layers.html 60.0 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115


<!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="DataFeeder" 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="../../../getstarted/index_en.html">GET STARTED</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../../howto/index_en.html">HOW TO</a></li>
<li class="toctree-l1 current"><a class="reference internal" href="../../index_en.html">API</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../../mobile/index_en.html">MOBILE</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="../../../getstarted/index_en.html">GET STARTED</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../../../getstarted/build_and_install/index_en.html">Install and Build</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../../../getstarted/build_and_install/pip_install_en.html">Install Using pip</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../../getstarted/build_and_install/docker_install_en.html">Run in Docker Containers</a></li>
116
<li class="toctree-l3"><a class="reference internal" href="../../../howto/dev/build_en.html">Build using Docker</a></li>
117 118 119 120 121 122 123 124 125 126 127 128 129 130 131
<li class="toctree-l3"><a class="reference internal" href="../../../getstarted/build_and_install/build_from_source_en.html">Build from Sources</a></li>
</ul>
</li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../../howto/index_en.html">HOW TO</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../../../howto/usage/cmd_parameter/index_en.html">Set Command-line Parameters</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../../../howto/usage/cmd_parameter/use_case_en.html">Use Case</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../../howto/usage/cmd_parameter/arguments_en.html">Argument Outline</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../../howto/usage/cmd_parameter/detail_introduction_en.html">Detail Description</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../../../howto/usage/cluster/cluster_train_en.html">PaddlePaddle Distributed Training</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../../howto/dev/new_layer_en.html">Write New Layers</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../../howto/dev/contribute_to_paddle_en.html">Contribute Code</a></li>
132
<li class="toctree-l2"><a class="reference internal" href="../../../howto/dev/write_docs_en.html">Contribute Documentation</a></li>
133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174
<li class="toctree-l2"><a class="reference internal" href="../../../howto/deep_model/rnn/index_en.html">RNN Models</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../../../howto/deep_model/rnn/rnn_config_en.html">RNN Configuration</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../../../howto/optimization/gpu_profiling_en.html">Tune GPU Performance</a></li>
</ul>
</li>
<li class="toctree-l1 current"><a class="reference internal" href="../../index_en.html">API</a><ul class="current">
<li class="toctree-l2"><a class="reference internal" href="../model_configs.html">Model Configuration</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../config/activation.html">Activation</a></li>
<li class="toctree-l3"><a class="reference internal" href="../config/layer.html">Layers</a></li>
<li class="toctree-l3"><a class="reference internal" href="../config/evaluators.html">Evaluators</a></li>
<li class="toctree-l3"><a class="reference internal" href="../config/optimizer.html">Optimizer</a></li>
<li class="toctree-l3"><a class="reference internal" href="../config/pooling.html">Pooling</a></li>
<li class="toctree-l3"><a class="reference internal" href="../config/networks.html">Networks</a></li>
<li class="toctree-l3"><a class="reference internal" href="../config/attr.html">Parameter Attribute</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../data.html">Data Reader Interface and DataSets</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../data/data_reader.html">Data Reader Interface</a></li>
<li class="toctree-l3"><a class="reference internal" href="../data/image.html">Image Interface</a></li>
<li class="toctree-l3"><a class="reference internal" href="../data/dataset.html">Dataset</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../run_logic.html">Training and Inference</a></li>
<li class="toctree-l2 current"><a class="reference internal" href="../fluid.html">Fluid</a><ul class="current">
<li class="toctree-l3 current"><a class="current reference internal" href="#">Layers</a></li>
<li class="toctree-l3"><a class="reference internal" href="data_feeder.html">DataFeeder</a></li>
<li class="toctree-l3"><a class="reference internal" href="executor.html">Executor</a></li>
<li class="toctree-l3"><a class="reference internal" href="initializer.html">Initializer</a></li>
<li class="toctree-l3"><a class="reference internal" href="evaluator.html">Evaluator</a></li>
<li class="toctree-l3"><a class="reference internal" href="nets.html">Nets</a></li>
<li class="toctree-l3"><a class="reference internal" href="optimizer.html">Optimizer</a></li>
<li class="toctree-l3"><a class="reference internal" href="param_attr.html">ParamAttr</a></li>
<li class="toctree-l3"><a class="reference internal" href="profiler.html">Profiler</a></li>
<li class="toctree-l3"><a class="reference internal" href="regularizer.html">Regularizer</a></li>
</ul>
</li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../../mobile/index_en.html">MOBILE</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../../../mobile/cross_compiling_for_android_en.html">Build PaddlePaddle for Android</a></li>
175
<li class="toctree-l2"><a class="reference internal" href="../../../mobile/cross_compiling_for_ios_en.html">PaddlePaddle Compiling Guide for iOS</a></li>
176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215
<li class="toctree-l2"><a class="reference internal" href="../../../mobile/cross_compiling_for_raspberry_en.html">Build PaddlePaddle for Raspberry Pi</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="../../index_en.html">API</a> > </li>
      
        <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="fc">
<h2>fc<a class="headerlink" href="#fc" title="Permalink to this headline"></a></h2>
216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250
<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>, <em>main_program=None</em>, <em>startup_program=None</em><span class="sig-paren">)</span></dt>
<dd><p>Fully Connected 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>input</strong> &#8211; The input tensor to the function</li>
<li><strong>size</strong> &#8211; The size of the layer</li>
<li><strong>num_flatten_dims</strong> &#8211; Number of columns in input</li>
<li><strong>param_attr</strong> &#8211; The parameters/weights to the FC Layer</li>
<li><strong>param_initializer</strong> &#8211; Initializer used for the weight/parameter. If None, XavierInitializer() is used</li>
<li><strong>bias_attr</strong> &#8211; The bias parameter for the FC layer</li>
<li><strong>bias_initializer</strong> &#8211; Initializer used for the bias. If None, then ConstantInitializer() is used</li>
<li><strong>act</strong> &#8211; Activation to be applied to the output of FC layer</li>
<li><strong>name</strong> &#8211; Name/alias of the function</li>
<li><strong>main_program</strong> &#8211; Name of the main program that calls this</li>
<li><strong>startup_program</strong> &#8211; Name of the startup program</li>
</ul>
</td>
</tr>
</tbody>
</table>
<p>This function can take in multiple inputs and performs the Fully Connected
function (linear transformation) on top of each of them.
So for input x, the output will be : Wx + b. Where W is the parameter,
b the bias and x is the input.</p>
<p>The function also applies an activation (non-linearity) on top of the
output, if activation is passed in the input.</p>
<p>All the input variables of this function are passed in as local variables
to the LayerHelper constructor.</p>
</dd></dl>

251 252 253
</div>
<div class="section" id="embedding">
<h2>embedding<a class="headerlink" href="#embedding" title="Permalink to this headline"></a></h2>
254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282
<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>param_attr=None</em>, <em>dtype='float32'</em>, <em>main_program=None</em>, <em>startup_program=None</em><span class="sig-paren">)</span></dt>
<dd><p>Embedding 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>param_initializer</strong> &#8211; </li>
<li><strong>input</strong> &#8211; The input to the function</li>
<li><strong>size</strong> &#8211; The size of the layer</li>
<li><strong>is_sparse</strong> &#8211; A flag that decleares whether the input is sparse</li>
<li><strong>param_attr</strong> &#8211; Parameters for this layer</li>
<li><strong>dtype</strong> &#8211; The type of data : float32, float_16, int etc</li>
<li><strong>main_program</strong> &#8211; Name of the main program that calls this</li>
<li><strong>startup_program</strong> &#8211; Name of the startup program</li>
</ul>
</td>
</tr>
</tbody>
</table>
<p>This function can take in the input (which is a vector of IDs) and
performs a lookup in the lookup_table using these IDs, to result into
the embedding of each ID in the input.</p>
<p>All the input variables of this function are passed in as local variables
to the LayerHelper constructor.</p>
</dd></dl>

283 284 285
</div>
<div class="section" id="dynamic-lstm">
<h2>dynamic_lstm<a class="headerlink" href="#dynamic-lstm" title="Permalink to this headline"></a></h2>
286 287 288 289 290
<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>main_program=None</em>, <em>startup_program=None</em><span class="sig-paren">)</span></dt>
<dd></dd></dl>

291 292 293
</div>
<div class="section" id="data">
<h2>data<a class="headerlink" href="#data" title="Permalink to this headline"></a></h2>
294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324
<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>main_program=None</em>, <em>startup_program=None</em>, <em>stop_gradient=True</em><span class="sig-paren">)</span></dt>
<dd><p>Data 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>name</strong> &#8211; The name/alias of the function</li>
<li><strong>shape</strong> &#8211; Tuple declaring the shape.</li>
<li><strong>append_batch_size</strong> &#8211; Whether or not to append the data as a batch.</li>
<li><strong>dtype</strong> &#8211; The type of data : float32, float_16, int etc</li>
<li><strong>type</strong> &#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> &#8211; Name of the main program that calls this</li>
<li><strong>startup_program</strong> &#8211; Name of the startup program</li>
<li><strong>stop_gradient</strong> &#8211; A boolean that mentions whether gradient should flow.</li>
</ul>
</td>
</tr>
</tbody>
</table>
<p>This function takes in input and based on whether data has
to be returned back as a minibatch, it creates the global variable using
the helper functions. The global variables can be accessed by all the
following operations and layers in the graph.</p>
<p>All the input variables of this function are passed in as local variables
to the LayerHelper constructor.</p>
</dd></dl>

325 326 327
</div>
<div class="section" id="mean">
<h2>mean<a class="headerlink" href="#mean" title="Permalink to this headline"></a></h2>
328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345
<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>

346 347 348
</div>
<div class="section" id="mul">
<h2>mul<a class="headerlink" href="#mul" title="Permalink to this headline"></a></h2>
349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384
<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>
<blockquote>
<div>$$Out = X * Y$$</div></blockquote>
<p>Both the input <cite>X</cite> and <cite>Y</cite> can carry the LoD (Level of Details) information,
or not. But the output only shares the LoD information 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; The first input of mul op
Duplicable: False  Optional: False</li>
<li><strong>y</strong> &#8211; The second input of mul op
Duplicable: False  Optional: False</li>
<li><strong>x_num_col_dims</strong> (<em>INT</em>) &#8211; (int, default 1) mul_op can take tensors with more than two dimensions as input <cite>X</cite>,
in that case, tensors will be reshaped to a matrix. The matrix&#8217;s first
dimension(column length) will be the product of tensor&#8217;s last
<cite>num_col_dims</cite> dimensions, and the matrix&#8217;s second dimension(row length)
will be the product of tensor&#8217;s first <cite>rank - num_col_dims</cite> dimensions.</li>
<li><strong>y_num_col_dims</strong> (<em>INT</em>) &#8211; (int, default 1) mul_op can take tensors with more than two dimensions as input <cite>Y</cite>,
in that case, tensors will be reshaped to a matrix. Just like input <cite>X</cite>.</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 mul op</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

385 386 387
</div>
<div class="section" id="elementwise-add">
<h2>elementwise_add<a class="headerlink" href="#elementwise-add" title="Permalink to this headline"></a></h2>
388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429
<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
the starting dimension index for broadcasting Y onto X.</p>
<p class="rubric">example</p>
<p>shape(X) = (2, 3, 4, 5), shape(Y) = (,)
shape(X) = (2, 3, 4, 5), shape(Y) = (5,)
shape(X) = (2, 3, 4, 5), shape(Y) = (4, 5)
shape(X) = (2, 3, 4, 5), shape(Y) = (3, 4), with axis=1
shape(X) = (2, 3, 4, 5), shape(Y) = (2), with axis=0</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 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 starting 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>

430 431 432
</div>
<div class="section" id="elementwise-div">
<h2>elementwise_div<a class="headerlink" href="#elementwise-div" title="Permalink to this headline"></a></h2>
433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474
<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
the starting dimension index for broadcasting Y onto X.</p>
<p class="rubric">example</p>
<p>shape(X) = (2, 3, 4, 5), shape(Y) = (,)
shape(X) = (2, 3, 4, 5), shape(Y) = (5,)
shape(X) = (2, 3, 4, 5), shape(Y) = (4, 5)
shape(X) = (2, 3, 4, 5), shape(Y) = (3, 4), with axis=1
shape(X) = (2, 3, 4, 5), shape(Y) = (2), with axis=0</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 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 starting 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>

475 476 477
</div>
<div class="section" id="dropout">
<h2>dropout<a class="headerlink" href="#dropout" title="Permalink to this headline"></a></h2>
478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">dropout</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Dropout Operator.</p>
<p>Dropout refers to randomly dropping out units in a nerual network. It 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 set equal to their corresponding inputs.</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 of dropout op.
Duplicable: False  Optional: False</li>
<li><strong>dropout_prob</strong> (<em>FLOAT</em>) &#8211; Probability of setting units to zero.</li>
<li><strong>is_test</strong> (<em>BOOLEAN</em>) &#8211; True if in test phase.</li>
<li><strong>seed</strong> (<em>INT</em>) &#8211; Dropout random seed.</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 dropout op.</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

507 508 509
</div>
<div class="section" id="reshape">
<h2>reshape<a class="headerlink" href="#reshape" title="Permalink to this headline"></a></h2>
510 511 512 513 514 515 516 517 518 519
<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</p>
<blockquote>
<div>[[1, 2], [3, 4]]</div></blockquote>
<p>and target shape = [1, 4], the reshape operator will transform
520
the tensor X into a 2-D tensor:</p>
521
<blockquote>
522
<div>[[1, 2, 3, 4]]</div></blockquote>
523 524 525
<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>
526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543
<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>

544 545 546
</div>
<div class="section" id="sigmoid">
<h2>sigmoid<a class="headerlink" href="#sigmoid" title="Permalink to this headline"></a></h2>
547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564
<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>$$y = frac{1}{1 + e^{-x}}$$</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>

565 566 567
</div>
<div class="section" id="scale">
<h2>scale<a class="headerlink" href="#scale" title="Permalink to this headline"></a></h2>
568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590
<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 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>

591 592 593
</div>
<div class="section" id="id1">
<h2>reshape<a class="headerlink" href="#id1" title="Permalink to this headline"></a></h2>
594 595 596 597 598 599 600 601 602 603
<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</p>
<blockquote>
<div>[[1, 2], [3, 4]]</div></blockquote>
<p>and target shape = [1, 4], the reshape operator will transform
604
the tensor X into a 2-D tensor:</p>
605
<blockquote>
606
<div>[[1, 2, 3, 4]]</div></blockquote>
607 608 609
<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>
610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627
<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>

628 629 630
</div>
<div class="section" id="transpose">
<h2>transpose<a class="headerlink" href="#transpose" title="Permalink to this headline"></a></h2>
631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674
<dl class="function">
<dt>
<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">transpose</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>Transpose Operator.</p>
<p>The input tensor will be permuted according to the axis values given.
The op functions similar to how numpy.transpose works in python.
For example:</p>
<blockquote>
<div><p>&gt;&gt; input = numpy.arange(6).reshape((2,3))
&gt;&gt; input
array([[0, 1, 2],</p>
<blockquote>
<div>[3, 4, 5]])</div></blockquote>
<p>&gt;&gt; axis = [1, 0]
&gt;&gt; output = input.transpose(axis)
&gt;&gt; output
array([[0, 3],</p>
<blockquote>
<div><dl class="docutils">
<dt>[1, 4],</dt>
<dd>[2, 5]])</dd>
</dl>
</div></blockquote>
</div></blockquote>
<p>So, given a input tensor of shape(N, C, H, W) and the axis is {0, 2, 3, 1},
the output tensor shape will be (N, H, W, C)</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 tensor, tensors with rank at most 6 are supported
Duplicable: False  Optional: False</li>
<li><strong>axis</strong> (<em>INTS</em>) &#8211; (vector&lt;int&gt;)A list of values, and the size of the list should be the same with the input tensor rank, the tensor will permute the axes according the the values given</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</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

675 676 677 678 679 680
</div>
<div class="section" id="sigmoid-cross-entropy-with-logits">
<h2>sigmoid_cross_entropy_with_logits<a class="headerlink" href="#sigmoid-cross-entropy-with-logits" title="Permalink to this headline"></a></h2>
</div>
<div class="section" id="cast">
<h2>cast<a class="headerlink" href="#cast" title="Permalink to this headline"></a></h2>
681 682 683 684 685 686 687
<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>, <em>main_program=None</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>

688 689 690
</div>
<div class="section" id="concat">
<h2>concat<a class="headerlink" href="#concat" title="Permalink to this headline"></a></h2>
691 692 693 694 695 696 697
<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</em>, <em>main_program=None</em>, <em>startup_program=None</em><span class="sig-paren">)</span></dt>
<dd><p>This function concats the input along the axis mentioned
and returns that as the output.</p>
</dd></dl>

698 699 700
</div>
<div class="section" id="sums">
<h2>sums<a class="headerlink" href="#sums" title="Permalink to this headline"></a></h2>
701 702 703 704 705 706 707
<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>, <em>main_program=None</em>, <em>startup_program=None</em><span class="sig-paren">)</span></dt>
<dd><p>This function takes in the input and performs the sum operation on it
and returns that as the output.</p>
</dd></dl>

708 709 710
</div>
<div class="section" id="linear-chain-crf">
<h2>linear_chain_crf<a class="headerlink" href="#linear-chain-crf" title="Permalink to this headline"></a></h2>
711 712 713 714 715
<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>, <em>main_program=None</em>, <em>startup_program=None</em><span class="sig-paren">)</span></dt>
<dd></dd></dl>

716 717 718
</div>
<div class="section" id="assign">
<h2>assign<a class="headerlink" href="#assign" title="Permalink to this headline"></a></h2>
719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747
<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>param_attr=None</em>, <em>dtype='float32'</em>, <em>main_program=None</em>, <em>startup_program=None</em><span class="sig-paren">)</span></dt>
<dd><p>Embedding 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>param_initializer</strong> &#8211; </li>
<li><strong>input</strong> &#8211; The input to the function</li>
<li><strong>size</strong> &#8211; The size of the layer</li>
<li><strong>is_sparse</strong> &#8211; A flag that decleares whether the input is sparse</li>
<li><strong>param_attr</strong> &#8211; Parameters for this layer</li>
<li><strong>dtype</strong> &#8211; The type of data : float32, float_16, int etc</li>
<li><strong>main_program</strong> &#8211; Name of the main program that calls this</li>
<li><strong>startup_program</strong> &#8211; Name of the startup program</li>
</ul>
</td>
</tr>
</tbody>
</table>
<p>This function can take in the input (which is a vector of IDs) and
performs a lookup in the lookup_table using these IDs, to result into
the embedding of each ID in the input.</p>
<p>All the input variables of this function are passed in as local variables
to the LayerHelper constructor.</p>
</dd></dl>

748 749 750
</div>
<div class="section" id="split-lod-tensor">
<h2>split_lod_tensor<a class="headerlink" href="#split-lod-tensor" title="Permalink to this headline"></a></h2>
751 752 753 754 755
<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>, <em>main_program=None</em>, <em>startup_program=None</em><span class="sig-paren">)</span></dt>
<dd></dd></dl>

756 757 758
</div>
<div class="section" id="merge-lod-tensor">
<h2>merge_lod_tensor<a class="headerlink" href="#merge-lod-tensor" title="Permalink to this headline"></a></h2>
759 760 761 762 763
<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>, <em>main_program=None</em>, <em>startup_program=None</em><span class="sig-paren">)</span></dt>
<dd></dd></dl>

764 765 766
</div>
<div class="section" id="cos-sim">
<h2>cos_sim<a class="headerlink" href="#cos-sim" title="Permalink to this headline"></a></h2>
767 768 769 770 771 772 773
<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>, <em>**kwargs</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>

774 775 776
</div>
<div class="section" id="cross-entropy">
<h2>cross_entropy<a class="headerlink" href="#cross-entropy" title="Permalink to this headline"></a></h2>
777 778 779 780 781 782
<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>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>This function computes cross_entropy using the input and label.</p>
</dd></dl>

783 784 785
</div>
<div class="section" id="square-error-cost">
<h2>square_error_cost<a class="headerlink" href="#square-error-cost" title="Permalink to this headline"></a></h2>
786 787 788 789 790 791 792
<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>, <em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>This functions returns the squared error cost using the input and label.
The output is appending the op to do the above.</p>
</dd></dl>

793 794 795
</div>
<div class="section" id="accuracy">
<h2>accuracy<a class="headerlink" href="#accuracy" title="Permalink to this headline"></a></h2>
796 797 798 799 800 801 802
<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>, <em>**kwargs</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>

803 804 805
</div>
<div class="section" id="sequence-conv">
<h2>sequence_conv<a class="headerlink" href="#sequence-conv" title="Permalink to this headline"></a></h2>
806 807 808 809 810 811 812 813
<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>, <em>main_program=None</em>, <em>startup_program=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>

814 815 816
</div>
<div class="section" id="conv2d">
<h2>conv2d<a class="headerlink" href="#conv2d" title="Permalink to this headline"></a></h2>
817 818
<dl class="function">
<dt>
819
<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>act=None</em>, <em>name=None</em>, <em>main_program=None</em>, <em>startup_program=None</em><span class="sig-paren">)</span></dt>
820 821 822 823 824 825 826
<dd><p>This function creates the op for a 2-dimensional Convolution.
This is performed using the parameters of filters(size, dimensionality etc)
, stride and other configurations for a Convolution operation.
This funciton can also append an activation on top of the
conv-2d output, if mentioned in the input parameters.</p>
</dd></dl>

827 828 829
</div>
<div class="section" id="sequence-pool">
<h2>sequence_pool<a class="headerlink" href="#sequence-pool" title="Permalink to this headline"></a></h2>
830 831 832 833 834 835 836 837
<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>, <em>**kwargs</em><span class="sig-paren">)</span></dt>
<dd><p>This function add the operator for sequence pooling.
This is applied on top of the input using pool_type mentioned
in the parameters.</p>
</dd></dl>

838 839 840
</div>
<div class="section" id="pool2d">
<h2>pool2d<a class="headerlink" href="#pool2d" title="Permalink to this headline"></a></h2>
841 842
<dl class="function">
<dt>
843
<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>main_program=None</em>, <em>startup_program=None</em><span class="sig-paren">)</span></dt>
844 845 846 847
<dd><p>This function adds the operator for pooling in 2 dimensions, using the
pooling configurations mentioned in input parameters.</p>
</dd></dl>

848 849 850
</div>
<div class="section" id="batch-norm">
<h2>batch_norm<a class="headerlink" href="#batch-norm" title="Permalink to this headline"></a></h2>
851 852 853 854 855 856 857
<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>main_program=None</em>, <em>startup_program=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>

858 859 860
</div>
<div class="section" id="beam-search-decode">
<h2>beam_search_decode<a class="headerlink" href="#beam-search-decode" title="Permalink to this headline"></a></h2>
861 862 863 864 865
<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>main_program=None</em>, <em>startup_program=None</em><span class="sig-paren">)</span></dt>
<dd></dd></dl>

866 867 868 869 870 871
</div>
<div class="section" id="lstm">
<h2>lstm<a class="headerlink" href="#lstm" title="Permalink to this headline"></a></h2>
</div>
<div class="section" id="lod-rank-table">
<h2>lod_rank_table<a class="headerlink" href="#lod-rank-table" title="Permalink to this headline"></a></h2>
872 873 874 875 876 877 878
<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>, <em>main_program=None</em><span class="sig-paren">)</span></dt>
<dd><p>This function creates an operator for creating a LOD_RANK_TABLE
using the input x.</p>
</dd></dl>

879 880 881
</div>
<div class="section" id="max-sequence-len">
<h2>max_sequence_len<a class="headerlink" href="#max-sequence-len" title="Permalink to this headline"></a></h2>
882 883 884 885 886 887 888
<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>, <em>main_program=None</em><span class="sig-paren">)</span></dt>
<dd><p>This function creates an operator to calculate the length of
max seqence through input rank_table(should be a lod_rank_table)</p>
</dd></dl>

889 890 891
</div>
<div class="section" id="topk">
<h2>topk<a class="headerlink" href="#topk" title="Permalink to this headline"></a></h2>
892 893 894 895 896
<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>, <em>main_program=None</em>, <em>startup_program=None</em><span class="sig-paren">)</span></dt>
<dd></dd></dl>

897 898 899
</div>
<div class="section" id="lod-tensor-to-array">
<h2>lod_tensor_to_array<a class="headerlink" href="#lod-tensor-to-array" title="Permalink to this headline"></a></h2>
900 901 902 903 904 905 906
<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>, <em>main_program=None</em><span class="sig-paren">)</span></dt>
<dd><p>This function creates an operator to convert an LOD_Tensor to
an array.</p>
</dd></dl>

907 908 909
</div>
<div class="section" id="array-to-lod-tensor">
<h2>array_to_lod_tensor<a class="headerlink" href="#array-to-lod-tensor" title="Permalink to this headline"></a></h2>
910 911 912 913 914 915 916
<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>, <em>main_program=None</em>, <em>startup_program=None</em><span class="sig-paren">)</span></dt>
<dd><p>This function creates an operator to convert an array to a
LOD_Tensor.</p>
</dd></dl>

917 918 919
</div>
<div class="section" id="fill-constant">
<h2>fill_constant<a class="headerlink" href="#fill-constant" title="Permalink to this headline"></a></h2>
920 921 922 923 924 925 926 927
<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>out=None</em>, <em>main_program=None</em>, <em>startup_program=None</em><span class="sig-paren">)</span></dt>
<dd><p>This function creates a tensor , with shape as mentioned in the input and
specified dtype and fills this up with a constant value that
comes in the input. It also sets the stop_gradient to be True.</p>
</dd></dl>

928 929 930
</div>
<div class="section" id="fill-constant-batch-size-like">
<h2>fill_constant_batch_size_like<a class="headerlink" href="#fill-constant-batch-size-like" title="Permalink to this headline"></a></h2>
931 932 933 934 935
<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>, <em>main_program=None</em>, <em>startup_program=None</em><span class="sig-paren">)</span></dt>
<dd></dd></dl>

936 937 938
</div>
<div class="section" id="ones">
<h2>ones<a class="headerlink" href="#ones" title="Permalink to this headline"></a></h2>
939 940 941 942 943 944 945
<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>main_program=None</em><span class="sig-paren">)</span></dt>
<dd><p>This function performs the same function as fill_constant() declared above
with the constant value being 1.0.</p>
</dd></dl>

946 947 948
</div>
<div class="section" id="zeros">
<h2>zeros<a class="headerlink" href="#zeros" title="Permalink to this headline"></a></h2>
949 950 951 952 953 954 955
<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>main_program=None</em><span class="sig-paren">)</span></dt>
<dd><p>This function performs the same function as fill_constant() declared above
with the constant value being 0.0.</p>
</dd></dl>

956 957 958
</div>
<div class="section" id="increment">
<h2>increment<a class="headerlink" href="#increment" title="Permalink to this headline"></a></h2>
959 960 961 962 963 964 965 966
<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>, <em>main_program=None</em>, <em>startup_program=None</em><span class="sig-paren">)</span></dt>
<dd><p>This function creates an operator to increment each value in the input
<cite>x</cite> by an amount: <cite>value</cite> as mentioned in the input parameter. This
operation is performed in-place by default.</p>
</dd></dl>

967 968 969
</div>
<div class="section" id="array-write">
<h2>array_write<a class="headerlink" href="#array-write" title="Permalink to this headline"></a></h2>
970 971 972 973 974 975 976
<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>, <em>main_program=None</em>, <em>startup_program=None</em><span class="sig-paren">)</span></dt>
<dd><p>This function creates an operator to write the data out as a
LOD_TENSOR_ARRAY.</p>
</dd></dl>

977 978 979
</div>
<div class="section" id="create-array">
<h2>create_array<a class="headerlink" href="#create-array" title="Permalink to this headline"></a></h2>
980 981 982 983 984
<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>, <em>main_program=None</em><span class="sig-paren">)</span></dt>
<dd></dd></dl>

985 986 987
</div>
<div class="section" id="less-than">
<h2>less_than<a class="headerlink" href="#less-than" title="Permalink to this headline"></a></h2>
988 989 990 991 992
<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>main_program=None</em>, <em>**ignored</em><span class="sig-paren">)</span></dt>
<dd></dd></dl>

993 994 995
</div>
<div class="section" id="array-read">
<h2>array_read<a class="headerlink" href="#array-read" title="Permalink to this headline"></a></h2>
996 997 998 999 1000 1001 1002
<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>, <em>main_program=None</em>, <em>startup_program=None</em><span class="sig-paren">)</span></dt>
<dd><p>This function creates an operator to read the data in as a
LOD_TENSOR_ARRAY.</p>
</dd></dl>

1003 1004 1005
</div>
<div class="section" id="shrink-memory">
<h2>shrink_memory<a class="headerlink" href="#shrink-memory" title="Permalink to this headline"></a></h2>
1006 1007 1008 1009 1010 1011 1012
<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>, <em>main_program=None</em>, <em>startup_program=None</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>

1013 1014 1015
</div>
<div class="section" id="array-length">
<h2>array_length<a class="headerlink" href="#array-length" title="Permalink to this headline"></a></h2>
1016 1017 1018 1019 1020 1021 1022
<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>, <em>main_program=None</em><span class="sig-paren">)</span></dt>
<dd><p>This function creates an operator to find the length of the
LOD_TENSOR_ARRAY.</p>
</dd></dl>

1023 1024 1025
</div>
<div class="section" id="conv2d-transpose">
<h2>conv2d_transpose<a class="headerlink" href="#conv2d-transpose" title="Permalink to this headline"></a></h2>
1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067
<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>param_attr=None</em>, <em>main_program=None</em>, <em>startup_program=None</em><span class="sig-paren">)</span></dt>
<dd><p>The transpose of conv2d layer.</p>
<p>This layer is also known as deconvolution 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 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>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.</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.</li>
<li><strong>param_attr</strong> &#8211; Parameter Attribute.</li>
<li><strong>main_program</strong> (<em>Program</em>) &#8211; the main program</li>
<li><strong>startup_program</strong> (<em>Program</em>) &#8211; the startup program</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">Output image.</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>
</dd></dl>

1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140
</div>
<div class="section" id="sequence-expand">
<h2>sequence_expand<a class="headerlink" href="#sequence-expand" title="Permalink to this headline"></a></h2>
<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>main_program=None</em>, <em>startup_program=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>main_program</strong> (<em>Program</em>) &#8211; The main program.</li>
<li><strong>startup_program</strong> (<em>Program</em>) &#8211; The startup program.</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>

1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209
</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="DataFeeder" 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>