</span>. Let <span class="math">\(I_i[j]\)</span> be the j-th row of the i-th input
variable. The given index variable should be a 2-D tensor with shape
[<span class="math">\(M\)</span>, 1]. Let <cite>ID[i]</cite> be the i-th index value of the index variable.
Then the output variable will be a tensor with shape [<span class="math">\(d_0\)</span>,
<span class="math">\(d_1\)</span>, ..., <span class="math">\(d_R\)</span>]. If we treat the output tensor as a 2-D
matrix with shape [<span class="math">\(M\)</span>, <span class="math">\(N\)</span>] and let <span class="math">\(O[i]\)</span> be the i-th
row of the matrix, then <cite>O[i]</cite> is equal to <span class="math">\(I_{ID[i]}[i]\)</span>.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>inputs</strong> (<em>list</em>) &#8211; A list of variables to gather from. All variables have the
same shape and the rank is at least 2.</li>
<li><strong>index</strong> (<em>Variable</em>) &#8211; Tensor&lt;int32&gt;, index variable which is a 2-D tensor
with shape [M, 1] where M is the batch size.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">Multiplex variable gathered from input variables.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">x1</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;x1&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">4</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">)</span>
<span class="n">x2</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;x2&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">4</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">)</span>
<span class="n">index</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;index&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;int32&#39;</span><span class="p">)</span>
<span class="n">out</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">multiplex</span><span class="p">(</span><span class="n">inputs</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">index</span><span class="o">=</span><span class="n">index</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
</div>
<div class="section" id="ops">
<h2>ops<a class="headerlink" href="#ops" title="Permalink to this headline">¶</a></h2>
<div class="section" id="mean">
<h3>mean<a class="headerlink" href="#mean" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">mean</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Mean Operator.</p>
<p>Out is a scalar which is the mean of all elements in X.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>x</strong> &#8211; The input of mean op
Duplicable: False  Optional: False</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">The output of mean op</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="mul">
<h3>mul<a class="headerlink" href="#mul" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">mul</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Mul Operator.</p>
<p>This operator is used to perform matrix multiplication for input $X$ and $Y$.</p>
<p>The equation is:</p>
<p>$$Out = X * Y$$</p>
<p>Both the input $X$ and $Y$ can carry the LoD (Level of Details) information,
or not. But the output only shares the LoD information with input $X$.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; (Tensor), The first input tensor of mul op.
Duplicable: False  Optional: False</li>
<li><strong>y</strong> &#8211; (Tensor), The second input tensor of mul op.
Duplicable: False  Optional: False</li>
<li><strong>x_num_col_dims</strong> (<em>INT</em>) &#8211; (int, default 1), The mul_op can take tensors with more than two
dimensions as its inputs. If the input $X$ is a tensor with more
than two dimensions, $X$ will be flattened into a two-dimensional
matrix first. The flattening rule is: the first <cite>num_col_dims</cite>
will be flattened to form the first dimension of the final matrix
(the height of the matrix), and the rest <cite>rank(X) - num_col_dims</cite>
dimensions are flattened to form the second dimension of the final
matrix (the width of the matrix). As a result, height of the
flattened matrix is equal to the product of $X$&#8217;s first
<cite>x_num_col_dims</cite> dimensions&#8217; sizes, and width of the flattened
matrix is equal to the product of $X$&#8217;s last <cite>rank(x) - num_col_dims</cite>
dimensions&#8217; size. For example, suppose $X$ is a 6-dimensional
tensor with the shape [2, 3, 4, 5, 6], and <cite>x_num_col_dims</cite> = 3.
Thus, the flattened matrix will have a shape [2 x 3 x 4, 5 x 6] =
[24, 30].</li>
<li><strong>y_num_col_dims</strong> (<em>INT</em>) &#8211; (int, default 1), The mul_op can take tensors with more than two,
dimensions as its inputs. If the input $Y$ is a tensor with more
than two dimensions, $Y$ will be flattened into a two-dimensional
matrix first. The attribute <cite>y_num_col_dims</cite> determines how $Y$ is
flattened. See comments of <cite>x_num_col_dims</cite> for more details.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">(Tensor), The output tensor of mul op.</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="reshape">
<h3>reshape<a class="headerlink" href="#reshape" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">reshape</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Reshape Operator.</p>
<p>Reshape Input(X) into the shape specified by Attr(shape).</p>
<p>An example:
Given a 2-D tensor X with 2 rows and 2 columns : [[1, 2], [3, 4]]</p>
<p>and target shape = [1, 4], the reshape operator will transform
the tensor X into a 2-D tensor: [[1, 2, 3, 4]]</p>
<p>One dimension in the target shape can be set -1, representing that its
size is unknown. In this case, the real dimension will be infered from
the original shape of Input(X) and other dimensions in the target shape.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; The input tensor of reshape operator.
Duplicable: False  Optional: False</li>
<li><strong>shape</strong> (<em>INTS</em>) &#8211; (vector&lt;int&gt;) Target shape of reshape operator.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">The output tensor of reshape operator.</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="scale">
<h3>scale<a class="headerlink" href="#scale" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">scale</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Scale operator</p>
<p>$$Out = scale*X$$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; (Tensor) Input tensor of scale operator.
Duplicable: False  Optional: False</li>
<li><strong>scale</strong> (<em>FLOAT</em>) &#8211; (float, default 1.0)The scaling factor of the scale operator.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">(Tensor) Output tensor of scale operator.</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="sigmoid-cross-entropy-with-logits">
<h3>sigmoid_cross_entropy_with_logits<a class="headerlink" href="#sigmoid-cross-entropy-with-logits" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">sigmoid_cross_entropy_with_logits</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>SigmoidCrossEntropyWithLogits Operator.</p>
<p>This measures the element-wise probability error in classification tasks
in which each class is independent. This can be thought of as predicting labels
for a data-point, where labels are not mutually exclusive.
For example, a news article can be about politics, technology or sports
at the same time or none of these.</p>
<p>The logistic loss is given as follows:</p>
<blockquote>
<div>$$loss = -Labels * log(sigma(X)) - (1 - Labels) * log(1 - sigma(X))$$</div></blockquote>
<p>We know that $$sigma(X) = (1 / (1 + exp(-X)))$$. By substituting this we get:</p>
<blockquote>
<div>$$loss = X - X * Labels + log(1 + exp(-X))$$</div></blockquote>
<p>For stability and to prevent overflow of $$exp(-X)$$ when X &lt; 0,
we reformulate the loss as follows:</p>
<blockquote>
<div>$$loss = max(X, 0) - X * Labels + log(1 + exp(-<a href="#id3"><span class="problematic" id="id4">|X|</span></a>))$$</div></blockquote>
<p>Both the input <cite>X</cite> and <cite>Labels</cite> can carry the LoD (Level of Details) information.
However the output only shares the LoD with input <cite>X</cite>.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; (Tensor, default Tensor&lt;float&gt;), a 2-D tensor with shape N x D, where N is the batch size and D is the number of classes. This input is a tensor of logits computed by the previous  operator. Logits are unscaled log probabilities given as log(p/(1-p)).
Duplicable: False  Optional: False</li>
<li><strong>label</strong> &#8211; (Tensor, default Tensor&lt;float&gt;), a 2-D tensor of the same type and shape as X. This input is a tensor of probabalistic labels for each logit
Duplicable: False  Optional: False</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">(Tensor, default Tensor&lt;float&gt;), a 2-D tensor with shape N x D  of elementwise logistic losses.</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="elementwise-add">
<h3>elementwise_add<a class="headerlink" href="#elementwise-add" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">elementwise_add</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Limited Elementwise Add Operator.</p>
<p>The equation is:</p>
<p>$$Out = X + Y$$</p>
<p>$X$ is a tensor of any dimension and the dimensions of tensor $Y$ must be
smaller than or equal to the dimensions of $X$.</p>
<p>There are two cases for this operator:
1. The shape of $Y$ is same with $X$;
2. The shape of $Y$ is a subset of $X$.</p>
<p>For case 2:
$Y$ will be broadcasted to match the shape of $X$ and axis should be
set to index of the start dimension to broadcast $Y$ onto $X$.</p>
<dl class="docutils">
<dt>For example</dt>
<dd><div class="first last highlight-python"><div class="highlight"><pre><span></span><span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(,)</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">5</span><span class="p">,)</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">)</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span> <span class="k">with</span> <span class="n">axis</span><span class="o">=</span><span class="mi">1</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="k">with</span> <span class="n">axis</span><span class="o">=</span><span class="mi">0</span>
</pre></div>
</div>
</dd>
</dl>
<p>Either of the inputs $X$ and $Y$ or none can carry the LoD (Level of Details)
information. However, the output only shares the LoD information with input $X$.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; (Tensor), The first input tensor of elementwise op.
Duplicable: False  Optional: False</li>
<li><strong>y</strong> &#8211; (Tensor), The second input tensor of elementwise op.
Duplicable: False  Optional: False</li>
<li><strong>axis</strong> (<em>INT</em>) &#8211; (int, default -1). The start dimension index for broadcasting Y onto X.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">The output of elementwise op.</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="elementwise-div">
<h3>elementwise_div<a class="headerlink" href="#elementwise-div" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">elementwise_div</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Limited Elementwise Div Operator.</p>
<p>The equation is:</p>
<p>$$Out = X / Y$$</p>
<p>$X$ is a tensor of any dimension and the dimensions of tensor $Y$ must be
smaller than or equal to the dimensions of $X$.</p>
<p>There are two cases for this operator:
1. The shape of $Y$ is same with $X$;
2. The shape of $Y$ is a subset of $X$.</p>
<p>For case 2:
$Y$ will be broadcasted to match the shape of $X$ and axis should be
set to index of the start dimension to broadcast $Y$ onto $X$.</p>
<dl class="docutils">
<dt>For example</dt>
<dd><div class="first last highlight-python"><div class="highlight"><pre><span></span><span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(,)</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">5</span><span class="p">,)</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">)</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span> <span class="k">with</span> <span class="n">axis</span><span class="o">=</span><span class="mi">1</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="k">with</span> <span class="n">axis</span><span class="o">=</span><span class="mi">0</span>
</pre></div>
</div>
</dd>
</dl>
<p>Either of the inputs $X$ and $Y$ or none can carry the LoD (Level of Details)
information. However, the output only shares the LoD information with input $X$.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; (Tensor), The first input tensor of elementwise op.
Duplicable: False  Optional: False</li>
<li><strong>y</strong> &#8211; (Tensor), The second input tensor of elementwise op.
Duplicable: False  Optional: False</li>
<li><strong>axis</strong> (<em>INT</em>) &#8211; (int, default -1). The start dimension index for broadcasting Y onto X.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">The output of elementwise op.</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="elementwise-sub">
<h3>elementwise_sub<a class="headerlink" href="#elementwise-sub" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">elementwise_sub</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Limited Elementwise Sub Operator.</p>
<p>The equation is:</p>
<p>$$Out = X - Y$$</p>
<p>$X$ is a tensor of any dimension and the dimensions of tensor $Y$ must be
smaller than or equal to the dimensions of $X$.</p>
<p>There are two cases for this operator:
1. The shape of $Y$ is same with $X$;
2. The shape of $Y$ is a subset of $X$.</p>
<p>For case 2:
$Y$ will be broadcasted to match the shape of $X$ and axis should be
set to index of the start dimension to broadcast $Y$ onto $X$.</p>
<dl class="docutils">
<dt>For example</dt>
<dd><div class="first last highlight-python"><div class="highlight"><pre><span></span><span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(,)</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">5</span><span class="p">,)</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">)</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span> <span class="k">with</span> <span class="n">axis</span><span class="o">=</span><span class="mi">1</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="k">with</span> <span class="n">axis</span><span class="o">=</span><span class="mi">0</span>
</pre></div>
</div>
</dd>
</dl>
<p>Either of the inputs $X$ and $Y$ or none can carry the LoD (Level of Details)
information. However, the output only shares the LoD information with input $X$.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; (Tensor), The first input tensor of elementwise op.
Duplicable: False  Optional: False</li>
<li><strong>y</strong> &#8211; (Tensor), The second input tensor of elementwise op.
Duplicable: False  Optional: False</li>
<li><strong>axis</strong> (<em>INT</em>) &#8211; (int, default -1). The start dimension index for broadcasting Y onto X.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">The output of elementwise op.</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="elementwise-mul">
<h3>elementwise_mul<a class="headerlink" href="#elementwise-mul" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">elementwise_mul</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Limited Elementwise Mul Operator.</p>
<p>The equation is:</p>
<p>$$Out = X odotY$$</p>
<p>$X$ is a tensor of any dimension and the dimensions of tensor $Y$ must be
smaller than or equal to the dimensions of $X$.</p>
<p>There are two cases for this operator:
1. The shape of $Y$ is same with $X$;
2. The shape of $Y$ is a subset of $X$.</p>
<p>For case 2:
$Y$ will be broadcasted to match the shape of $X$ and axis should be
set to index of the start dimension to broadcast $Y$ onto $X$.</p>
<dl class="docutils">
<dt>For example</dt>
<dd><div class="first last highlight-python"><div class="highlight"><pre><span></span><span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(,)</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">5</span><span class="p">,)</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">)</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span> <span class="k">with</span> <span class="n">axis</span><span class="o">=</span><span class="mi">1</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="k">with</span> <span class="n">axis</span><span class="o">=</span><span class="mi">0</span>
</pre></div>
</div>
</dd>
</dl>
<p>Either of the inputs $X$ and $Y$ or none can carry the LoD (Level of Details)
information. However, the output only shares the LoD information with input $X$.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; (Tensor), The first input tensor of elementwise op.
Duplicable: False  Optional: False</li>
<li><strong>y</strong> &#8211; (Tensor), The second input tensor of elementwise op.
Duplicable: False  Optional: False</li>
<li><strong>axis</strong> (<em>INT</em>) &#8211; (int, default -1). The start dimension index for broadcasting Y onto X.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">The output of elementwise op.</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="elementwise-max">
<h3>elementwise_max<a class="headerlink" href="#elementwise-max" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">elementwise_max</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Limited Elementwise Max Operator.</p>
<p>The equation is:</p>
<p>$$Out = max(X, Y)$$</p>
<p>$X$ is a tensor of any dimension and the dimensions of tensor $Y$ must be
smaller than or equal to the dimensions of $X$.</p>
<p>There are two cases for this operator:
1. The shape of $Y$ is same with $X$;
2. The shape of $Y$ is a subset of $X$.</p>
<p>For case 2:
$Y$ will be broadcasted to match the shape of $X$ and axis should be
set to index of the start dimension to broadcast $Y$ onto $X$.</p>
<dl class="docutils">
<dt>For example</dt>
<dd><div class="first last highlight-python"><div class="highlight"><pre><span></span><span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(,)</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">5</span><span class="p">,)</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">)</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span> <span class="k">with</span> <span class="n">axis</span><span class="o">=</span><span class="mi">1</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="k">with</span> <span class="n">axis</span><span class="o">=</span><span class="mi">0</span>
</pre></div>
</div>
</dd>
</dl>
<p>Either of the inputs $X$ and $Y$ or none can carry the LoD (Level of Details)
information. However, the output only shares the LoD information with input $X$.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; (Tensor), The first input tensor of elementwise op.
Duplicable: False  Optional: False</li>
<li><strong>y</strong> &#8211; (Tensor), The second input tensor of elementwise op.
Duplicable: False  Optional: False</li>
<li><strong>axis</strong> (<em>INT</em>) &#8211; (int, default -1). The start dimension index for broadcasting Y onto X.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">The output of elementwise op.</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="elementwise-min">
<h3>elementwise_min<a class="headerlink" href="#elementwise-min" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">elementwise_min</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Limited Elementwise Max Operator.</p>
<p>The equation is:</p>
<p>$$Out = min(X, Y)$$</p>
<p>$X$ is a tensor of any dimension and the dimensions of tensor $Y$ must be
smaller than or equal to the dimensions of $X$.</p>
<p>There are two cases for this operator:
1. The shape of $Y$ is same with $X$;
2. The shape of $Y$ is a subset of $X$.</p>
<p>For case 2:
$Y$ will be broadcasted to match the shape of $X$ and axis should be
set to index of the start dimension to broadcast $Y$ onto $X$.</p>
<dl class="docutils">
<dt>For example</dt>
<dd><div class="first last highlight-python"><div class="highlight"><pre><span></span><span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(,)</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">5</span><span class="p">,)</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">)</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span> <span class="k">with</span> <span class="n">axis</span><span class="o">=</span><span class="mi">1</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="k">with</span> <span class="n">axis</span><span class="o">=</span><span class="mi">0</span>
</pre></div>
</div>
</dd>
</dl>
<p>Either of the inputs $X$ and $Y$ or none can carry the LoD (Level of Details)
information. However, the output only shares the LoD information with input $X$.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; (Tensor), The first input tensor of elementwise op.
Duplicable: False  Optional: False</li>
<li><strong>y</strong> &#8211; (Tensor), The second input tensor of elementwise op.
Duplicable: False  Optional: False</li>
<li><strong>axis</strong> (<em>INT</em>) &#8211; (int, default -1). The start dimension index for broadcasting Y onto X.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">The output of elementwise op.</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="elementwise-pow">
<h3>elementwise_pow<a class="headerlink" href="#elementwise-pow" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">elementwise_pow</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Limited Elementwise Pow Operator.</p>
<p>The equation is:</p>
<p>$$Out = X ^ Y$$</p>
<p>$X$ is a tensor of any dimension and the dimensions of tensor $Y$ must be
smaller than or equal to the dimensions of $X$.</p>
<p>There are two cases for this operator:
1. The shape of $Y$ is same with $X$;
2. The shape of $Y$ is a subset of $X$.</p>
<p>For case 2:
$Y$ will be broadcasted to match the shape of $X$ and axis should be
set to index of the start dimension to broadcast $Y$ onto $X$.</p>
<dl class="docutils">
<dt>For example</dt>
<dd><div class="first last highlight-python"><div class="highlight"><pre><span></span><span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(,)</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">5</span><span class="p">,)</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">)</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span> <span class="k">with</span> <span class="n">axis</span><span class="o">=</span><span class="mi">1</span>
<span class="n">shape</span><span class="p">(</span><span class="n">X</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="n">shape</span><span class="p">(</span><span class="n">Y</span><span class="p">)</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="k">with</span> <span class="n">axis</span><span class="o">=</span><span class="mi">0</span>
</pre></div>
</div>
</dd>
</dl>
<p>Either of the inputs $X$ and $Y$ or none can carry the LoD (Level of Details)
information. However, the output only shares the LoD information with input $X$.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; (Tensor), The first input tensor of elementwise op.
Duplicable: False  Optional: False</li>
<li><strong>y</strong> &#8211; (Tensor), The second input tensor of elementwise op.
Duplicable: False  Optional: False</li>
<li><strong>axis</strong> (<em>INT</em>) &#8211; (int, default -1). The start dimension index for broadcasting Y onto X.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">The output of elementwise op.</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="clip">
<h3>clip<a class="headerlink" href="#clip" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">clip</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Clip Operator.</p>
<p>The clip operator limits the value of given input within an interval. The
interval is specified with arguments &#8216;min&#8217; and &#8216;max&#8217;:</p>
<p>$$
Out = min(max(X, min), max)
$$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; (Tensor)The input of clip op.The number of dimensions must be between [1, 9].
Duplicable: False  Optional: False</li>
<li><strong>min</strong> (<em>FLOAT</em>) &#8211; (float)Minimum value, under which element is replaced by min.</li>
<li><strong>max</strong> (<em>FLOAT</em>) &#8211; (float)Maximum value, above which element is replaced by max</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">(Tensor)The output of clip op with shape as input(X)</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="clip-by-norm">
<h3>clip_by_norm<a class="headerlink" href="#clip-by-norm" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">clip_by_norm</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>ClipByNorm Operator.</p>
<p>This operator limits the L2 norm of the input $X$ within $max_norm$.
If the L2 norm of $X$ is less than or equal to $max_norm$, $Out$ will be
the same as $X$. If the L2 norm of $X$ is greater than $max_norm$, $X$ will
be linearly scaled to make the L2 norm of $Out$ equal to $max_norm$, as
shown in the following formula:</p>
<p>$$
Out = frac{max_norm * X}{norm(X)},
$$</p>
<p>where $norm(X)$ represents the L2 norm of $X$.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; (Tensor) The input of clip_by_norm op.The number of dimensions must be between [1, 9].
Duplicable: False  Optional: False</li>
<li><strong>max_norm</strong> (<em>FLOAT</em>) &#8211; (float) The maximum norm value.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">(Tensor) The output of clip_by_norm op with shape as input(X)</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="sequence-softmax">
<h3>sequence_softmax<a class="headerlink" href="#sequence-softmax" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">sequence_softmax</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Sequence Softmax Operator.</p>
<p>SequenceSoftmaxOp computes the softmax activation among all time-steps for each
sequence. The dimension of each time-step should be 1. Thus, the shape of
input Tensor can be either [N, 1] or [N], where N is the sum of the length
of all sequences.</p>
<p>The algorithm works as follows:</p>
<blockquote>
<div>for i-th sequence in a mini-batch:</div></blockquote>
<p>$$
Out(X[lod[i]:lod[i+1]], :) = frac{exp(X[lod[i]:lod[i+1], :])} {sum(exp(X[lod[i]:lod[i+1], :]))}
$$</p>
<p>For example, for a mini-batch of 3 sequences with variable-length,
each containing 2, 3, 2 time-steps, the lod of which is [0, 2, 5, 7],
then softmax will be computed among X[0:2, :], X[2:5, :], X[5:7, :]
and N turns out to be 7.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>x</strong> &#8211; (LoDTensor) 1-D or 2-D input LoDTensor with the 2-nd dimension of length 1.
Duplicable: False  Optional: False</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">(LoDTensor) 1-D or 2-D output LoDTensor with the 2-nd dimension of length 1.</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="sigmoid">
<h3>sigmoid<a class="headerlink" href="#sigmoid" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">sigmoid</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Sigmoid Activation Operator</p>
<p>$$out = frac{1}{1 + e^{-x}}$$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>x</strong> &#8211; Input of Sigmoid operator
Duplicable: False  Optional: False</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">Output of Sigmoid operator</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="logsigmoid">
<h3>logsigmoid<a class="headerlink" href="#logsigmoid" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">logsigmoid</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Logsigmoid Activation Operator</p>
<p>$$out = log frac{1}{1 + e^{-x}}$$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>x</strong> &#8211; Input of LogSigmoid operator
Duplicable: False  Optional: False</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">Output of LogSigmoid operator</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="exp">
<h3>exp<a class="headerlink" href="#exp" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">exp</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Exp Activation Operator.</p>
<p>$out = e^x$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>x</strong> &#8211; Input of Exp operator
Duplicable: False  Optional: False</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">Output of Exp operator</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="relu">
<h3>relu<a class="headerlink" href="#relu" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">relu</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Relu Activation Operator.</p>
<p>$out = max(x, 0)$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>x</strong> &#8211; Input of Relu operator
Duplicable: False  Optional: False</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">Output of Relu operator</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="tanh">
<h3>tanh<a class="headerlink" href="#tanh" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">tanh</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Tanh Activation Operator.</p>
<p>$$out = frac{e^{x} - e^{-x}}{e^{x} + e^{-x}}$$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>x</strong> &#8211; Input of Tanh operator
Duplicable: False  Optional: False</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">Output of Tanh operator</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="tanh-shrink">
<h3>tanh_shrink<a class="headerlink" href="#tanh-shrink" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">tanh_shrink</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>TanhShrink Activation Operator.</p>
<p>$$out = x - frac{e^{x} - e^{-x}}{e^{x} + e^{-x}}$$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>x</strong> &#8211; Input of TanhShrink operator
Duplicable: False  Optional: False</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">Output of TanhShrink operator</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="softshrink">
<h3>softshrink<a class="headerlink" href="#softshrink" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">softshrink</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Softshrink Activation Operator.</p>
<p>$$
out = begin{cases}</p>
<blockquote>
<div>x - lambda, text{if } x &gt; lambda \
x + lambda, text{if } x &lt; -lambda \
0,  text{otherwise}
end{cases}</div></blockquote>
<p>$$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; Input of Softshrink operator
Duplicable: False  Optional: False</li>
<li><strong>lambda</strong> (<em>FLOAT</em>) &#8211; non-negative offset</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">Output of Softshrink operator</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="sqrt">
<h3>sqrt<a class="headerlink" href="#sqrt" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">sqrt</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Sqrt Activation Operator.</p>
<p>$out = sqrt{x}$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>x</strong> &#8211; Input of Sqrt operator
Duplicable: False  Optional: False</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">Output of Sqrt operator</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="abs">
<h3>abs<a class="headerlink" href="#abs" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">abs</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Abs Activation Operator.</p>
<p>$out = <a href="#id1"><span class="problematic" id="id2">|</span></a>x|$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>x</strong> &#8211; Input of Abs operator
Duplicable: False  Optional: False</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">Output of Abs operator</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="ceil">
<h3>ceil<a class="headerlink" href="#ceil" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">ceil</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Ceil Activation Operator.</p>
<p>$out = ceil(x)$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>x</strong> &#8211; Input of Ceil operator
Duplicable: False  Optional: False</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">Output of Ceil operator</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="floor">
<h3>floor<a class="headerlink" href="#floor" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">floor</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Floor Activation Operator.</p>
<p>$out = floor(x)$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>x</strong> &#8211; Input of Floor operator
Duplicable: False  Optional: False</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">Output of Floor operator</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="round">
<h3>round<a class="headerlink" href="#round" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">round</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Round Activation Operator.</p>
<p>$out = [x]$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>x</strong> &#8211; Input of Round operator
Duplicable: False  Optional: False</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">Output of Round operator</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="reciprocal">
<h3>reciprocal<a class="headerlink" href="#reciprocal" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">reciprocal</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Reciprocal Activation Operator.</p>
<p>$$out = frac{1}{x}$$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>x</strong> &#8211; Input of Reciprocal operator
Duplicable: False  Optional: False</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">Output of Reciprocal operator</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="log">
<h3>log<a class="headerlink" href="#log" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">log</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Log Activation Operator.</p>
<p>$out = ln(x)$</p>
<p>Natural logarithm of x.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>x</strong> &#8211; Input of Log operator
Duplicable: False  Optional: False</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">Output of Log operator</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="square">
<h3>square<a class="headerlink" href="#square" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">square</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Square Activation Operator.</p>
<p>$out = x^2$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>x</strong> &#8211; Input of Square operator
Duplicable: False  Optional: False</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">Output of Square operator</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="softplus">
<h3>softplus<a class="headerlink" href="#softplus" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">softplus</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Softplus Activation Operator.</p>
<p>$out = ln(1 + e^{x})$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>x</strong> &#8211; Input of Softplus operator
Duplicable: False  Optional: False</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">Output of Softplus operator</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="softsign">
<h3>softsign<a class="headerlink" href="#softsign" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">softsign</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Softsign Activation Operator.</p>
<p>$$out = frac{x}{1 + <a href="#id5"><span class="problematic" id="id6">|x|</span></a>}$$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>x</strong> &#8211; Input of Softsign operator
Duplicable: False  Optional: False</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">Output of Softsign operator</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="brelu">
<h3>brelu<a class="headerlink" href="#brelu" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">brelu</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>BRelu Activation Operator.</p>
<p>$out = max(min(x, t_{min}), t_{max})$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; Input of BRelu operator
Duplicable: False  Optional: False</li>
<li><strong>t_min</strong> (<em>FLOAT</em>) &#8211; The min marginal value of BRelu</li>
<li><strong>t_max</strong> (<em>FLOAT</em>) &#8211; The max marginal value of BRelu</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">Output of BRelu operator</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="leaky-relu">
<h3>leaky_relu<a class="headerlink" href="#leaky-relu" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">leaky_relu</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>LeakyRelu Activation Operator.</p>
<p>$out = max(x, alpha * x)$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; Input of LeakyRelu operator
Duplicable: False  Optional: False</li>
<li><strong>alpha</strong> (<em>FLOAT</em>) &#8211; The small negative slope</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">Output of LeakyRelu operator</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="soft-relu">
<h3>soft_relu<a class="headerlink" href="#soft-relu" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">soft_relu</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>SoftRelu Activation Operator.</p>
<p>$out = ln(1 + exp(max(min(x, threshold), threshold))$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; Input of SoftRelu operator
Duplicable: False  Optional: False</li>
<li><strong>threshold</strong> (<em>FLOAT</em>) &#8211; The threshold value of SoftRelu</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">Output of SoftRelu operator</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="elu">
<h3>elu<a class="headerlink" href="#elu" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">elu</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>ELU Activation Operator.</p>
<p>Applies the following element-wise computation on the input according to
<a class="reference external" href="https://arxiv.org/abs/1511.07289">https://arxiv.org/abs/1511.07289</a>.</p>
<p>$out = max(0, x) + min(0, alpha * (e^x - 1))$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; Input of ELU operator
Duplicable: False  Optional: False</li>
<li><strong>alpha</strong> (<em>FLOAT</em>) &#8211; The alpha value of ELU</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">Output of ELU operator</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="relu6">
<h3>relu6<a class="headerlink" href="#relu6" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">relu6</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Relu6 Activation Operator.</p>
<p>$out = min(max(0, x), 6)$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; Input of Relu6 operator
Duplicable: False  Optional: False</li>
<li><strong>threshold</strong> (<em>FLOAT</em>) &#8211; The threshold value of Relu6</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">Output of Relu6 operator</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="pow">
<h3>pow<a class="headerlink" href="#pow" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">pow</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Pow Activation Operator.</p>
<p>$out = x^{factor}$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; Input of Pow operator
Duplicable: False  Optional: False</li>
<li><strong>factor</strong> (<em>FLOAT</em>) &#8211; The exponential factor of Pow</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">Output of Pow operator</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="stanh">
<h3>stanh<a class="headerlink" href="#stanh" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">stanh</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>STanh Activation Operator.</p>
<p>$$out = b * frac{e^{a * x} - e^{-a * x}}{e^{a * x} + e^{-a * x}}$$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; Input of STanh operator
Duplicable: False  Optional: False</li>
<li><strong>scale_a</strong> (<em>FLOAT</em>) &#8211; The scale parameter of a for the input</li>
<li><strong>scale_b</strong> (<em>FLOAT</em>) &#8211; The scale parameter of b for the input</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">Output of STanh operator</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="hard-shrink">
<h3>hard_shrink<a class="headerlink" href="#hard-shrink" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">hard_shrink</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>HardShrink Activation Operator.</p>
<p>$$
out = begin{cases}</p>
<blockquote>
<div>x, text{if } x &gt; lambda \
x, text{if } x &lt; -lambda \
0,  text{otherwise}
end{cases}</div></blockquote>
<p>$$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; Input of HardShrink operator
Duplicable: False  Optional: False</li>
<li><strong>threshold</strong> (<em>FLOAT</em>) &#8211; The value of threshold for HardShrink</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">Output of HardShrink operator</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="thresholded-relu">
<h3>thresholded_relu<a class="headerlink" href="#thresholded-relu" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">thresholded_relu</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>ThresholdedRelu Activation Operator.</p>
<p>$$
out = begin{cases}</p>
<blockquote>
<div>x, text{if } x &gt; threshold \
0,  text{otherwise}
end{cases}</div></blockquote>
<p>$$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; Input of ThresholdedRelu operator
Duplicable: False  Optional: False</li>
<li><strong>threshold</strong> (<em>FLOAT</em>) &#8211; The threshold location of activation</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">Output of ThresholdedRelu operator</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="hard-sigmoid">
<h3>hard_sigmoid<a class="headerlink" href="#hard-sigmoid" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">hard_sigmoid</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>HardSigmoid Activation Operator.</p>
<p>Segment-wise linear approximation of sigmoid(<a class="reference external" href="https://arxiv.org/abs/1603.00391">https://arxiv.org/abs/1603.00391</a>),
which is much faster than sigmoid.</p>
<p>$out = max(0, min(1, slope * x + shift))$</p>
<p>The slope should be positive. The offset can be either positive or negative.
The default slope and shift are set according to the above reference.
It is recommended to use the defaults for this activation.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; Input of HardSigmoid operator
Duplicable: False  Optional: False</li>
<li><strong>slope</strong> (<em>FLOAT</em>) &#8211; Slope for linear approximation of sigmoid</li>
<li><strong>offset</strong> (<em>FLOAT</em>) &#8211; Offset for linear approximation of sigmoid</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">Output of HardSigmoid operator</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="swish">
<h3>swish<a class="headerlink" href="#swish" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">swish</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Swish Activation Operator.</p>
<p>$$out = frac{x}{1 + e^{- beta x}}$$</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>x</strong> &#8211; Input of Swish operator
Duplicable: False  Optional: False</li>
<li><strong>beta</strong> (<em>FLOAT</em>) &#8211; Constant beta of swish operator</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last">Output of Swish operator</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
</div>
<div class="section" id="tensor">
<h2>tensor<a class="headerlink" href="#tensor" title="Permalink to this headline">¶</a></h2>
<div class="section" id="create-tensor">
<h3>create_tensor<a class="headerlink" href="#create-tensor" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">create_tensor</code><span class="sig-paren">(</span><em>dtype</em>, <em>name=None</em>, <em>persistable=False</em><span class="sig-paren">)</span></dt>
<dd></dd></dl>

</div>
<div class="section" id="create-parameter">
<h3>create_parameter<a class="headerlink" href="#create-parameter" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">create_parameter</code><span class="sig-paren">(</span><em>shape</em>, <em>dtype</em>, <em>name=None</em>, <em>attr=None</em>, <em>is_bias=False</em>, <em>default_initializer=None</em><span class="sig-paren">)</span></dt>
<dd><p>Create a parameter
:param shape: shape of the parameter
:type shape: list[int]
:param dtype: element type of the parameter
:type dtype: string
:param attr: attributes of the parameter
:type attr: ParamAttr
:param is_bias: This can affect which default initializer is chosen</p>
<blockquote>
<div>when default_initializer is None. If is_bias,
initializer.Constant(0.0) will be used. Otherwise,
Xavier() will be used.</div></blockquote>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>default_initializer</strong> (<em>Initializer</em>) &#8211; initializer for the parameter</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">the created parameter</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body">Parameter</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="create-global-var">
<h3>create_global_var<a class="headerlink" href="#create-global-var" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">create_global_var</code><span class="sig-paren">(</span><em>shape</em>, <em>value</em>, <em>dtype</em>, <em>persistable=False</em>, <em>force_cpu=False</em>, <em>name=None</em><span class="sig-paren">)</span></dt>
<dd><p>Create a global variable. such as global_step
:param shape: shape of the variable
:type shape: list[int]
:param value: the value of the variable
:type value: float
:param dtype: element type of the parameter
:type dtype: string
:param persistable: if this variable is persistable
:type persistable: bool
:param force_cpu: force this variable to be on CPU
:type force_cpu: bool</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Returns:</th><td class="field-body">the created Variable</td>
</tr>
<tr class="field-even field"><th class="field-name">Return type:</th><td class="field-body">Variable</td>
</tr>
</tbody>
</table>
</dd></dl>

</div>
<div class="section" id="cast">
<h3>cast<a class="headerlink" href="#cast" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">cast</code><span class="sig-paren">(</span><em>x</em>, <em>dtype</em><span class="sig-paren">)</span></dt>
<dd><p>This function takes in the input with input_dtype
and casts it to the output_dtype as the output.</p>
</dd></dl>

</div>
<div class="section" id="concat">
<h3>concat<a class="headerlink" href="#concat" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">concat</code><span class="sig-paren">(</span><em>input</em>, <em>axis=0</em><span class="sig-paren">)</span></dt>
<dd><p><strong>Concat</strong></p>
<p>This function concatenates the input along the axis mentioned
and returns that as the output.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>list</em>) &#8211; List of tensors to be concatenated</li>
<li><strong>axis</strong> (<em>int</em>) &#8211; Integer axis along which the tensors will be concatenated</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">Output variable of the concatenation</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
</dd></dl>

</div>
<div class="section" id="sums">
<h3>sums<a class="headerlink" href="#sums" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">sums</code><span class="sig-paren">(</span><em>input</em>, <em>out=None</em><span class="sig-paren">)</span></dt>
<dd><p>This function performs the sum operation on the input and returns the
result as the output.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>input</strong> (<em>Variable|list</em>) &#8211; The input tensor that has the elements
that need to be summed up.</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><dl class="docutils">
<dt>The tensor type variable that has the sum of input</dt>
<dd>written to it.</dd>
</dl>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body">Variable</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
</dd></dl>

</div>
<div class="section" id="assign">
<h3>assign<a class="headerlink" href="#assign" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">assign</code><span class="sig-paren">(</span><em>input</em>, <em>output</em><span class="sig-paren">)</span></dt>
<dd><p><strong>Assign</strong></p>
<p>This function copies the <em>input</em> Variable to the <em>output</em> Variable.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable|numpy.ndarray</em>) &#8211; The source variable</li>
<li><strong>output</strong> (<em>Variable</em>) &#8211; The destination variable</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The destination variable that was supplied as the <em>output</em>.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
</dd></dl>

</div>
<div class="section" id="fill-constant-batch-size-like">
<h3>fill_constant_batch_size_like<a class="headerlink" href="#fill-constant-batch-size-like" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">fill_constant_batch_size_like</code><span class="sig-paren">(</span><em>input</em>, <em>shape</em>, <em>dtype</em>, <em>value</em>, <em>input_dim_idx=0</em>, <em>output_dim_idx=0</em><span class="sig-paren">)</span></dt>
<dd><p><strong>fill_constant_batch_size_like</strong></p>
<p>This function creates a tensor of specified <em>shape</em>, <em>dtype</em> and batch size,
and initializes this with a constant supplied in <em>value</em>. The batch size is
obtained from the <cite>input</cite> tensor.</p>
<p>It also sets <em>stop_gradient</em> to True.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>Variable</em>) &#8211; Tensor whose dimensions will be used to get batch size</li>
<li><strong>shape</strong> (<em>tuple|list|None</em>) &#8211; Shape of output tensor</li>
<li><strong>dtype</strong> (<em>np.dtype|core.VarDesc.VarType|str</em>) &#8211; Data type of output tensor</li>
<li><strong>value</strong> (<em>float</em>) &#8211; Constant value to initialize the output tensor</li>
<li><strong>input_dim_idx</strong> (<em>int</em>) &#8211; Index of input&#8217;s batch size dimension</li>
<li><strong>output_dim_idx</strong> (<em>int</em>) &#8211; Index of output&#8217;s batch size dimension</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The tensor variable storing the output</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">data</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">fill_constant_batch_size_like</span><span class="p">(</span>
    <span class="nb">input</span><span class="o">=</span><span class="n">like</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">value</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;int64&#39;</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="fill-constant">
<h3>fill_constant<a class="headerlink" href="#fill-constant" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">fill_constant</code><span class="sig-paren">(</span><em>shape</em>, <em>dtype</em>, <em>value</em>, <em>force_cpu=False</em>, <em>out=None</em><span class="sig-paren">)</span></dt>
<dd><p><strong>fill_constant</strong></p>
<p>This function creates a tensor with specified <cite>shape</cite> and <cite>dtype</cite>, and
initializes it with a constant specifed by <cite>value</cite>.</p>
<p>The attribute <cite>stop_gradient</cite> of the created tensor is set to True.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>shape</strong> (<em>tuple|list|None</em>) &#8211; Shape of the output tensor.</li>
<li><strong>dtype</strong> (<em>np.dtype|core.VarDesc.VarType|str</em>) &#8211; Data type of the output tensor.</li>
<li><strong>value</strong> (<em>float</em>) &#8211; The constant value used to initialize the output tensor.</li>
<li><strong>out</strong> (<em>Variable</em>) &#8211; The output tensor.</li>
<li><strong>force_cpu</strong> (<em>True|False</em>) &#8211; data should be on CPU if set true.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The tensor variable storing the output.</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">data</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">fill_constant</span><span class="p">(</span><span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">value</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;int64&#39;</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="ones">
<h3>ones<a class="headerlink" href="#ones" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">ones</code><span class="sig-paren">(</span><em>shape</em>, <em>dtype</em>, <em>force_cpu=False</em><span class="sig-paren">)</span></dt>
<dd><p><strong>ones</strong></p>
<p>This function creates a tensor of specified <em>shape</em> and
<em>dtype</em>, and initializes this with 1.</p>
<p>It also sets <em>stop_gradient</em> to True.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>shape</strong> (<em>tuple|list|None</em>) &#8211; Shape of output tensor</li>
<li><strong>dtype</strong> (<em>np.dtype|core.VarDesc.VarType|str</em>) &#8211; Data type of output tensor</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The tensor variable storing the output</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">data</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">ones</span><span class="p">(</span><span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;int64&#39;</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
<div class="section" id="zeros">
<h3>zeros<a class="headerlink" href="#zeros" title="Permalink to this headline">¶</a></h3>
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">zeros</code><span class="sig-paren">(</span><em>shape</em>, <em>dtype</em>, <em>force_cpu=False</em><span class="sig-paren">)</span></dt>
<dd><p><strong>zeros</strong></p>
<p>This function creates a tensor of specified <em>shape</em> and
<em>dtype</em>, and initializes this with 0.</p>
<p>It also sets <em>stop_gradient</em> to True.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>shape</strong> (<em>tuple|list|None</em>) &#8211; Shape of output tensor</li>
<li><strong>dtype</strong> (<em>np.dtype|core.VarDesc.VarType|str</em>) &#8211; Data type of output tensor</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The tensor variable storing the output</p>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
</td>
</tr>
</tbody>
</table>
<p class="rubric">Examples</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">data</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;int64&#39;</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>

</div>
</div>
</div>


           </div>
          </div>
          <footer>
  
    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
      
        <a href="data_feeder.html" class="btn btn-neutral float-right" title="data_feeder" accesskey="n">Next <span class="fa fa-arrow-circle-right"></span></a>
      
      
        <a href="../fluid.html" class="btn btn-neutral" title="Fluid" accesskey="p"><span class="fa fa-arrow-circle-left"></span> Previous</a>
      
    </div>
  

  <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>