Skip to content

P
Paddle

PaddlePaddle / Paddle
大约 1 年 前同步成功

通知 2298
Star 20931
Fork 5422
P
Paddle
切换分支/标签
查找文件 普通视图历史永久链接Permalink
detail_introduction_cn.html 33.2 KB
Newer Older
T
Deploy to GitHub Pages: 2c98becba4fb48311b80c53c1793fae47e7fe975  
Travis CI 已提交
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 


<!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>细节描述 &mdash; PaddlePaddle  文档</title>
  

  
  

  

  
  
    

  

  
  
    <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="设置命令行参数" href="index_cn.html"/>
T
Deploy to GitHub Pages: 1ff47b284c565d030b198705d5f18b4bd4ce53e5  
Travis CI 已提交
37 
        <link rel="next" title="PaddlePaddle分布式训练" href="../cluster/cluster_train_cn.html"/>
T
Deploy to GitHub Pages: 2c98becba4fb48311b80c53c1793fae47e7fe975  
Travis CI 已提交
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 
        <link rel="prev" title="参数概述" href="arguments_cn.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">
T
Deploy to GitHub Pages: 1ff47b284c565d030b198705d5f18b4bd4ce53e5  
Travis CI 已提交
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>
T
Deploy to GitHub Pages: 2c98becba4fb48311b80c53c1793fae47e7fe975  
Travis CI 已提交
69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 
        <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_cn.html">新手入门</a></li>
<li class="toctree-l1 current"><a class="reference internal" href="../../index_cn.html">进阶指南</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../../api/index_cn.html">API</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
T
Deploy to GitHub Pages: 1ff47b284c565d030b198705d5f18b4bd4ce53e5  
Travis CI 已提交
91 
<li class="toctree-l1"><a class="reference internal" href="../../../mobile/index_cn.html">MOBILE</a></li>
T
Deploy to GitHub Pages: 2c98becba4fb48311b80c53c1793fae47e7fe975  
Travis CI 已提交
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>
T
Deploy to GitHub Pages: 1ff47b284c565d030b198705d5f18b4bd4ce53e5  
Travis CI 已提交
115 116 117 118 
<li class="toctree-l3"><a class="reference internal" href="../../../getstarted/build_and_install/pip_install_cn.html">使用pip安装</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../../getstarted/build_and_install/docker_install_cn.html">使用Docker安装运行</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../dev/build_cn.html">用Docker编译和测试PaddlePaddle</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../../getstarted/build_and_install/build_from_source_cn.html">从源码编译</a></li>
T
Deploy to GitHub Pages: 2c98becba4fb48311b80c53c1793fae47e7fe975  
Travis CI 已提交
119 120 121 122 123 124 125 126 127 128 129 130 
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../../../getstarted/concepts/use_concepts_cn.html">基本使用概念</a></li>
</ul>
</li>
<li class="toctree-l1 current"><a class="reference internal" href="../../index_cn.html">进阶指南</a><ul class="current">
<li class="toctree-l2 current"><a class="reference internal" href="index_cn.html">设置命令行参数</a><ul class="current">
<li class="toctree-l3"><a class="reference internal" href="use_case_cn.html">使用案例</a></li>
<li class="toctree-l3"><a class="reference internal" href="arguments_cn.html">参数概述</a></li>
<li class="toctree-l3 current"><a class="current reference internal" href="#">细节描述</a></li>
</ul>
</li>
T
Deploy to GitHub Pages: 1ff47b284c565d030b198705d5f18b4bd4ce53e5  
Travis CI 已提交
131 
<li class="toctree-l2"><a class="reference internal" href="../cluster/cluster_train_cn.html">PaddlePaddle分布式训练</a></li>
T
Deploy to GitHub Pages: 2c98becba4fb48311b80c53c1793fae47e7fe975  
Travis CI 已提交
132 133 134 135 
<li class="toctree-l2"><a class="reference internal" href="../k8s/k8s_basis_cn.html">Kubernetes 简介</a></li>
<li class="toctree-l2"><a class="reference internal" href="../k8s/k8s_cn.html">Kubernetes单机训练</a></li>
<li class="toctree-l2"><a class="reference internal" href="../k8s/k8s_distributed_cn.html">Kubernetes分布式训练</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../dev/contribute_to_paddle_cn.html">如何贡献代码</a></li>
T
Deploy to GitHub Pages: 1ff47b284c565d030b198705d5f18b4bd4ce53e5  
Travis CI 已提交
136 
<li class="toctree-l2"><a class="reference internal" href="../../dev/write_docs_cn.html">如何贡献/修改文档</a></li>
T
Deploy to GitHub Pages: 2c98becba4fb48311b80c53c1793fae47e7fe975  
Travis CI 已提交
137 
<li class="toctree-l2"><a class="reference internal" href="../../deep_model/rnn/index_cn.html">RNN相关模型</a><ul>
T
Deploy to GitHub Pages: 1ff47b284c565d030b198705d5f18b4bd4ce53e5  
Travis CI 已提交
138 
<li class="toctree-l3"><a class="reference internal" href="../../deep_model/rnn/rnn_config_cn.html">RNN配置</a></li>
T
Deploy to GitHub Pages: 2c98becba4fb48311b80c53c1793fae47e7fe975  
Travis CI 已提交
139 140 141 142 143 144 145 146 147 148 149 150 
<li class="toctree-l3"><a class="reference internal" href="../../deep_model/rnn/recurrent_group_cn.html">Recurrent Group教程</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../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="../../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="../../optimization/gpu_profiling_cn.html">GPU性能分析与调优</a></li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../../api/index_cn.html">API</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../../../api/v2/model_configs.html">模型配置</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../../../api/v2/config/activation.html">Activation</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../../api/v2/config/layer.html">Layers</a></li>
T
Deploy to GitHub Pages: 1ff47b284c565d030b198705d5f18b4bd4ce53e5  
Travis CI 已提交
151 
<li class="toctree-l3"><a class="reference internal" href="../../../api/v2/config/evaluators.html">Evaluators</a></li>
T
Deploy to GitHub Pages: 2c98becba4fb48311b80c53c1793fae47e7fe975  
Travis CI 已提交
152 153 154 155 156 157 
<li class="toctree-l3"><a class="reference internal" href="../../../api/v2/config/optimizer.html">Optimizer</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../../api/v2/config/pooling.html">Pooling</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../../api/v2/config/networks.html">Networks</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../../api/v2/config/attr.html">Parameter Attribute</a></li>
</ul>
</li>
T
Deploy to GitHub Pages: 1ff47b284c565d030b198705d5f18b4bd4ce53e5  
Travis CI 已提交
158 159 160 161 162 163 
<li class="toctree-l2"><a class="reference internal" href="../../../api/v2/data.html">数据访问</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../../../api/v2/data/data_reader.html">Data Reader Interface</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../../api/v2/data/image.html">Image Interface</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../../api/v2/data/dataset.html">Dataset</a></li>
</ul>
</li>
T
Deploy to GitHub Pages: 2c98becba4fb48311b80c53c1793fae47e7fe975  
Travis CI 已提交
164 165 166 
<li class="toctree-l2"><a class="reference internal" href="../../../api/v2/run_logic.html">训练与应用</a></li>
</ul>
</li>
T
Deploy to GitHub Pages: 1ff47b284c565d030b198705d5f18b4bd4ce53e5  
Travis CI 已提交
167 168 169 170 171 172 173 174 175 176 177 178 179 180 
<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>
<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">Android平台编译指南</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../../mobile/cross_compiling_for_ios_cn.html">iOS平台编译指南</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../../mobile/cross_compiling_for_raspberry_cn.html">Raspberry Pi平台编译指南</a></li>
</ul>
</li>
T
Deploy to GitHub Pages: 2c98becba4fb48311b80c53c1793fae47e7fe975  
Travis CI 已提交
181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 
</ul>

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

      

 







<div role="navigation" aria-label="breadcrumbs navigation">
  <ul class="wy-breadcrumbs">
      
        <li><a href="../../index_cn.html">进阶指南</a> > </li>
      
        <li><a href="index_cn.html">设置命令行参数</a> > </li>
      
    <li>细节描述</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="">
<span id="id1"></span><h1>细节描述<a class="headerlink" href="#" title="永久链接至标题"></a></h1>
<div class="section" id="">
<span id="id2"></span><h2>通用<a class="headerlink" href="#" title="永久链接至标题"></a></h2>
<ul class="simple">
<li><code class="docutils literal"><span class="pre">--job</span></code><ul>
<li>工作模式，包括: <strong>train, test, checkgrad</strong>，其中checkgrad主要为开发者使用，使用者不需要关心。</li>
<li>类型: string (默认: train)</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--config</span></code><ul>
<li>用于指定网络配置文件。</li>
<li>类型: string (默认: null).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--use_gpu</span></code><ul>
<li>训练过程是否使用GPU，设置为true使用GPU模式，否则使用CPU模式。</li>
<li>类型: bool (默认: 1).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--local</span></code>
&nbsp;- 训练过程是否为本地模式，设置为true使用本地训练或者使用集群上的一个节点，否则使用多机训练。<ul>
<li>类型: bool (默认: 1).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--trainer_count</span></code><ul>
<li>指定一台机器上使用的线程数。例如，trainer_count = 4, 意思是在GPU模式下使用4个GPU，或者在CPU模式下使用4个线程。每个线程（或GPU）分配到当前数据块样本数的四分之一。也就是说，如果在训练配置中设置batch_size为512，每个线程分配到128个样本用于训练。</li>
<li>类型: int32 (默认: 1).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--num_passes</span></code><ul>
<li>当模式为<code class="docutils literal"><span class="pre">--job=train</span></code>时, 该参数的意思是训练num_passes轮。每轮会将数据集中的所有训练样本使用一次。当模式为<code class="docutils literal"><span class="pre">--job=test</span></code>时，意思是使用第test_pass个模型到第 num_passes-1 个模型测试数据。</li>
<li>类型: int32 (默认: 100).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--config_args</span></code><ul>
<li>传递给配置文件的参数。格式: key1=value1,key2=value2.</li>
<li>类型: string (默认: null).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--version</span></code><ul>
<li>是否打印版本信息。</li>
<li>类型: bool (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--show_layer_stat</span></code><ul>
<li>是否显示<strong>每个批次数据</strong>中每层的数值统计.</li>
<li>类型: bool (默认: 0).</li>
</ul>
</li>
</ul>
</div>
<div class="section" id="">
<span id="id3"></span><h2>训练<a class="headerlink" href="#" title="永久链接至标题"></a></h2>
<ul class="simple">
<li><code class="docutils literal"><span class="pre">--log_period</span></code><ul>
<li>每log_period个批次打印日志进度.</li>
<li>类型: int32 (默认: 100).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--dot_period</span></code><ul>
<li>每dot_period个批次输出符号&#8217;.&#8217;.</li>
<li>类型: int32 (默认: 1).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--saving_period</span></code><ul>
<li>每saving_period轮保存训练参数.</li>
<li>类型: int32 (默认: 1).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--save_dir</span></code><ul>
<li>保存模型参数的目录，需要明确指定，但不需要提前创建。</li>
<li>类型: string (默认: null).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--start_pass</span></code><ul>
<li>从start_pass轮开始训练，会加载上一轮的参数。</li>
<li>类型: int32 (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--show_parameter_stats_period</span></code><ul>
<li>在训练过程中每show_parameter_stats_period个批次输出参数统计。默认不显示。</li>
<li>类型: int32 (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--save_only_one</span></code><ul>
<li>只保存最后一轮的参数，而之前的参数将会被删除。</li>
<li>类型: bool (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--load_missing_parameter_strategy</span></code><ul>
<li>当模型参数不存在时，指定加载的方式。目前支持fail/rand/zero三种操作.<ul>
<li><code class="docutils literal"><span class="pre">fail</span></code>: 程序直接退出.</li>
<li><code class="docutils literal"><span class="pre">rand</span></code>: 根据网络配置中的<strong>initial_strategy</strong>采用均匀分布或者高斯分布初始化。均匀分布的范围是: <strong>[mean - std, mean + std]</strong>, 其中mean和std是训练配置中的参数.</li>
<li><code class="docutils literal"><span class="pre">zero</span></code>: 所有参数置为零.</li>
</ul>
</li>
<li>类型: string (默认: fail).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--init_model_path</span></code><ul>
<li>初始化模型的路径。如果设置该参数，start_pass将不起作用。同样也可以在测试模式中指定模型路径。</li>
<li>类型: string (默认: null).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--saving_period_by_batches</span></code><ul>
<li>在一轮中每saving_period_by_batches个批次保存一次参数。</li>
<li>类型: int32 (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--log_error_clipping</span></code><ul>
<li>当在网络层配置中设置<strong>error_clipping_threshold</strong>时，该参数指示是否打印错误截断日志。如果为true，<strong>每批次</strong>的反向传播将会打印日志信息。该截断会影响<strong>输出的梯度</strong>.</li>
<li>类型: bool (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--log_clipping</span></code><ul>
<li>当在训练配置中设置<strong>gradient_clipping_threshold</strong>时，该参数指示是否打印日志截断信息。该截断会影响<strong>权重更新的梯度</strong>.</li>
<li>类型: bool (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--use_old_updater</span></code><ul>
<li>是否使用旧的RemoteParameterUpdater。 默认使用ConcurrentRemoteParameterUpdater，主要为开发者使用，使用者通常无需关心.</li>
<li>类型: bool (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--enable_grad_share</span></code><ul>
<li>启用梯度参数的阈值，在多CPU训练时共享该参数.</li>
<li>类型: int32 (默认: 100 * 1024 * 1024).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--grad_share_block_num</span></code><ul>
<li>梯度参数的分块数目，在多CPU训练时共享该参数.</li>
<li>类型: int32 (默认: 64).</li>
</ul>
</li>
</ul>
</div>
<div class="section" id="">
<span id="id4"></span><h2>测试<a class="headerlink" href="#" title="永久链接至标题"></a></h2>
<ul class="simple">
<li><code class="docutils literal"><span class="pre">--test_pass</span></code><ul>
<li>加载test_pass轮的模型用于测试.</li>
<li>类型: int32 (默认: -1).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--test_period</span></code><ul>
<li>如果为0，每轮结束时对所有测试数据进行测试；如果不为0，每test_period个批次对所有测试数据进行测试.</li>
<li>类型: int32 (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--test_wait</span></code><ul>
<li>指示当指定轮的测试模型不存在时，是否需要等待该轮模型参数。如果在训练期间同时发起另外一个进程进行测试，可以使用该参数.</li>
<li>类型: bool (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--model_list</span></code><ul>
<li>测试时指定的存储模型列表的文件.</li>
<li>类型: string (默认: &#8220;&#8221;, null).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--predict_output_dir</span></code><ul>
<li>保存网络层输出结果的目录。该参数在网络配置的Outputs()中指定，默认为null，意思是不保存结果。在测试阶段，如果你想要保存某些层的特征图，请指定该目录。需要注意的是，网络层的输出是经过激活函数之后的值.</li>
<li>类型: string (默认: &#8220;&#8221;, null).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--average_test_period</span></code><ul>
<li>使用<code class="docutils literal"><span class="pre">average_test_period</span></code>个批次的参数平均值进行测试。该参数必须能被FLAGS_log_period整除，默认为0，意思是不使用平均参数执行测试.</li>
<li>类型: int32 (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--distribute_test</span></code><ul>
<li>在分布式环境中测试，将多台机器的测试结果合并.</li>
<li>类型: bool (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--predict_file</span></code><ul>
<li>保存预测结果的文件名。该参数默认为null，意思是不保存结果。目前该参数仅用于AucValidationLayer和PnpairValidationLayer层，每轮都会保存预测结果.</li>
<li>类型: string (默认: &#8220;&#8221;, null).</li>
</ul>
</li>
</ul>
</div>
<div class="section" id="gpu">
<span id="gpu"></span><h2>GPU<a class="headerlink" href="#gpu" title="永久链接至标题"></a></h2>
<ul class="simple">
<li><code class="docutils literal"><span class="pre">--gpu_id</span></code><ul>
<li>指示使用哪个GPU核.</li>
<li>类型: int32 (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--allow_only_one_model_on_one_gpu</span></code><ul>
<li>如果为true，一个GPU设备上不允许配置多个模型.</li>
<li>类型: bool (默认: 1).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--parallel_nn</span></code><ul>
<li>指示是否使用多线程来计算一个神经网络。如果为false，设置gpu_id指定使用哪个GPU核（训练配置中的设备属性将会无效）。如果为true，GPU核在训练配置中指定（gpu_id无效）.</li>
<li>类型: bool (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--cudnn_dir</span></code><ul>
<li>选择路径来动态加载NVIDIA CuDNN库，例如，/usr/local/cuda/lib64. [默认]: LD_LIBRARY_PATH</li>
<li>类型: string (默认: &#8220;&#8221;, null)</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--cuda_dir</span></code><ul>
<li>选择路径来动态加载NVIDIA CUDA库，例如，/usr/local/cuda/lib64. [默认]: LD_LIBRARY_PATH</li>
<li>类型: string (默认: &#8220;&#8221;, null)</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--cudnn_conv_workspace_limit_in_mb</span></code><ul>
<li>指定cuDNN的最大工作空间容限，单位是MB，默认为4096MB=4GB.</li>
<li>类型: int32 (默认: 4096MB=4GB)</li>
</ul>
</li>
</ul>
</div>
<div class="section" id="nlp-rnn-lstm-gru">
<span id="nlp-rnn-lstm-gru"></span><h2>自然语言处理(NLP): RNN/LSTM/GRU<a class="headerlink" href="#nlp-rnn-lstm-gru" title="永久链接至标题"></a></h2>
<ul class="simple">
<li><code class="docutils literal"><span class="pre">--rnn_use_batch</span></code><ul>
<li>指示在简单的RecurrentLayer层的计算中是否使用批处理方法.</li>
<li>类型: bool (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--prev_batch_state</span></code><ul>
<li>标识是否为连续的batch计算.</li>
<li>类型: bool (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--beam_size</span></code><ul>
<li>集束搜索使用广度优先搜索的方式构建查找树。在树的每一层上，都会产生当前层状态的所有继承结果，按启发式损失的大小递增排序。然而，每层上只能保存固定数目个最好的状态，该数目是提前定义好的，称之为集束大小.</li>
<li>类型: int32 (默认: 1).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--diy_beam_search_prob_so</span></code>
&nbsp;- 用户可以自定义beam search的方法，编译成动态库，供PaddlePaddle加载。 该参数用于指定动态库路径.<ul>
<li>类型: string (默认: &#8220;&#8221;, null).</li>
</ul>
</li>
</ul>
</div>
<div class="section" id="dataprovider">
<span id="dataprovider"></span><h2>数据支持(DataProvider)<a class="headerlink" href="#dataprovider" title="永久链接至标题"></a></h2>
<ul class="simple">
<li><code class="docutils literal"><span class="pre">--memory_threshold_on_load_data</span></code><ul>
<li>内存容限阈值，当超过该阈值时，停止加载数据.</li>
<li>类型: double (默认: 1.0).</li>
</ul>
</li>
</ul>
</div>
<div class="section" id="">
<span id="id5"></span><h2>单元测试<a class="headerlink" href="#" title="永久链接至标题"></a></h2>
<ul class="simple">
<li><code class="docutils literal"><span class="pre">--checkgrad_eps</span></code><ul>
<li>使用checkgrad模式时的参数变化大小.</li>
<li>类型: double (默认: 1e-05).</li>
</ul>
</li>
</ul>
</div>
<div class="section" id="">
<span id="id6"></span><h2>参数服务器和分布式通信<a class="headerlink" href="#" title="永久链接至标题"></a></h2>
<ul class="simple">
<li><code class="docutils literal"><span class="pre">--start_pserver</span></code><ul>
<li>指示是否开启参数服务器(parameter server).</li>
<li>类型: bool (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--pservers</span></code><ul>
<li>参数服务器的IP地址，以逗号间隔.</li>
<li>类型: string (默认: &#8220;127.0.0.1&#8221;).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--port</span></code><ul>
<li>参数服务器的监听端口.</li>
<li>类型: int32 (默认: 20134).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--ports_num</span></code><ul>
<li>发送参数的端口号，根据默认端口号递增.</li>
<li>类型: int32 (默认: 1).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--trainer_id</span></code>
&nbsp;- 在分布式训练中，每个训练节点必须指定一个唯一的id号，从0到num_trainers-1。0号训练节点是主训练节点。使用者无需关心这个参数.<ul>
<li>类型: int32 (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--num_gradient_servers</span></code><ul>
<li>梯度服务器的数量，该参数在集群提交环境中自动设置.</li>
<li>类型: int32 (默认: 1).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--small_messages</span></code><ul>
<li>如果消息数据太小，建议将该参数设为true，启动快速应答，无延迟.</li>
<li>类型: bool (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--sock_send_buf_size</span></code><ul>
<li>限制套接字发送缓冲区的大小。如果仔细设置的话，可以有效减小网络的阻塞.</li>
<li>类型: int32 (默认: 1024 * 1024 * 40).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--sock_recv_buf_size</span></code><ul>
<li>限制套接字接收缓冲区的大小.</li>
<li>类型: int32 (默认: 1024 * 1024 * 40).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--parameter_block_size</span></code><ul>
<li>参数服务器的参数分块大小。如果未设置，将会自动计算出一个合适的值.</li>
<li>类型: int32 (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--parameter_block_size_for_sparse</span></code><ul>
<li>参数服务器稀疏更新的参数分块大小。如果未设置，将会自动计算出一个合适的值.</li>
<li>类型: int32 (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--log_period_server</span></code><ul>
<li>在参数服务器终端每log_period_server个批次打印日志进度.</li>
<li>类型: int32 (默认: 500).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--loadsave_parameters_in_pserver</span></code><ul>
<li>在参数服务器上加载和保存参数，只有当设置了sparse_remote_update参数时才有效.</li>
<li>类型: bool (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--pserver_num_threads</span></code><ul>
<li>同步执行操作的线程数.</li>
<li>类型: bool (默认: 1).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--ports_num_for_sparse</span></code><ul>
<li>发送参数的端口号，根据默认值递增(port + ports_num)，用于稀疏训练中.</li>
<li>类型: int32 (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--nics</span></code><ul>
<li>参数服务器的网络设备名称，已经在集群提交环境中完成设置.</li>
<li>类型: string (默认: &#8220;xgbe0,xgbe1&#8221;).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--rdma_tcp</span></code><ul>
<li>使用rdma还是tcp传输协议，该参数已经在集群提交环境中完成设置.</li>
<li>类型: string (默认: &#8220;tcp&#8221;).</li>
</ul>
</li>
</ul>
</div>
<div class="section" id="async-sgd">
<span id="async-sgd"></span><h2>异步随机梯度下降(Async SGD)<a class="headerlink" href="#async-sgd" title="永久链接至标题"></a></h2>
<ul class="simple">
<li><code class="docutils literal"><span class="pre">--async_count</span></code><ul>
<li>定义异步训练的长度，如果为0，则使用同步训练.</li>
<li>类型: int32 (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--async_lagged_ratio_min</span></code><ul>
<li>控制<code class="docutils literal"><span class="pre">config_.async_lagged_grad_discard_ratio()</span></code>的最小值.</li>
<li>类型: double (默认: 1.0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--async_lagged_ratio_default</span></code><ul>
<li>如果在网络配置中未设置async_lagged_grad_discard_ratio，则使用该参数作为默认值.</li>
<li>类型: double (默认: 1.5).</li>
</ul>
</li>
</ul>
</div>
<div class="section" id="performance-tuning">
<span id="performance-tuning"></span><h2>性能调优(Performance Tuning)<a class="headerlink" href="#performance-tuning" title="永久链接至标题"></a></h2>
<ul class="simple">
<li><code class="docutils literal"><span class="pre">--log_barrier_abstract</span></code><ul>
<li>如果为true，则显示阻隔性能的摘要信息.</li>
<li>类型: bool (默认: 1).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--log_barrier_show_log</span></code><ul>
<li>如果为true，则总会显示阻隔摘要信息，即使间隔很小.</li>
<li>类型: bool (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--log_barrier_lowest_nodes</span></code><ul>
<li>最少显示多少个节点.</li>
<li>类型: int32 (默认: 5).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--check_sparse_distribution_in_pserver</span></code><ul>
<li>指示是否检查所有参数服务器上的稀疏参数的分布是均匀的.</li>
<li>类型: bool (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--show_check_sparse_distribution_log</span></code><ul>
<li>指示是否显示参数服务器上的稀疏参数分布的日志细节.</li>
<li>类型: bool (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--check_sparse_distribution_batches</span></code><ul>
<li>每运行多少个批次执行一次稀疏参数分布的检查.</li>
<li>类型: int32 (默认: 100).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--check_sparse_distribution_ratio</span></code><ul>
<li>如果检查到分配在不同参数服务器上的参数的分布不均匀次数大于check_sparse_distribution_ratio *  check_sparse_distribution_batches次，程序停止.</li>
<li>类型: double (默认: 0.6).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--check_sparse_distribution_unbalance_degree</span></code><ul>
<li>不同参数服务器上数据大小的最大值与最小值的比率.</li>
<li>类型: double (默认: 2).</li>
</ul>
</li>
</ul>
</div>
<div class="section" id="">
<span id="id7"></span><h2>矩阵/向量/随机数<a class="headerlink" href="#" title="永久链接至标题"></a></h2>
<ul class="simple">
<li><code class="docutils literal"><span class="pre">--enable_parallel_vector</span></code><ul>
<li>启动并行向量的阈值.</li>
<li>类型: int32 (默认: 0).</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--seed</span></code><ul>
<li>随机数的种子。srand(time)的为0.</li>
<li>类型: int32 (默认: 1)</li>
</ul>
</li>
<li><code class="docutils literal"><span class="pre">--thread_local_rand_use_global_seed</span></code><ul>
<li>是否将全局种子应用于本地线程的随机数.</li>
<li>类型: bool (默认: 0).</li>
</ul>
</li>
</ul>
</div>
</div>


           </div>
          </div>
          <footer>
  
    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
T
Deploy to GitHub Pages: 1ff47b284c565d030b198705d5f18b4bd4ce53e5  
Travis CI 已提交
660 
        <a href="../cluster/cluster_train_cn.html" class="btn btn-neutral float-right" title="PaddlePaddle分布式训练" accesskey="n">Next <span class="fa fa-arrow-circle-right"></span></a>
T
Deploy to GitHub Pages: 2c98becba4fb48311b80c53c1793fae47e7fe975  
Travis CI 已提交
661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 
      
      
        <a href="arguments_cn.html" class="btn btn-neutral" title="参数概述" 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="../../../_static/translations.js"></script>
      <script type="text/javascript" src="https://cdn.bootcss.com/mathjax/2.7.0/MathJax.js"></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>