run_logic.html 35.0 KB
Newer Older
1 2 3 4 5 6 7 8 9 10


<!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">
  
11
  <title>Training and Inference &mdash; PaddlePaddle  文档</title>
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
  

  
  

  

  
  
    

  

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

  
  
        <link rel="index" title="索引"
              href="../../genindex.html"/>
        <link rel="search" title="搜索" href="../../search.html"/>
    <link rel="top" title="PaddlePaddle  文档" href="../../index.html"/>
        <link rel="up" title="API" href="../index_cn.html"/>
        <link rel="next" title="FAQ" href="../../faq/index_cn.html"/>
38
        <link rel="prev" title="Dataset" href="data/dataset.html"/> 
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

  <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">
68
        <a class="fork-on-github" href="https://github.com/PaddlePaddle/Paddle" target="_blank"><i class="fa fa-github"></i>Fork me on Github</a>
69 70 71 72 73 74 75 76 77 78 79 80
        <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">
81
          <li><a href="/">Home</a></li>
82 83 84 85 86 87 88 89 90
        </ul>
      </div>
      <div class="doc-module">
        
        <ul class="current">
<li class="toctree-l1"><a class="reference internal" href="../../getstarted/index_cn.html">新手入门</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../howto/index_cn.html">进阶指南</a></li>
<li class="toctree-l1 current"><a class="reference internal" href="../index_cn.html">API</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
91
<li class="toctree-l1"><a class="reference internal" href="../../mobile/index_cn.html">MOBILE</a></li>
92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114
</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_cn.html">新手入门</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../../getstarted/build_and_install/index_cn.html">安装与编译</a><ul>
115
<li class="toctree-l3"><a class="reference internal" href="../../getstarted/build_and_install/docker_install_cn.html">PaddlePaddle的Docker容器使用方式</a></li>
116 117 118
<li class="toctree-l3"><a class="reference internal" href="../../getstarted/build_and_install/cmake/build_from_source_cn.html">PaddlePaddle的编译选项</a></li>
</ul>
</li>
119
<li class="toctree-l2"><a class="reference internal" href="../../getstarted/concepts/use_concepts_cn.html">基本使用概念</a></li>
120 121 122 123 124 125 126 127 128
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../howto/index_cn.html">进阶指南</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/cmd_parameter/index_cn.html">设置命令行参数</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../../howto/usage/cmd_parameter/use_case_cn.html">使用案例</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/usage/cmd_parameter/arguments_cn.html">参数概述</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/usage/cmd_parameter/detail_introduction_cn.html">细节描述</a></li>
</ul>
</li>
129
<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/cluster/cluster_train_cn.html">PaddlePaddle分布式训练</a></li>
130 131 132
<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/k8s/k8s_basis_cn.html">Kubernetes 简介</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/k8s/k8s_cn.html">Kubernetes单机训练</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/k8s/k8s_distributed_cn.html">Kubernetes分布式训练</a></li>
133
<li class="toctree-l2"><a class="reference internal" href="../../howto/dev/build_cn.html">编译PaddlePaddle和运行单元测试</a></li>
134 135
<li class="toctree-l2"><a class="reference internal" href="../../howto/dev/write_docs_cn.html">如何贡献/修改文档</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../howto/deep_model/rnn/index_cn.html">RNN相关模型</a><ul>
136
<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/rnn_config_cn.html">RNN配置</a></li>
137 138 139 140 141 142 143 144 145
<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/recurrent_group_cn.html">Recurrent Group教程</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/hierarchical_layer_cn.html">支持双层序列作为输入的Layer</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/hrnn_rnn_api_compare_cn.html">单双层RNN API对比介绍</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../../howto/optimization/gpu_profiling_cn.html">GPU性能分析与调优</a></li>
</ul>
</li>
<li class="toctree-l1 current"><a class="reference internal" href="../index_cn.html">API</a><ul class="current">
146 147 148
<li class="toctree-l2"><a class="reference internal" href="model_configs.html">模型配置</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>
149
<li class="toctree-l3"><a class="reference internal" href="config/evaluators.html">Evaluators</a></li>
150 151 152 153 154 155
<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>
156 157 158 159 160 161
<li class="toctree-l2"><a class="reference internal" href="data.html">数据访问</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>
162
<li class="toctree-l2 current"><a class="current reference internal" href="#">训练与应用</a></li>
163 164
</ul>
</li>
165 166 167 168 169 170 171 172
<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
</ul>
</li>
173 174 175 176 177 178
<li class="toctree-l1"><a class="reference internal" href="../../mobile/index_cn.html">MOBILE</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../../mobile/cross_compiling_for_android_cn.html">构建Android平台上的PaddlePaddle库</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../mobile/cross_compiling_for_ios_cn.html">构建iOS平台上的PaddlePaddle库</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../mobile/cross_compiling_for_raspberry_cn.html">构建Raspberry Pi平台上的PaddlePaddle库</a></li>
</ul>
</li>
179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200
</ul>

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

      

 







<div role="navigation" aria-label="breadcrumbs navigation">
  <ul class="wy-breadcrumbs">
      
        <li><a href="../index_cn.html">API</a> > </li>
      
201
    <li>Training and Inference</li>
202 203 204 205 206 207 208 209
  </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">
            
210 211 212 213
  <div class="section" id="training-and-inference">
<h1>Training and Inference<a class="headerlink" href="#training-and-inference" title="永久链接至标题"></a></h1>
<div class="section" id="parameters">
<h2>Parameters<a class="headerlink" href="#parameters" title="永久链接至标题"></a></h2>
214 215 216
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.parameters.</code><code class="descname">Parameters</code></dt>
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
<dd><p><cite>Parameters</cite> manages all the learnable parameters in a neural network.
It stores parameters&#8217; information in an OrderedDict. The key is
the name of a parameter, and value is a parameter&#8217;s configuration(in
protobuf format), such as initialization mean and std, its size, whether it
is a static parameter, and so on.</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">参数:</th><td class="field-body"><ul class="first last simple">
<li><strong>__param_conf__</strong> (<em>OrderedDict</em>) &#8211; store the configurations of learnable parameters in
the network in an OrderedDict. Parameter is added one by one into the
dict by following their created order in the network: parameters of
the previous layers in a network are careted first. You can visit the
parameters from bottom to top by iterating over this dict.</li>
<li><strong>__gradient_machines__</strong> (<em>list</em>) &#8211; all of the parameters in a neural network are
appended to a PaddlePaddle gradient machine, which is used internally to
copy parameter values between C++ and Python end.</li>
<li><strong>__tmp_params__</strong> (<em>dict</em>) &#8211; a dict to store dummy parameters if no
__gradient_machines__ is appended to <cite>Parameters</cite>.</li>
</ul>
</td>
</tr>
</tbody>
</table>
242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339
<p>Basically usage is</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">data</span> <span class="o">=</span> <span class="n">paddle</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="o">...</span><span class="p">)</span>
<span class="o">...</span>
<span class="n">out</span> <span class="o">=</span> <span class="n">paddle</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="o">...</span><span class="p">)</span>

<span class="n">parameters</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">.</span><span class="n">parameters</span><span class="o">.</span><span class="n">create</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>

<span class="n">parameter_names</span> <span class="o">=</span> <span class="n">parameters</span><span class="o">.</span><span class="n">names</span><span class="p">()</span>
<span class="n">fc_mat</span> <span class="o">=</span> <span class="n">parameters</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">&#39;fc&#39;</span><span class="p">)</span>
<span class="k">print</span> <span class="n">fc_mat</span>
</pre></div>
</div>
<dl class="method">
<dt>
<code class="descname">keys</code><span class="sig-paren">(</span><span class="sig-paren">)</span></dt>
<dd><p>keys are the names of each parameter.</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">返回:</th><td class="field-body">list of parameter name</td>
</tr>
<tr class="field-even field"><th class="field-name">返回类型:</th><td class="field-body">list</td>
</tr>
</tbody>
</table>
</dd></dl>

<dl class="method">
<dt>
<code class="descname">names</code><span class="sig-paren">(</span><span class="sig-paren">)</span></dt>
<dd><p>names of each parameter.</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">返回:</th><td class="field-body">list of parameter name</td>
</tr>
<tr class="field-even field"><th class="field-name">返回类型:</th><td class="field-body">list</td>
</tr>
</tbody>
</table>
</dd></dl>

<dl class="method">
<dt>
<code class="descname">has_key</code><span class="sig-paren">(</span><em>key</em><span class="sig-paren">)</span></dt>
<dd><p>has_key return true if there are such parameter name == key</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">参数:</th><td class="field-body"><strong>key</strong> (<em>basestring</em>) &#8211; Parameter name</td>
</tr>
<tr class="field-even field"><th class="field-name">返回:</th><td class="field-body">True if contains such key</td>
</tr>
</tbody>
</table>
</dd></dl>

<dl class="method">
<dt>
<code class="descname">get_shape</code><span class="sig-paren">(</span><em>key</em><span class="sig-paren">)</span></dt>
<dd><p>get shape of the parameter.</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">参数:</th><td class="field-body"><strong>key</strong> (<em>basestring</em>) &#8211; parameter name</td>
</tr>
<tr class="field-even field"><th class="field-name">返回:</th><td class="field-body">parameter&#8217;s shape</td>
</tr>
<tr class="field-odd field"><th class="field-name">返回类型:</th><td class="field-body">tuple</td>
</tr>
</tbody>
</table>
</dd></dl>

<dl class="method">
<dt>
<code class="descname">get</code><span class="sig-paren">(</span><em>parameter_name</em><span class="sig-paren">)</span></dt>
<dd><p>Get parameter by parameter name.</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">Note:</th><td class="field-body">It will always copy the parameter from C++ side.</td>
</tr>
<tr class="field-even field"><th class="field-name">参数:</th><td class="field-body"><strong>parameter_name</strong> (<em>basestring</em>) &#8211; parameter name</td>
</tr>
<tr class="field-odd field"><th class="field-name">返回:</th><td class="field-body">The parameter matrix.</td>
</tr>
<tr class="field-even field"><th class="field-name">返回类型:</th><td class="field-body">np.ndarray</td>
</tr>
</tbody>
</table>
</dd></dl>

340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359
<dl class="method">
<dt>
<code class="descname">get_grad</code><span class="sig-paren">(</span><em>key</em><span class="sig-paren">)</span></dt>
<dd><p>Get grandient by parameter name.</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">Note:</th><td class="field-body">It will always copy the parameter from C++ side.</td>
</tr>
<tr class="field-even field"><th class="field-name">参数:</th><td class="field-body"><strong>key</strong> (<em>basestring</em>) &#8211; parameter name</td>
</tr>
<tr class="field-odd field"><th class="field-name">返回:</th><td class="field-body">The grandient matrix.</td>
</tr>
<tr class="field-even field"><th class="field-name">返回类型:</th><td class="field-body">np.ndarray</td>
</tr>
</tbody>
</table>
</dd></dl>

360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389
<dl class="method">
<dt>
<code class="descname">set</code><span class="sig-paren">(</span><em>parameter_name</em>, <em>value</em><span class="sig-paren">)</span></dt>
<dd><p>Set parameter by parameter name &amp; matrix.</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">参数:</th><td class="field-body"><ul class="first simple">
<li><strong>parameter_name</strong> (<em>basestring</em>) &#8211; parameter name</li>
<li><strong>value</strong> (<em>np.ndarray</em>) &#8211; parameter matrix</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">返回:</th><td class="field-body"><p class="first last">Nothing.</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

<dl class="method">
<dt>
<code class="descname">append_gradient_machine</code><span class="sig-paren">(</span><em>gradient_machine</em><span class="sig-paren">)</span></dt>
<dd><p>append gradient machine to parameters. This method is used internally in
Trainer.train.</p>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
390
<tr class="field-odd field"><th class="field-name">参数:</th><td class="field-body"><strong>gradient_machine</strong> (<em>api.GradientMachine</em>) &#8211; PaddlePaddle C++ GradientMachine object.</td>
391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437
</tr>
<tr class="field-even field"><th class="field-name">返回:</th><td class="field-body"></td>
</tr>
</tbody>
</table>
</dd></dl>

<dl class="method">
<dt>
<code class="descname">serialize</code><span class="sig-paren">(</span><em>name</em>, <em>f</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">参数:</th><td class="field-body"><ul class="first simple">
<li><strong>name</strong> &#8211; </li>
<li><strong>f</strong> (<em>file</em>) &#8211; </li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">返回:</th><td class="field-body"><p class="first last"></p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

<dl class="method">
<dt>
<code class="descname">deserialize</code><span class="sig-paren">(</span><em>name</em>, <em>f</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">参数:</th><td class="field-body"><ul class="first simple">
<li><strong>name</strong> &#8211; </li>
<li><strong>f</strong> (<em>file</em>) &#8211; </li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">返回:</th><td class="field-body"><p class="first last"></p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458
<dl class="method">
<dt>
<code class="descname">to_tar</code><span class="sig-paren">(</span><em>f</em><span class="sig-paren">)</span></dt>
<dd><p>Save parameters to a tar file.</p>
<dl class="docutils">
<dt>WARNING: You should use <cite>paddle.v2.trainer.SGD.save_parameter_to_tar(f)</cite></dt>
<dd>to save parameters most of the time. Otherwise, some settings such
as model average will not take effect.</dd>
</dl>
<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">参数:</th><td class="field-body"><strong>f</strong> (<em>file</em>) &#8211; </td>
</tr>
<tr class="field-even field"><th class="field-name">返回:</th><td class="field-body"></td>
</tr>
</tbody>
</table>
</dd></dl>

459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497
<dl class="staticmethod">
<dt>
<em class="property">static </em><code class="descname">from_tar</code><span class="sig-paren">(</span><em>f</em><span class="sig-paren">)</span></dt>
<dd><p>Create a <cite>Parameters</cite> object from the given file. And
the <cite>Parameters</cite> only contains the parameters in this
file. It is adapted the parameters are same in the
defined network and the given file. For example, it
can be used in the inference.</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">参数:</th><td class="field-body"><strong>f</strong> (<em>tar file</em>) &#8211; the initialized model file.</td>
</tr>
<tr class="field-even field"><th class="field-name">返回:</th><td class="field-body">A Parameters object.</td>
</tr>
<tr class="field-odd field"><th class="field-name">返回类型:</th><td class="field-body">Parameters.</td>
</tr>
</tbody>
</table>
</dd></dl>

<dl class="method">
<dt>
<code class="descname">init_from_tar</code><span class="sig-paren">(</span><em>f</em><span class="sig-paren">)</span></dt>
<dd><p>Different from <cite>from_tar</cite>, this interface can be used to
init partial network parameters from another saved model.</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">参数:</th><td class="field-body"><strong>f</strong> (<em>tar file</em>) &#8211; the initialized model file.</td>
</tr>
<tr class="field-even field"><th class="field-name">返回:</th><td class="field-body">Nothing.</td>
</tr>
</tbody>
</table>
</dd></dl>

498 499
</dd></dl>

500
</div>
501 502
<div class="section" id="trainer">
<h2>Trainer<a class="headerlink" href="#trainer" title="永久链接至标题"></a></h2>
503 504 505
<p>Module Trainer</p>
<dl class="class">
<dt>
506
<em class="property">class </em><code class="descclassname">paddle.v2.trainer.</code><code class="descname">SGD</code><span class="sig-paren">(</span><em>cost</em>, <em>parameters</em>, <em>update_equation</em>, <em>extra_layers=None</em>, <em>is_local=True</em>, <em>pserver_spec=None</em>, <em>use_etcd=True</em><span class="sig-paren">)</span></dt>
507 508 509 510 511 512 513 514 515 516
<dd><p>Simple SGD Trainer.
SGD Trainer combines data reader, network topolopy and update_equation together
to train/test a neural network.</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">参数:</th><td class="field-body"><ul class="first last simple">
<li><strong>cost</strong> (<em>paddle.v2.config_base.Layer</em>) &#8211; Target cost that neural network should be optimized.</li>
<li><strong>parameters</strong> (<em>paddle.v2.parameters.Parameters</em>) &#8211; The parameters dictionary.</li>
517
<li><strong>update_equation</strong> (<em>paddle.v2.optimizer.Optimizer</em>) &#8211; The optimizer object.</li>
518 519
<li><strong>extra_layers</strong> (<em>paddle.v2.config_base.Layer</em>) &#8211; Some layers in the neural network graph are not
in the path of cost layer.</li>
520 521 522 523 524 525 526
<li><strong>is_local</strong> (<em>bool</em>) &#8211; Whether trainning locally</li>
<li><strong>pserver_spec</strong> (<em>string</em>) &#8211; comma string for pserver location,
eg:127.10.0.10:3000,127.10.0.11:3000,
and this parameter is only used for fault
tolerant mode cluster training.</li>
<li><strong>use_etcd</strong> &#8211; Whether using etcd pserver.</li>
<li><strong>use_etcd</strong> &#8211; bool</li>
527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543
</ul>
</td>
</tr>
</tbody>
</table>
<dl class="method">
<dt>
<code class="descname">train</code><span class="sig-paren">(</span><em>reader</em>, <em>num_passes=1</em>, <em>event_handler=None</em>, <em>feeding=None</em><span class="sig-paren">)</span></dt>
<dd><p>Training method. Will train num_passes of input data.</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">参数:</th><td class="field-body"><ul class="first simple">
<li><strong>reader</strong> (<em>collections.Iterable</em>) &#8211; A reader that reads and yeilds data items. Usually we use a
batched reader to do mini-batch training.</li>
<li><strong>num_passes</strong> &#8211; The total train passes.</li>
544
<li><strong>event_handler</strong> (<em>(</em><em>BaseEvent</em><em>) </em><em>=&gt; None</em>) &#8211; Event handler. A method will be invoked when event
545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566
occurred.</li>
<li><strong>feeding</strong> (<em>dict|list</em>) &#8211; Feeding is a map of neural network input name and array
index that reader returns.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">返回:</th><td class="field-body"><p class="first last"></p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

<dl class="method">
<dt>
<code class="descname">test</code><span class="sig-paren">(</span><em>reader</em>, <em>feeding=None</em><span class="sig-paren">)</span></dt>
<dd><p>Testing method. Will test input data.</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">参数:</th><td class="field-body"><ul class="first simple">
567 568
<li><strong>reader</strong> (<em>collections.Iterable</em>) &#8211; A batch reader that reads and yeilds data items,
it should be a paddle.v2.batch.</li>
569 570 571 572 573 574 575 576 577 578 579 580 581 582
<li><strong>feeding</strong> (<em>dict</em>) &#8211; Feeding is a map of neural network input name and array
index that reader returns.</li>
</ul>
</td>
</tr>
<tr class="field-even field"><th class="field-name">返回:</th><td class="field-body"><p class="first last"></p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

</dd></dl>

583
</div>
584 585
<div class="section" id="event">
<h2>Event<a class="headerlink" href="#event" title="永久链接至标题"></a></h2>
586
<p>Testing and training events.</p>
587 588
<p>There are:</p>
<ul class="simple">
589
<li>TestResult</li>
590 591 592 593 594
<li>BeginIteration</li>
<li>EndIteration</li>
<li>BeginPass</li>
<li>EndPass</li>
</ul>
595 596 597 598 599 600 601 602 603 604 605 606 607 608
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.event.</code><code class="descname">TestResult</code><span class="sig-paren">(</span><em>evaluator</em>, <em>cost</em><span class="sig-paren">)</span></dt>
<dd><p>Result that trainer.test return.</p>
</dd></dl>

<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.event.</code><code class="descname">BeginPass</code><span class="sig-paren">(</span><em>pass_id</em><span class="sig-paren">)</span></dt>
<dd><p>Event On One Pass Training Start.</p>
</dd></dl>

<dl class="class">
<dt>
609 610 611 612
<em class="property">class </em><code class="descclassname">paddle.v2.event.</code><code class="descname">EndPass</code><span class="sig-paren">(</span><em>pass_id</em>, <em>evaluator</em>, <em>gm</em><span class="sig-paren">)</span></dt>
<dd><p>Event On One Pass Training Complete.
To get the output of a specific layer, add &#8220;event.gm.getLayerOutputs(&#8216;predict_layer&#8217;)&#8221;
in your event_handler call back</p>
613 614 615 616 617 618 619 620
</dd></dl>

<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.event.</code><code class="descname">BeginIteration</code><span class="sig-paren">(</span><em>pass_id</em>, <em>batch_id</em><span class="sig-paren">)</span></dt>
<dd><p>Event On One Batch Training Start.</p>
</dd></dl>

621 622 623 624 625 626
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.event.</code><code class="descname">EndForwardBackward</code><span class="sig-paren">(</span><em>pass_id</em>, <em>batch_id</em>, <em>gm</em><span class="sig-paren">)</span></dt>
<dd><p>Event On One Batch ForwardBackward Complete.</p>
</dd></dl>

627 628
<dl class="class">
<dt>
629 630 631 632
<em class="property">class </em><code class="descclassname">paddle.v2.event.</code><code class="descname">EndIteration</code><span class="sig-paren">(</span><em>pass_id</em>, <em>batch_id</em>, <em>cost</em>, <em>evaluator</em>, <em>gm</em><span class="sig-paren">)</span></dt>
<dd><p>Event On One Batch Training Complete.
To get the output of a specific layer, add &#8220;event.gm.getLayerOutputs(&#8216;predict_layer&#8217;)&#8221;
in your event_handler call back</p>
633 634
</dd></dl>

635 636 637 638
</div>
<div class="section" id="inference">
<h2>Inference<a class="headerlink" href="#inference" title="永久链接至标题"></a></h2>
<dl class="function">
639 640
<dt>
<code class="descclassname">paddle.v2.</code><code class="descname">infer</code><span class="sig-paren">(</span><em>output_layer</em>, <em>parameters</em>, <em>input</em>, <em>feeding=None</em>, <em>field='value'</em><span class="sig-paren">)</span></dt>
641 642
<dd><p>Infer a neural network by given neural network output and parameters.  The
user should pass either a batch of input data or reader method.</p>
643
<p>Example usage for sinlge output_layer:</p>
644
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">result</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">.</span><span class="n">infer</span><span class="p">(</span><span class="n">output_layer</span><span class="o">=</span><span class="n">prediction</span><span class="p">,</span>
645 646
                      <span class="n">parameters</span><span class="o">=</span><span class="n">parameters</span><span class="p">,</span>
                      <span class="nb">input</span><span class="o">=</span><span class="n">SomeData</span><span class="p">)</span>
647 648 649
<span class="k">print</span> <span class="n">result</span>
</pre></div>
</div>
650 651 652 653 654 655 656 657
<p>Example usage for multiple outout_layers and fields:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">result</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">.</span><span class="n">infer</span><span class="p">(</span><span class="n">output_layer</span><span class="o">=</span><span class="p">[</span><span class="n">prediction1</span><span class="p">,</span> <span class="n">prediction2</span><span class="p">],</span>
                      <span class="n">parameters</span><span class="o">=</span><span class="n">parameters</span><span class="p">,</span>
                      <span class="nb">input</span><span class="o">=</span><span class="n">SomeData</span><span class="p">,</span>
                      <span class="n">field</span><span class="o">=</span><span class="p">[</span><span class="nb">id</span><span class="p">,</span> <span class="n">value</span><span class="p">]])</span>
<span class="k">print</span> <span class="n">result</span>
</pre></div>
</div>
658 659 660 661 662
<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">参数:</th><td class="field-body"><ul class="first simple">
663 664
<li><strong>output_layer</strong> (<em>paddle.v2.config_base.Layer</em><em> or </em><em>a list of
paddle.v2.config_base.Layer</em>) &#8211; output of the neural network that would be inferred</li>
665
<li><strong>parameters</strong> (<em>paddle.v2.parameters.Parameters</em>) &#8211; parameters of the neural network.</li>
666 667 668 669
<li><strong>input</strong> (<em>collections.Iterable</em>) &#8211; input data batch. Should be a python iterable object, and each
element is the data batch.</li>
<li><strong>feeding</strong> &#8211; Reader dictionary. Default could generate from input
value.</li>
670 671 672 673 674
<li><strong>field</strong> (<em>str</em>) &#8211; The prediction field. It should in [<cite>value</cite>, <cite>id</cite>, <cite>prob</cite>].
<cite>value</cite> and <cite>prob</cite> mean return the prediction probabilities,
<cite>id</cite> means return the prediction labels. Default is <cite>value</cite>.
Note that <cite>prob</cite> only used when output_layer is beam_search
or max_id.</li>
675 676 677
</ul>
</td>
</tr>
678 679 680
<tr class="field-even field"><th class="field-name">返回:</th><td class="field-body"><p class="first">The prediction result. If there are multiple outout_layers and fields,
the return order is outout_layer1.field1, outout_layer2.field1, ...,
outout_layer1.field2, outout_layer2.field2 ...</p>
681 682 683 684 685 686 687 688 689
</td>
</tr>
<tr class="field-odd field"><th class="field-name">返回类型:</th><td class="field-body"><p class="first last">numpy.ndarray</p>
</td>
</tr>
</tbody>
</table>
</dd></dl>

690 691 692 693 694 695 696 697 698 699 700 701 702
</div>
</div>


           </div>
          </div>
          <footer>
  
    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
      
        <a href="../../faq/index_cn.html" class="btn btn-neutral float-right" title="FAQ" accesskey="n">Next <span class="fa fa-arrow-circle-right"></span></a>
      
      
703
        <a href="data/dataset.html" class="btn btn-neutral" title="Dataset" accesskey="p"><span class="fa fa-arrow-circle-left"></span> Previous</a>
704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736
      
    </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',
737 738
            HAS_SOURCE:  true,
            SOURCELINK_SUFFIX: ".txt",
739 740 741 742 743 744
        };
    </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="../../_static/translations.js"></script>
745
      <script type="text/javascript" src="https://cdn.bootcss.com/mathjax/2.7.0/MathJax.js"></script>
746 747 748 749 750 751 752 753 754 755 756 757 758
       
  

  
  
    <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>
759
</html>