Deploy to GitHub Pages: 90b9cba7

14768828 · Travis CI · 4b97303a · 14768828 · 14768828 · 14768828
10 changed file
--- a/doc/_sources/ui/data_provider/pydataprovider2.txt
+++ b/doc/_sources/ui/data_provider/pydataprovider2.txt
@@ -24,7 +24,7 @@ A small part of the original data as an example is shown as below:
 .. literalinclude:: ../../../doc_cn/ui/data_provider/mnist_train.txt
-Each line of the data contains two parts, separated by ';'. The first part is
+Each line of the data contains two parts, separated by :code:`;`. The first part is
 label of an image. The second part contains 28x28 pixel float values.
 Just write path of the above data into train.list. It looks like this:
@@ -74,7 +74,20 @@ you can take this as an example.
 .. literalinclude:: ../../../doc_cn/ui/data_provider/mnist_config.py
-Here we specify training data by 'train.list', and no testing data is specified.
+Here we specify training data by :code:`train.list`, and no testing data is specified.
+The method which actually provide data is :code:`process`.
+User also can use another style to provide data, which defines the
+:code:`data_layer`'s name explicitly when `yield`. For example,
+the :code:`dataprovider` is shown as below.
+.. literalinclude:: ../../../doc_cn/ui/data_provider/mnist_provider.dict.py
+   :linenos:
+If user did't give the :code:`data_layer`'s name, PaddlePaddle will use
+the order of :code:`data_layer` definition roughly to determine which feature to
+which :code:`data_layer`. This order may be not correct, so TO DEFINE THE
+:code:`data_layer`'s NAMES EXPLICITLY IS THE RECOMMANDED WAY TO PROVIDER DATA.
 Now, this simple example of using PyDataProvider is finished.
 The only thing that the user should know is how to generte **one sample** from
@@ -93,7 +106,7 @@ DataProvider for the sequential model
 -------------------------------------
 A sequence model takes sequences as its input. A sequence is made up of several
 timesteps. The so-called timestep, is not necessary to have something to do
-with 'time'. It can also be explained to that the order of data are taken into
+with time. It can also be explained to that the order of data are taken into
 consideration into model design and training.
 For example, the sentence can be interpreted as a kind of sequence data in NLP
 tasks.
@@ -155,23 +168,7 @@ Reference
 @provider
 +++++++++
-'@provider' is a Python `Decorator`_, it can construct a PyDataProvider in
+.. autofunction:: paddle.trainer.PyDataProvider2.provider
-PaddlePaddle from a user defined function. Its parameters are:
-* `input_types`_ defines format of the data input.
-* should_shuffle defines whether to shuffle data or not. By default, it is set
-  true during training, and false during testing.
-* pool_size is the memory pool size (in sample number) in DataProvider.
-  -1 means no limit.
-* can_over_batch_size defines whether PaddlePaddle can store little more
-  samples than pool_size. It is better to set True to avoid some deadlocks.
-* calc_batch_size is a function define how to calculate batch size. This is
-  usefull in sequential model, that defines batch size is counted upon sequence
-  or token. By default, each sample or sequence counts to 1 when calculating
-  batch size.
-* cache is a data cache strategy, see `cache`_.
-* Init_hook function is invoked once the data provider is initialized,
-  see `init_hook`_.
 input_types
 +++++++++++

--- a/doc/genindex.html
+++ b/doc/genindex.html
@@ -6389,7 +6389,7 @@
  </dt>
-  <dt><a href="source/gserver/dataprovider/dataproviders.html#_CPPv2N6paddle12DataProvider6createERK10DataConfigb">paddle::DataProvider::create (C++ function)</a>
+  <dt><a href="source/gserver/dataprovider/dataproviders.html#_CPPv2N6paddle12DataProvider6createERK10DataConfigRK11ModelConfigb">paddle::DataProvider::create (C++ function)</a>, <a href="source/gserver/dataprovider/dataproviders.html#_CPPv2N6paddle12DataProvider6createERK10DataConfigb">[1]</a>
  </dt>
@@ -9672,12 +9672,12 @@
  <dt><a href="source/math/matrix/matrix.html#_CPPv2N6paddle12MatrixOffset5cCol_E">paddle::MatrixOffset::cCol_ (C++ member)</a>
  </dt>
-  </dl></td>
-  <td style="width: 33%" valign="top"><dl>
  <dt><a href="source/math/matrix/matrix.html#_CPPv2N6paddle12MatrixOffset5cRow_E">paddle::MatrixOffset::cRow_ (C++ member)</a>
  </dt>
+  </dl></td>
+  <td style="width: 33%" valign="top"><dl>
  <dt><a href="source/math/matrix/matrix.html#_CPPv2N6paddle12MatrixOffset5dCol_E">paddle::MatrixOffset::dCol_ (C++ member)</a>
  </dt>
@@ -10143,7 +10143,7 @@
  </dt>
-  <dt><a href="source/gserver/dataprovider/dataproviders.html#_CPPv2N6paddle17MultiDataProvider17MultiDataProviderERK10DataConfigb">paddle::MultiDataProvider::MultiDataProvider (C++ function)</a>
+  <dt><a href="source/gserver/dataprovider/dataproviders.html#_CPPv2N6paddle17MultiDataProvider17MultiDataProviderERK10DataConfigRK11ModelConfigb">paddle::MultiDataProvider::MultiDataProvider (C++ function)</a>
  </dt>
@@ -13163,7 +13163,7 @@
  </dt>
-  <dt><a href="source/gserver/dataprovider/dataproviders.html#_CPPv2N6paddle15PyDataProvider215PyDataProvider2ERK10DataConfigb">paddle::PyDataProvider2::PyDataProvider2 (C++ function)</a>
+  <dt><a href="source/gserver/dataprovider/dataproviders.html#_CPPv2N6paddle15PyDataProvider215PyDataProvider2ERK10DataConfigRK11ModelConfigb">paddle::PyDataProvider2::PyDataProvider2 (C++ function)</a>
  </dt>
@@ -16814,6 +16814,10 @@
  <dt><a href="source/api/api.html#_CPPv2N25ParameterTraverseCallbackD0Ev">ParameterTraverseCallback::~ParameterTraverseCallback (C++ function)</a>
  </dt>
+  <dt><a href="ui/data_provider/pydataprovider2.html#paddle.trainer.PyDataProvider2.provider">provider() (in module paddle.trainer.PyDataProvider2)</a>
+  </dt>
  </dl></td>
 </tr></table>

--- a/doc/objects.inv
+++ b/doc/objects.inv
--- a/doc/searchindex.js
+++ b/doc/searchindex.js
--- a/doc/source/gserver/dataprovider/dataproviders.html
+++ b/doc/source/gserver/dataprovider/dataproviders.html
@@ -164,16 +164,22 @@
 <div class="breathe-sectiondef container">
 <p class="breathe-sectiondef-title rubric">Public Static Functions</p>
 <dl class="function">
-<dt id="_CPPv2N6paddle12DataProvider6createERK10DataConfigb">
+<dt id="_CPPv2N6paddle12DataProvider6createERK10DataConfigRK11ModelConfigb">
-<span id="paddle::DataProvider::create__DataConfigCR.b"></span><span class="target" id="paddleclasspaddle_1_1DataProvider_1ae6b6c293f49094fa37e06504da998248"></span><a class="reference internal" href="#_CPPv2N6paddle12DataProviderE" title="paddle::DataProvider">DataProvider</a> *<code class="descname">create</code><span class="sig-paren">(</span><em class="property">const</em> DataConfig &amp;<em>config</em>, bool <em>useGpu</em> = FLAGS_use_gpu<span class="sig-paren">)</span><a class="headerlink" href="#_CPPv2N6paddle12DataProvider6createERK10DataConfigb" title="Permalink to this definition">¶</a></dt>
+<span id="paddle::DataProvider::create__DataConfigCR.ModelConfigCR.b"></span><span class="target" id="paddleclasspaddle_1_1DataProvider_1a761ef962e6ef45b14562632db6be39c9"></span><a class="reference internal" href="#_CPPv2N6paddle12DataProviderE" title="paddle::DataProvider">DataProvider</a> *<code class="descname">create</code><span class="sig-paren">(</span><em class="property">const</em> DataConfig &amp;<em>config</em>, <em class="property">const</em> <a class="reference internal" href="../../api/api.html#_CPPv211ModelConfig" title="ModelConfig">ModelConfig</a> &amp;<em>modelConfig</em>, bool <em>useGpu</em> = FLAGS_use_gpu<span class="sig-paren">)</span><a class="headerlink" href="#_CPPv2N6paddle12DataProvider6createERK10DataConfigRK11ModelConfigb" title="Permalink to this definition">¶</a></dt>
 <dd></dd></dl>
+<dl class="function">
+<dt id="_CPPv2N6paddle12DataProvider6createERK10DataConfigb">
+<span id="paddle::DataProvider::create__DataConfigCR.b"></span><span class="target" id="paddleclasspaddle_1_1DataProvider_1ad782dc59f7366c19ba4375101159ba95"></span><em class="property">static</em> <a class="reference internal" href="#_CPPv2N6paddle12DataProviderE" title="paddle::DataProvider">DataProvider</a> *<code class="descname">create</code><span class="sig-paren">(</span><em class="property">const</em> DataConfig &amp;<em>config</em>, bool <em>useGpu</em><span class="sig-paren">)</span><a class="headerlink" href="#_CPPv2N6paddle12DataProvider6createERK10DataConfigb" title="Permalink to this definition">¶</a></dt>
+<dd><p>create only used for unittest. </p>
+</dd></dl>
 </div>
 <div class="breathe-sectiondef container">
 <p class="breathe-sectiondef-title rubric">Public Static Attributes</p>
 <dl class="member">
 <dt id="_CPPv2N6paddle12DataProvider10registrar_E">
-<span id="paddle::DataProvider::registrar___ClassRegistrar:DataProvider.DataConfig.b:"></span><span class="target" id="paddleclasspaddle_1_1DataProvider_1acc7ff8754097b2ecdd6c85ba98e78e18"></span>ClassRegistrar&lt;<a class="reference internal" href="#_CPPv2N6paddle12DataProviderE" title="paddle::DataProvider">DataProvider</a>, DataConfig, bool&gt; <code class="descname">registrar_</code><a class="headerlink" href="#_CPPv2N6paddle12DataProvider10registrar_E" title="Permalink to this definition">¶</a></dt>
+<span id="paddle::DataProvider::registrar___ClassRegistrar:DataProvider.DataConfig.ModelConfig.b:"></span><span class="target" id="paddleclasspaddle_1_1DataProvider_1ae40e5169aa51da3fb0e903ff75b8ab01"></span>ClassRegistrar&lt;<a class="reference internal" href="#_CPPv2N6paddle12DataProviderE" title="paddle::DataProvider">DataProvider</a>, DataConfig, <a class="reference internal" href="../../api/api.html#_CPPv211ModelConfig" title="ModelConfig">ModelConfig</a>, bool&gt; <code class="descname">registrar_</code><a class="headerlink" href="#_CPPv2N6paddle12DataProvider10registrar_E" title="Permalink to this definition">¶</a></dt>
 <dd></dd></dl>
 </div>
@@ -357,8 +363,8 @@
 <div class="breathe-sectiondef container">
 <p class="breathe-sectiondef-title rubric">Public Functions</p>
 <dl class="function">
-<dt id="_CPPv2N6paddle17MultiDataProvider17MultiDataProviderERK10DataConfigb">
+<dt id="_CPPv2N6paddle17MultiDataProvider17MultiDataProviderERK10DataConfigRK11ModelConfigb">
-<span id="paddle::MultiDataProvider::MultiDataProvider__DataConfigCR.b"></span><span class="target" id="paddleclasspaddle_1_1MultiDataProvider_1a9335b9b57d19fccd0b374746b0e28336"></span><code class="descname">MultiDataProvider</code><span class="sig-paren">(</span><em class="property">const</em> DataConfig &amp;<em>config</em>, bool <em>useGpu</em><span class="sig-paren">)</span><a class="headerlink" href="#_CPPv2N6paddle17MultiDataProvider17MultiDataProviderERK10DataConfigb" title="Permalink to this definition">¶</a></dt>
+<span id="paddle::MultiDataProvider::MultiDataProvider__DataConfigCR.ModelConfigCR.b"></span><span class="target" id="paddleclasspaddle_1_1MultiDataProvider_1acd993858e31a0f829e7ca3034bcdb655"></span><code class="descname">MultiDataProvider</code><span class="sig-paren">(</span><em class="property">const</em> DataConfig &amp;<em>config</em>, <em class="property">const</em> <a class="reference internal" href="../../api/api.html#_CPPv211ModelConfig" title="ModelConfig">ModelConfig</a> &amp;<em>modelConfig</em>, bool <em>useGpu</em><span class="sig-paren">)</span><a class="headerlink" href="#_CPPv2N6paddle17MultiDataProvider17MultiDataProviderERK10DataConfigRK11ModelConfigb" title="Permalink to this definition">¶</a></dt>
 <dd></dd></dl>
 <dl class="function">
@@ -977,8 +983,8 @@ virtual <span class="target" id="paddleclasspaddle_1_1SparseValueScanner_1a7ce57
 <div class="breathe-sectiondef container">
 <p class="breathe-sectiondef-title rubric">Public Functions</p>
 <dl class="function">
-<dt id="_CPPv2N6paddle15PyDataProvider215PyDataProvider2ERK10DataConfigb">
+<dt id="_CPPv2N6paddle15PyDataProvider215PyDataProvider2ERK10DataConfigRK11ModelConfigb">
-<span id="paddle::PyDataProvider2::PyDataProvider2__DataConfigCR.b"></span><span class="target" id="paddleclasspaddle_1_1PyDataProvider2_1a493a2635a8b83aeb9be3ce6db0d3aa55"></span><code class="descname">PyDataProvider2</code><span class="sig-paren">(</span><em class="property">const</em> DataConfig &amp;<em>config</em>, bool <em>useGpu</em><span class="sig-paren">)</span><a class="headerlink" href="#_CPPv2N6paddle15PyDataProvider215PyDataProvider2ERK10DataConfigb" title="Permalink to this definition">¶</a></dt>
+<span id="paddle::PyDataProvider2::PyDataProvider2__DataConfigCR.ModelConfigCR.b"></span><span class="target" id="paddleclasspaddle_1_1PyDataProvider2_1a2281a7aca7c68247414746864688e7bb"></span><code class="descname">PyDataProvider2</code><span class="sig-paren">(</span><em class="property">const</em> DataConfig &amp;<em>config</em>, <em class="property">const</em> <a class="reference internal" href="../../api/api.html#_CPPv211ModelConfig" title="ModelConfig">ModelConfig</a> &amp;<em>modelConfig</em>, bool <em>useGpu</em><span class="sig-paren">)</span><a class="headerlink" href="#_CPPv2N6paddle15PyDataProvider215PyDataProvider2ERK10DataConfigRK11ModelConfigb" title="Permalink to this definition">¶</a></dt>
 <dd><p>Ctor </p>
 </dd></dl>

--- a/doc/ui/data_provider/pydataprovider2.html
+++ b/doc/ui/data_provider/pydataprovider2.html
@@ -79,7 +79,7 @@ of 28 x 28 pixels.</p>
 <span class="mi">4</span><span class="p">;</span><span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mf">0.180392</span> <span class="mf">0.470588</span> <span class="mf">0.623529</span> <span class="mf">0.623529</span> <span class="mf">0.623529</span> <span class="mf">0.588235</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mf">0.243137</span> <span class="mf">0.494118</span> <span class="mf">0.862745</span> <span class="mf">0.870588</span> <span class="mf">0.960784</span> <span class="mf">0.996078</span> <span class="mf">0.996078</span> <span class="mf">0.996078</span> <span class="mf">0.996078</span> <span class="mf">0.992157</span> <span class="mf">0.466667</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mf">0.317647</span> <span class="mf">0.639216</span> <span class="mf">0.639216</span> <span class="mf">0.639216</span> <span class="mf">0.639216</span> <span class="mf">0.639216</span> <span class="mf">0.470588</span> <span class="mf">0.262745</span> <span class="mf">0.333333</span> <span class="mf">0.929412</span> <span class="mf">0.694118</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mf">0.811765</span> <span class="mf">0.694118</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mf">0.811765</span> <span class="mf">0.694118</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mf">0.811765</span> <span class="mf">0.694118</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mf">0.184314</span> <span class="mf">0.992157</span> <span class="mf">0.694118</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mf">0.192157</span> <span class="mf">0.996078</span> <span class="mf">0.384314</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mf">0.454902</span> <span class="mf">0.980392</span> <span class="mf">0.219608</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mf">0.564706</span> <span class="mf">0.941176</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mf">0.588235</span> <span class="mf">0.776471</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mf">0.945098</span> <span class="mf">0.560784</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mf">0.054902</span> <span class="mf">0.952941</span> <span class="mf">0.356863</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mf">0.337255</span> <span class="mf">0.917647</span> <span class="mf">0.109804</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mf">0.698039</span> <span class="mf">0.701961</span> <span class="mf">0.019608</span> <span class="mf">0.4</span> <span class="mf">0.662745</span> <span class="mf">0.662745</span> <span class="mf">0.662745</span> <span class="mf">0.662745</span> <span class="mf">0.662745</span> <span class="mf">0.662745</span> <span class="mf">0.662745</span> <span class="mf">0.376471</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mf">0.090196</span> <span class="mf">0.639216</span> <span class="mf">0.972549</span> <span class="mf">0.945098</span> <span class="mf">0.913725</span> <span class="mf">0.996078</span> <span class="mf">0.996078</span> <span class="mf">0.996078</span> <span class="mf">0.996078</span> <span class="mi">1</span> <span class="mf">0.996078</span> <span class="mf">0.996078</span> <span class="mi">1</span> <span class="mf">0.996078</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mf">0.007843</span> <span class="mf">0.105882</span> <span class="mf">0.717647</span> <span class="mf">0.776471</span> <span class="mf">0.905882</span> <span class="mf">0.996078</span> <span class="mf">0.996078</span> <span class="mf">0.988235</span> <span class="mf">0.980392</span> <span class="mf">0.862745</span> <span class="mf">0.537255</span> <span class="mf">0.223529</span> <span class="mf">0.223529</span> <span class="mf">0.368627</span> <span class="mf">0.376471</span> <span class="mf">0.6</span> <span class="mf">0.6</span> <span class="mf">0.6</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mf">0.262745</span> <span class="mf">0.470588</span> <span class="mf">0.6</span> <span class="mf">0.996078</span> <span class="mf">0.996078</span> <span class="mf">0.996078</span> <span class="mf">0.996078</span> <span class="mf">0.847059</span> <span class="mf">0.356863</span> <span class="mf">0.156863</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mf">0.909804</span> <span class="mf">0.705882</span> <span class="mf">0.823529</span> <span class="mf">0.635294</span> <span class="mf">0.490196</span> <span class="mf">0.219608</span> <span class="mf">0.113725</span> <span class="mf">0.062745</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mf">0.152941</span> <span class="mf">0.152941</span> <span class="mf">0.156863</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span> <span class="mi">0</span><span class="p">;</span>
 </pre></div>
 </div>
-<p>Each line of the data contains two parts, separated by &#8216;;&#8217;. The first part is
+<p>Each line of the data contains two parts, separated by <code class="code docutils literal"><span class="pre">;</span></code>. The first part is
 label of an image. The second part contains 28x28 pixel float values.</p>
 <p>Just write path of the above data into train.list. It looks like this:</p>
 <div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">mnist_train</span><span class="o">.</span><span class="n">txt</span>
@@ -148,9 +148,70 @@ you can take this as an example.</p>
                        <span class="n">test_list</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
                        <span class="n">module</span><span class="o">=</span><span class="s1">&#39;mnist_provider&#39;</span><span class="p">,</span>
                        <span class="n">obj</span><span class="o">=</span><span class="s1">&#39;process&#39;</span><span class="p">)</span>
+<span class="n">img</span> <span class="o">=</span> <span class="n">data_layer</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;pixel&#39;</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="mi">784</span><span class="p">)</span>
+<span class="n">label</span> <span class="o">=</span> <span class="n">data_layer</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;label&#39;</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="mi">10</span><span class="p">)</span>
 </pre></div>
 </div>
-<p>Here we specify training data by &#8216;train.list&#8217;, and no testing data is specified.</p>
+<p>Here we specify training data by <code class="code docutils literal"><span class="pre">train.list</span></code>, and no testing data is specified.
+The method which actually provide data is <code class="code docutils literal"><span class="pre">process</span></code>.</p>
+<p>User also can use another style to provide data, which defines the
+<code class="code docutils literal"><span class="pre">data_layer</span></code>&#8216;s name explicitly when <cite>yield</cite>. For example,
+the <code class="code docutils literal"><span class="pre">dataprovider</span></code> is shown as below.</p>
+<div class="highlight-default"><table class="highlighttable"><tr><td class="linenos"><div class="linenodiv"><pre> 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</pre></div></td><td class="code"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">paddle.trainer.PyDataProvider2</span> <span class="k">import</span> <span class="o">*</span>
+<span class="c1"># Define a py data provider</span>
+<span class="nd">@provider</span><span class="p">(</span><span class="n">input_types</span><span class="o">=</span><span class="p">[</span>
+    <span class="n">dense_vector</span><span class="p">(</span><span class="mi">28</span> <span class="o">*</span> <span class="mi">28</span><span class="p">),</span>
+    <span class="n">integer_value</span><span class="p">(</span><span class="mi">10</span><span class="p">)</span>
+<span class="p">])</span>
+<span class="k">def</span> <span class="nf">process</span><span class="p">(</span><span class="n">settings</span><span class="p">,</span> <span class="n">filename</span><span class="p">):</span>  <span class="c1"># settings is not used currently.</span>
+    <span class="n">f</span> <span class="o">=</span> <span class="nb">open</span><span class="p">(</span><span class="n">filename</span><span class="p">,</span> <span class="s1">&#39;r&#39;</span><span class="p">)</span>  <span class="c1"># open one of training file</span>
+    <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">f</span><span class="p">:</span>  <span class="c1"># read each line</span>
+        <span class="n">label</span><span class="p">,</span> <span class="n">pixel</span> <span class="o">=</span> <span class="n">line</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s1">&#39;;&#39;</span><span class="p">)</span>
+        <span class="c1"># get features and label</span>
+        <span class="n">pixels_str</span> <span class="o">=</span> <span class="n">pixel</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s1">&#39; &#39;</span><span class="p">)</span>
+        <span class="n">pixels_float</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">each_pixel_str</span> <span class="ow">in</span> <span class="n">pixels_str</span><span class="p">:</span>
+            <span class="n">pixels_float</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="nb">float</span><span class="p">(</span><span class="n">each_pixel_str</span><span class="p">))</span>
+        <span class="c1"># give data to paddle.</span>
+        <span class="k">yield</span> <span class="p">{</span> <span class="s2">&quot;pixel&quot;</span><span class="p">:</span> <span class="n">pixels_float</span><span class="p">,</span> <span class="s1">&#39;label&#39;</span><span class="p">:</span> <span class="nb">int</span><span class="p">(</span><span class="n">label</span><span class="p">)</span> <span class="p">}</span>
+    <span class="n">f</span><span class="o">.</span><span class="n">close</span><span class="p">()</span>  <span class="c1"># close file</span>
+</pre></div>
+</td></tr></table></div>
+<p>If user did&#8217;t give the <code class="code docutils literal"><span class="pre">data_layer</span></code>&#8216;s name, PaddlePaddle will use
+the order of <code class="code docutils literal"><span class="pre">data_layer</span></code> definition roughly to determine which feature to
+which <code class="code docutils literal"><span class="pre">data_layer</span></code>. This order may be not correct, so TO DEFINE THE
+<code class="code docutils literal"><span class="pre">data_layer</span></code>&#8216;s NAMES EXPLICITLY IS THE RECOMMANDED WAY TO PROVIDER DATA.</p>
 <p>Now, this simple example of using PyDataProvider is finished.
 The only thing that the user should know is how to generte <strong>one sample</strong> from
 <strong>one data file</strong>.
@@ -168,7 +229,7 @@ And PaddlePadle will do all of the rest things:</p>
 <h2>DataProvider for the sequential model<a class="headerlink" href="#dataprovider-for-the-sequential-model" title="Permalink to this headline">¶</a></h2>
 <p>A sequence model takes sequences as its input. A sequence is made up of several
 timesteps. The so-called timestep, is not necessary to have something to do
-with &#8216;time&#8217;. It can also be explained to that the order of data are taken into
+with time. It can also be explained to that the order of data are taken into
 consideration into model design and training.
 For example, the sentence can be interpreted as a kind of sequence data in NLP
 tasks.</p>
@@ -278,24 +339,73 @@ Please refer to the following section reference for details.</p>
 <h2>Reference<a class="headerlink" href="#reference" title="Permalink to this headline">¶</a></h2>
 <div class="section" id="provider">
 <h3>&#64;provider<a class="headerlink" href="#provider" title="Permalink to this headline">¶</a></h3>
-<p><a class="reference external" href="mailto:'&#37;&#52;&#48;provider">'<span>&#64;</span>provider</a>&#8216; is a Python <a class="reference external" href="http://www.learnpython.org/en/Decorators">Decorator</a>, it can construct a PyDataProvider in
+<dl class="function">
-PaddlePaddle from a user defined function. Its parameters are:</p>
+<dt id="paddle.trainer.PyDataProvider2.provider">
-<ul class="simple">
+<code class="descclassname">paddle.trainer.PyDataProvider2.</code><code class="descname">provider</code><span class="sig-paren">(</span><em>input_types=None</em>, <em>should_shuffle=None</em>, <em>pool_size=-1</em>, <em>min_pool_size=-1</em>, <em>can_over_batch_size=True</em>, <em>calc_batch_size=None</em>, <em>cache=0</em>, <em>check=False</em>, <em>check_fail_continue=False</em>, <em>use_dynamic_order=True</em>, <em>init_hook=None</em>, <em>**kwargs</em><span class="sig-paren">)</span><a class="headerlink" href="#paddle.trainer.PyDataProvider2.provider" title="Permalink to this definition">¶</a></dt>
-<li><a class="reference internal" href="#input-types">input_types</a> defines format of the data input.</li>
+<dd><p>Provider decorator. Use it to make a function into PyDataProvider2 object.
-<li>should_shuffle defines whether to shuffle data or not. By default, it is set
+In this function, user only need to get each sample for some train/test
-true during training, and false during testing.</li>
+file.</p>
-<li>pool_size is the memory pool size (in sample number) in DataProvider.
+<p>The basic usage is:</p>
-1 means no limit.</li>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="nd">@provider</span><span class="p">(</span><span class="n">some</span> <span class="n">data</span> <span class="n">provider</span> <span class="n">config</span> <span class="n">here</span><span class="o">...</span><span class="p">)</span>
-<li>can_over_batch_size defines whether PaddlePaddle can store little more
+<span class="k">def</span> <span class="nf">process</span><span class="p">(</span><span class="n">settings</span><span class="p">,</span> <span class="n">file_name</span><span class="p">):</span>
-samples than pool_size. It is better to set True to avoid some deadlocks.</li>
+    <span class="k">while</span> <span class="ow">not</span> <span class="n">at</span> <span class="n">end</span> <span class="n">of</span> <span class="n">file_name</span><span class="p">:</span>
-<li>calc_batch_size is a function define how to calculate batch size. This is
+        <span class="n">sample</span> <span class="o">=</span> <span class="n">readOneSampleFromFile</span><span class="p">(</span><span class="n">file_name</span><span class="p">)</span>
-usefull in sequential model, that defines batch size is counted upon sequence
+        <span class="k">yield</span> <span class="n">sample</span><span class="o">.</span>
-or token. By default, each sample or sequence counts to 1 when calculating
+</pre></div>
-batch size.</li>
+</div>
-<li>cache is a data cache strategy, see <a class="reference internal" href="#cache">cache</a>.</li>
+<p>The configuration of data provider should be setup by:</p>
-<li>Init_hook function is invoked once the data provider is initialized,
+<table class="docutils field-list" frame="void" rules="none">
-see <a class="reference internal" href="#init-hook">init_hook</a>.</li>
+<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_types</strong> (<em>list|tuple</em>) &#8211; Specify the input types, can also be set in init_hook.
+It is a list of InputType object. For example, input_types=                         [dense_vector(9), integer_value(2)].</li>
+<li><strong>should_shuffle</strong> (<em>bool</em>) &#8211; True if data should shuffle. Pass None means shuffle
+when is training and not to shuffle when is testing.</li>
+<li><strong>pool_size</strong> (<em>int</em>) &#8211; Max number of sample in data pool.</li>
+<li><strong>min_pool_size</strong> (<em>int</em>) &#8211; Set minimal sample in data pool. The PaddlePaddle will
+random pick sample in pool. So the min_pool_size
+effect the randomize of data.</li>
+<li><strong>can_over_batch_size</strong> (<em>bool</em>) &#8211; <p>True if paddle can return a mini-batch larger
+than batch size in settings. It is useful when
+custom calculate one sample&#8217;s batch_size.</p>
+<p>It is very danger to set it to false and use
+calc_batch_size together. Default is false.</p>
+</li>
+<li><strong>calc_batch_size</strong> (<em>callable</em>) &#8211; a method to calculate each sample&#8217;s batch size.
+Default each sample&#8217;s batch size is 1. But to you
+can customize each sample&#8217;s batch size.</li>
+<li><strong>cache</strong> (<em>int</em>) &#8211; Cache strategy of Data Provider. Default is CacheType.NO_CACHE</li>
+<li><strong>init_hook</strong> (<em>callable</em>) &#8211; <p>Initialize hook. Useful when data provider need load some
+external data like dictionary. The parameter is
+(settings, file_list, **kwargs).</p>
+<ul>
+<li>settings. It is the global settings object. User can set
+settings.input_types here.</li>
+<li>file_list. All file names for passed to data provider.</li>
+<li>is_train. Is this data provider used for training or not.</li>
+<li>kwargs. Other keyword arguments passed from
+trainer_config&#8217;s args parameter.</li>
+</ul>
+</li>
+<li><strong>check</strong> (<em>bool</em>) &#8211; Check the yield data format is as same as input_types. Enable
+this will make data provide process slow but it is very useful
+for debug. Default is disabled.</li>
+<li><strong>check_fail_continue</strong> (<em>bool</em>) &#8211; Continue train or not when check failed. Just
+drop the wrong format data when it is True. Has
+no effect when check set to False.</li>
+<li><strong>use_dynamic_order</strong> (<em>bool</em>) &#8211; Allow provider to yield a dictionary object, whose
+key is a input data layer name, and value is the
+feature value. The tuples are still allowed when
+use_dynmaic_order is True.</li>
 </ul>
+</td>
+</tr>
+</tbody>
+</table>
+</dd></dl>
 </div>
 <div class="section" id="input-types">
 <h3>input_types<a class="headerlink" href="#input-types" title="Permalink to this headline">¶</a></h3>

--- a/doc_cn/_sources/ui/data_provider/pydataprovider2.txt
+++ b/doc_cn/_sources/ui/data_provider/pydataprovider2.txt
@@ -56,6 +56,14 @@ process函数调用多次 :code:`yield` 即可。 :code:`yield` 是Python的一
 这里说明了训练数据是 'train.list'，而没有测试数据。引用的DataProvider是 'mnist_provider' 
 这个模块中的 'process' 函数。
+同时，根据模型配置文件中 :code:`data_layer` 的名字，用户也可以显式指定返回的数据对应关系。例如:
+.. literalinclude:: mnist_provider.dict.py
+   :linenos:
+如果用户不指定返回数据的对应关系，那么PaddlePaddle会粗略的根据layer的声明顺序，
+来确定对应关系。这个对应关系可能不正确。所以推荐使用显式指定返回值和数据对应关系。
 至此，简单的PyDataProvider样例就说明完毕了。对于用户来说，讲数据发送给PaddlePaddle，仅仅需要
 知道如何从 **一个文件** 里面读取 **一条** 样本。而PaddlePaddle进程帮助用户做了
@@ -119,11 +127,13 @@ DataProvider创建的时候执行。这个初始化函数具有如下参数:
 @provider
 +++++++++
-'@provider'是一个Python的 `Decorator`_ ，他可以将某一个函数标记成一个PyDataProvider。它包含的参数有:
+:code:`@provider` 是一个Python的 `Decorator`_ ，他可以将某一个函数标记成一个PyDataProvider。它包含的参数有:
 *  `input_types`_ 是数据输入格式。具体有哪些格式，参考 `input_types`_ 。
 *  should_shuffle 是个DataProvider是不是要做shuffle，如果不设置的话，训练的时候默认shuffle，
-   测试的时候默认不shuffle
+   测试的时候默认不shuffle。
+*  min_pool_size 是设置DataProvider在内存中最小暂存的数据条数。这个也是PaddlePaddle所能够保证的shuffle粒度。
+   设置成-1的话，会预先读取全部数据到内存中。
 *  pool_size 是设置DataProvider在内存中暂存的数据条数。设置成-1的话，即不在乎内存暂存多少条数据。
 *  can_over_batch_size 表示是否允许Paddle暂存略微多余pool_size的数据。这样做可以避免很多死锁问题。
   一般推荐设置成True
@@ -131,6 +141,11 @@ DataProvider创建的时候执行。这个初始化函数具有如下参数:
   是一个batch size，但是有时为了计算均衡性，可以将一条数据设置成多个batch size
 *  cache 是数据缓存的策略，参考 `cache`_
 *  init_hook 是初始化时调用的函数，参考 `init_hook`_
+*  use_dynamic_order 如果是true的话，可以返回一个dict，key是data_layer的名字，value是特征值。同时，也可以
+   返回一个list或者tuple。如果是false的话，只能够返回list或者tuple
+*  check 设置成true的话，会根据input_types检查数据的合法性。
+*  check_fail_continue 如果设置成true的话，即使在check中数据不合法，也会扔到这条数据，继续训练。 如果
+   check是false的话，没有作用。
 input_types
 +++++++++++
@@ -190,3 +205,55 @@ DataProvider提供了两种简单的Cache策略。他们是
 * CacheType.NO_CACHE 不缓存任何数据，每次都会从python端读取数据
 * CacheType.CACHE_PASS_IN_MEM 第一个pass会从python端读取数据，剩下的pass会直接从内存里
  读取数据。 
+注意事项
+--------
+可能的内存泄露问题
++++++++++++++++++
+PaddlePaddle将train.list中的每一行，都传递给process函数，从而生成多个generator。
+即如果train.list中，有100个训练文件，即会生成100个generator。这个本身不是一个很
+严重的问题。
+但是，如果在训练时，每一条训练数据都是一个文件，并且，训练数据非常多的情况下，就
+会生成多个generator。每个generator在没有调用的时候，是几乎不占内存的。但是，当调
+用过一次的时候，generator便会存下当前的上下文(Context)。而这个Context可能会非常
+大。并且，generator至少调用两次才会知道是否停止。所以，即使在process里面只会有一
+个yield，也需要两次随机选择到同样的generator的时候，才会释放该段内存。
+..  code-block:: python
+    def func():
+        yield 0
+    f = func()  # 创建generator
+    tmp = next(f)  # 调用一次，返回0
+    tmp = next(f)  # 调用第二次的时候，才会Stop Iteration
+而如果按顺序调用这些generator就不会出现这个问题。
+所以最佳实践推荐不要将每一个样本都放入train.list。而是将样本的地址放入另一个文本
+文件，train.list写入那个文本文件的地址。 或者在python generator的上下文中尽量留
+下非常少的变量引用。例如
+..  code-block:: python
+    def real_process(fn):
+        # ... read from fn
+        return result   # 当函数返回的时候，python可以解除掉内部变量的引用。
+    def process(fn):
+        yield real_process(fn)
+这个问题是PyDataProvider读数据时候的逻辑问题，基本上不能整体修正。
+内存不够用的情况
++++++++++++++++
+PyDataProvider2会尽量使用内存。所以如果对于内存比较小的机器，推荐设置
+:code:`pool_size` 变量，而这个变量推荐大于训练的batch size，并且在内存足够
+的情况下越大越好。
--- a/doc_cn/searchindex.js
+++ b/doc_cn/searchindex.js
-Search.setIndex({envversion:49,filenames:["build/docker/build_docker_image","build_and_install/cmake/compile_options","build_and_install/cmake/index","build_and_install/cmake/install_deps","build_and_install/cmake/make_and_install","build_and_install/index","build_and_install/install/docker_install","build_and_install/install/ubuntu_install","cluster/index","demo/index","demo/quick_start/index","index","ui/cmd/dump_config","ui/cmd/index","ui/cmd/make_diagram","ui/cmd/merge_model","ui/cmd/paddle_pserver","ui/cmd/paddle_train","ui/cmd/paddle_version","ui/data_provider/index","ui/data_provider/pydataprovider2","ui/data_provider/write_new_dataprovider","ui/index","ui/predict/swig_py_paddle"],objects:{},objnames:{},objtypes:{},terms:{"0000x":10,"000\u5f20\u7070\u5ea6\u56fe\u7247\u7684\u6570\u5b57\u5206\u7c7b\u6570\u636e\u96c6":20,"00186201e":23,"04\u7248\u672c\u4f7f\u7528deb\u5305\u5b89\u88c5":7,"08823112e":23,"0b0":7,"10\u4ee5\u4e0a\u7684linux":6,"10\u7ef4\u7684\u6574\u6570\u503c":20,"10gbe":6,"10m":0,"12194102e":23,"12\u7248\u672c\u6d4b\u8bd5\u901a\u8fc7":0,"12\u7248\u672c\u7684\u60c5\u51b5\u4e0b\u5e76\u6ca1\u6709\u6d4b\u8bd5":0,"15501715e":23,"15mb":10,"16mb":10,"1\u7684\u8bdd":20,"252kb":10,"25639710e":23,"25k":10,"27787406e":23,"28\u7684\u50cf\u7d20\u7070\u5ea6\u503c":20,"28\u7684\u7a20\u5bc6\u5411\u91cf\u548c\u4e00\u4e2a":20,"2\u8fdb\u884c\u8fdb\u4e00\u6b65\u6f14\u5316":10,"32777140e":23,"36540484e":23,"40gbe":6,"43630644e":23,"48565123e":23,"48684503e":23,"49316648e":23,"50k":0,"51111044e":23,"53018653e":23,"56gbe":6,"5\u5230\u672c\u5730\u73af\u5883\u4e2d":7,"70634608e":23,"72296313e":23,"85625684e":23,"93137714e":23,"96644767e":23,"99982715e":23,"9\u7684\u6570\u5b57":20,"\u4e00\u4e2a\u6587\u4ef6":20,"\u4e00\u4e2alogging\u5bf9\u8c61":20,"\u4e00\u4e2apass\u8868\u793a\u8fc7\u4e00\u904d\u6240\u6709\u8bad\u7ec3\u6837\u672c":10,"\u4e00\u6761":20,"\u4e00\u81f4\u7684\u7279\u5f81":20,"\u4e00\u822c\u60c5\u51b5\u4e0b":19,"\u4e00\u822c\u63a8\u8350\u8bbe\u7f6e\u6210true":20,"\u4e00\u884c\u4e3a\u4e00\u4e2a\u6837\u672c":10,"\u4e00\u884c\u5bf9\u5e94\u4e00\u4e2a\u6570\u636e\u6587\u4ef6":19,"\u4e0a\u7684\u6548\u679c":10,"\u4e0b\u8f7d\u8fdb\u7a0b\u4f1a\u91cd\u542f":0,"\u4e0b\u8ff0\u5185\u5bb9\u5c06\u5206\u4e3a\u5982\u4e0b\u51e0\u4e2a\u7c7b\u522b\u63cf\u8ff0":6,"\u4e0b\u9762dataprovid":10,"\u4e0d\u4e00\u5b9a\u548c\u65f6\u95f4\u6709\u5173\u7cfb":20,"\u4e0d\u4f1a\u6267\u884c\u6d4b\u8bd5\u64cd\u4f5c":19,"\u4e0d\u5305\u542blabel":23,"\u4e0d\u540c\u7684\u6570\u636e\u7c7b\u578b\u548c\u5e8f\u5217\u6a21\u5f0f\u8fd4\u56de\u7684\u683c\u5f0f\u4e0d\u540c":20,"\u4e0d\u540c\u7684\u7a33\u5b9a\u7248\u672c":6,"\u4e0d\u7f13\u5b58\u4efb\u4f55\u6570\u636e":20,"\u4e0e\u8bad\u7ec3\u7f51\u7edc\u914d\u7f6e\u4e0d\u540c\u7684\u662f":10,"\u4e14\u5e8f\u5217\u7684\u6bcf\u4e00\u4e2a\u5143\u7d20\u8fd8\u662f\u4e00\u4e2a\u65f6\u95f4\u5e8f\u5217":20,"\u4e24\u4e2a\u6587\u6863":6,"\u4e24\u7c7b":10,"\u4e2a":10,"\u4e2d":10,"\u4e2d\u5b9a\u4e49\u4f7f\u7528\u54ea\u79cddataprovider\u53ca\u5176\u53c2\u6570":19,"\u4e2d\u5b9a\u4e49\u7684\u987a\u5e8f\u4e00\u81f4":20,"\u4e2d\u5bfb\u627e\u8fd9\u4e9bblas\u7684\u5b9e\u73b0":1,"\u4e2d\u7684":20,"\u4e2d\u7684\u4e8c\u8fdb\u5236\u4f7f\u7528\u4e86":6,"\u4e2d\u7684set":20,"\u4e2d\u914d\u7f6e":20,"\u4e3a":20,"\u4e3a\u4e86\u8fd0\u884cpaddlepaddle\u7684docker\u955c\u50cf":6,"\u4e3a\u4f8b\u8fdb\u884c\u9884\u6d4b":10,"\u4e3b\u8981\u51fd\u6570\u662fprocess\u51fd\u6570":20,"\u4e3b\u8981\u5206\u4e3a\u4ee5\u4e0b\u51e0\u4e2a\u6b65\u9aa4":23,"\u4e3b\u8981\u5305\u62ec\u4e24\u90e8\u5206":10,"\u4e3b\u8981\u662f\u589e\u52a0\u4e86\u521d\u59cb\u5316\u673a\u5236":20,"\u4e3b\u8981\u6b65\u9aa4\u4e3a":23,"\u4e3b\u8981\u7531\u4e8e\u65e7\u7248\u672c":0,"\u4e3b\u8981\u7684\u8f6f\u4ef6\u5305\u4e3apy_paddl":23,"\u4e5f\u4f1a\u8bfb\u53d6\u76f8\u5173\u8def\u5f84\u53d8\u91cf\u6765\u8fdb\u884c\u641c\u7d22":1,"\u4e5f\u53ef\u4ee5\u4f7f\u7528":20,"\u4e5f\u53ef\u4ee5\u548cpaddl":13,"\u4e5f\u53ef\u4ee5\u76f4\u63a5\u6267\u884c":6,"\u4e5f\u53ef\u4ee5\u901a\u8fc7\u5982\u4e0b\u65b9\u5f0f\u9884\u6d4b":10,"\u4e5f\u53ef\u4ee5\u901a\u8fc7save":10,"\u4e5f\u53ef\u4ee5\u9884\u6d4b\u6ca1\u6709label\u7684\u6d4b\u8bd5\u96c6":10,"\u4e5f\u5c31\u662f\u5c06\u8bcd\u5411\u91cf\u6a21\u578b\u989d\u6b65":10,"\u4e5f\u5c31\u662f\u76f4\u63a5\u5199\u5185\u5b58\u7684float\u6570\u7ec4":23,"\u4e86\u975e\u5e38\u65b9\u4fbf\u7684\u4e8c\u8fdb\u5236\u5206\u53d1\u624b\u6bb5":6,"\u4e8c\u6b21\u5f00\u53d1\u53ef\u4ee5":6,"\u4eba\u5458\u7b49\u7b49":0,"\u4ec5\u4ec5\u9700\u8981":20,"\u4ecb\u7ecdpaddlepaddle\u4f7f\u7528\u6d41\u7a0b\u548c\u5e38\u7528\u7684\u7f51\u7edc\u57fa\u7840\u5355\u5143\u7684\u914d\u7f6e\u65b9\u6cd5":10,"\u4ece\u6587\u4ef6\u4e2d\u8bfb\u53d6\u6bcf\u4e00\u6761\u6570\u636e":20,"\u4ece\u6587\u672c\u6587\u4ef6\u4e2d\u8bfb\u53d6":20,"\u4ece\u800c\u4e0d\u80fd\u5728\u8fd0\u884c\u7f16\u8bd1\u547d\u4ee4\u7684\u65f6\u5019\u63a5\u53d7\u53c2\u6570":0,"\u4ece\u800c\u9632\u6b62\u8fc7\u62df\u5408":19,"\u4ed6\u4eec\u662f":[13,20],"\u4ed6\u4eec\u7684\u4e0b\u8f7d\u5730\u5740\u662f":7,"\u4ed6\u53ef\u4ee5\u5c06\u67d0\u4e00\u4e2a\u51fd\u6570\u6807\u8bb0\u6210\u4e00\u4e2apydataprovid":20,"\u4ee4\u884c\u53c2\u6570\u6587\u6863":10,"\u4ee5\u592a\u7f51\u5361":6,"\u4ee5\u76f8\u5bf9\u8def\u5f84\u5f15\u7528":19,"\u4efb\u610f\u4e00\u79cdcblas\u5b9e\u73b0":1,"\u4f1a\u5c06latest\u66ff\u6362\u6210\u7a33\u5b9a\u7248\u672c\u7684\u7248\u672c\u53f7":6,"\u4f1a\u62a5\u5bfb\u627e\u4e0d\u5230\u8fd9\u4e9b\u52a8\u6001\u5e93":7,"\u4f1a\u6839\u636e\u547d\u4ee4\u884c\u53c2\u6570\u6307\u5b9a\u7684\u6d4b\u8bd5\u65b9\u5f0f":19,"\u4f20\u5165\u547d\u4ee4\u884c\u53c2\u6570\u521d\u59cb\u5316":23,"\u4f20\u5165\u7684\u662f\u4e00\u4e2a\u51fd\u6570":20,"\u4f20\u5165\u7684\u914d\u7f6e\u53c2\u6570\u5305\u62ec":0,"\u4f20\u5165\u8fd9\u4e2a\u53d8\u91cf\u7684\u65b9\u5f0f\u4e3a":20,"\u4f46\u662f":0,"\u4f46\u662f\u5728":20,"\u4f46\u662f\u5982\u679c\u4f7f\u7528\u4e86\u9ad8\u6027\u80fd\u7684\u7f51\u5361":6,"\u4f46\u662f\u65b9\u4fbf\u8c03\u8bd5\u548cbenchmark":1,"\u4f46\u662f\u6709\u65f6\u4e3a\u4e86\u8ba1\u7b97\u5747\u8861\u6027":20,"\u4f46\u9700\u8981\u6ce8\u610f\u7684\u662f\u7f16\u8bd1\u548c":1,"\u4f4e\u4e8edocker":0,"\u4f53\u53ef\u4ee5\u53c2\u8003":20,"\u4f7f\u5728python\u73af\u5883\u4e0b\u7684\u9884\u6d4b\u63a5\u53e3\u66f4\u52a0\u7b80\u5355":23,"\u4f7f\u7528":[1,6,23],"\u4f7f\u7528\u7684\u547d\u4ee4\u4e5f\u662f":1,"\u4f7f\u7528\u8be5\u63a5\u53e3\u7528\u6237\u53ef\u4ee5\u53ea\u5173\u6ce8\u5982\u4f55":20,"\u4f7f\u7528\u8be5dockerfile\u6784\u5efa\u51fa\u955c\u50cf":6,"\u4f7f\u7528\u8d77\u6765\u6216\u8bb8\u4f1a\u4e0d\u662f\u5f88\u987a\u7545":5,"\u4f7f\u7528\u8fd9\u4e2a\u5173\u952e\u8bcd":20,"\u4f7f\u7528deb\u5305\u5728ubuntu\u4e0a\u5b89\u88c5paddlepaddl":5,"\u4f7f\u7528dockerfile\u6784\u5efa\u4e00\u4e2a\u5168\u65b0\u7684dock":6,"\u4f7f\u7528mnist\u624b\u5199\u8bc6\u522b\u4f5c\u4e3a\u6837\u4f8b":23,"\u4f7f\u7528ssh\u8bbf\u95eepaddlepaddle\u955c\u50cf":6,"\u4f8b\u5982":[1,7,10,20],"\u4f8b\u5982\u6587\u4ef6\u540d\u662f":20,"\u4f8b\u5982rdma\u7f51\u5361":6,"\u4f8b\u5982sigmoid\u53d8\u6362":10,"\u4f9d\u6b21\u8fd4\u56de\u4e86\u6587\u4ef6\u4e2d\u7684\u6bcf\u6761\u6570\u636e":20,"\u4fe1\u606f":6,"\u5173\u95edcontain":6,"\u5176\u4e2d":[0,6,19,20,23],"\u5176\u4e2d\u6587\u672c\u8f93\u5165\u7c7b\u578b\u5b9a\u4e49\u4e3a\u6574\u6570\u65f6\u5e8f\u7c7b\u578binteg":10,"\u5176\u4e2d\u7b2c\u4e00\u884c\u662f\u5f15\u5165paddlepaddle\u7684pydataprovider2\u5305":20,"\u5176\u4ed6\u53c2\u6570\u5747\u4f7f\u7528kei":20,"\u5176\u4ed6\u53c2\u6570\u8bf7\u53c2\u8003":10,"\u5176\u4ed6\u53c2\u6570\u90fd\u4f7f\u7528kei":20,"\u5176\u4f5c\u7528\u662f\u5c06\u8bad\u7ec3\u6570\u636e\u4f20\u5165\u5185\u5b58\u6216\u8005\u663e\u5b58":19,"\u5176\u5b83\u90e8\u5206\u548c\u903b\u8f91\u56de\u5f52\u7f51\u7edc\u7ed3\u6784\u4e00\u81f4":10,"\u5176\u6570\u636e\u4f7f\u7528":20,"\u5176\u7b2c\u4e00\u884c\u8bf4\u660e\u4e86paddle\u7684\u7248\u672c":18,"\u5177":20,"\u5177\u4f53\u53ef\u4ee5\u8bbe\u7f6e\u6210\u4ec0\u4e48\u5176\u4ed6\u683c":20,"\u5177\u4f53\u6709\u54ea\u4e9b\u683c\u5f0f":20,"\u5177\u4f53\u8bf7\u53c2\u8003\u6ce8\u610f\u4e8b\u9879\u4e2d\u7684":6,"\u5177\u4f53dataprovider\u8fd8\u5177\u6709\u4ec0\u4e48\u529f\u80fd":20,"\u5177\u6709\u4e24\u4e2a\u53c2\u6570":20,"\u5177\u6709\u548c\u5bbf\u4e3b\u673a\u76f8\u8fd1\u7684\u8fd0\u884c\u6548\u7387":6,"\u5177\u6709\u7684\u5c5e\u6027\u6709":20,"\u5178\u578b\u7684\u8f93\u51fa\u7ed3\u679c\u4e3a":23,"\u5178\u578b\u7684\u9884\u6d4b\u4ee3\u7801\u5982\u4e0b":23,"\u518d\u6307\u5b9a":1,"\u5199\u5165train":20,"\u5199\u5728train":19,"\u51c6\u5907\u6570\u636e":23,"\u51fd\u6570":20,"\u51fd\u6570\u4e2d":20,"\u51fd\u6570\u4e2d\u4f7f\u7528":20,"\u51fd\u6570\u4e2d\u7684":20,"\u51fd\u6570\u662f\u4f7f\u7528":20,"\u51fd\u6570\u6765\u4fdd\u8bc1\u517c\u5bb9\u6027":20,"\u51fd\u6570\u67e5\u8be2\u6587\u6863":23,"\u5206\u5e03\u5f0f\u8bad\u7ec3":10,"\u5206\u7c7b\u6210\u6b63\u9762\u60c5\u7eea\u548c":20,"\u5217\u8868\u5982\u4e0b":20,"\u5219\u5b57\u4e0e\u5b57\u4e4b\u95f4\u7528\u7a7a\u683c\u5206\u9694":10,"\u5219\u9700\u8981\u5148\u5c06":6,"\u5219\u9700\u8981\u8fdb\u884c\u4e00\u5b9a\u7684\u4e8c\u6b21\u5f00\u53d1":6,"\u521b\u5efa\u4e00\u4e2a":23,"\u5220\u9664contain":6,"\u5229\u7528\u5355\u8bcdid\u67e5\u627e\u5bf9\u5e94\u7684\u8be5\u5355\u8bcd\u7684\u8fde\u7eed\u8868\u793a\u5411\u91cf":10,"\u5229\u7528\u903b\u8f91\u56de\u5f52\u6a21\u578b\u5bf9\u8be5\u5411\u91cf\u8fdb\u884c\u5206\u7c7b":10,"\u522b":10,"\u5269\u4e0b\u7684pass\u4f1a\u76f4\u63a5\u4ece\u5185\u5b58\u91cc":20,"\u52a0\u4e86l2\u6b63\u5219\u548c\u68af\u5ea6\u622a\u65ad":10,"\u52a0\u8f7d\u6570\u636e":10,"\u5305":6,"\u5305\u548c":6,"\u5305\u62ec":10,"\u5305\u62ec\u7b80\u5355\u7684rnn\u6a21\u578b":10,"\u5305\u62ecdocker\u955c\u50cf":5,"\u5373":[6,10],"\u5373\u4e0d\u5728\u4e4e\u5185\u5b58\u6682\u5b58\u591a\u5c11\u6761\u6570\u636e":20,"\u5373\u4e0d\u662f\u4e00\u6761\u5e8f\u5217":20,"\u5373\u4ece\u5355\u8bcd\u5b57\u7b26\u4e32\u5230\u5355\u8bcdid\u7684\u5b57\u5178":20,"\u5373\u5305\u542b\u65f6\u95f4\u6b65\u4fe1\u606f":20,"\u5373\u53ef":20,"\u5373\u53ef\u4ee5\u4f7f\u7528ssh\u8bbf\u95ee\u5bbf\u4e3b\u673a\u76848022\u7aef\u53e3":6,"\u5373\u53ef\u542f\u52a8\u548c\u8fdb\u5165paddlepaddle\u7684contain":6,"\u5373\u53ef\u5728\u672c\u5730\u7f16\u8bd1\u51fapaddlepaddle\u7684\u955c\u50cf":0,"\u5373\u53ef\u6253\u5370\u51fapaddlepaddle\u7684\u7248\u672c\u548c\u6784\u5efa":6,"\u5373\u5927\u90e8\u5206\u503c\u4e3a0":20,"\u5373\u5bf9\u7b2c3\u6b65\u8fdb\u884c\u66ff\u6362":10,"\u5373\u662f\u4e00\u6761\u65f6\u95f4\u5e8f\u5217":20,"\u5373cpu\u7248\u672c":7,"\u5373train":20,"\u5377\u79ef\u7f51\u7edc\u662f\u4e00\u79cd\u7279\u6b8a\u7684\u4ece\u8bcd\u5411\u91cf\u8868\u793a\u5230\u53e5\u5b50\u8868\u793a\u7684\u65b9\u6cd5":10,"\u53c2\u6570":0,"\u53c2\u6570\u6570\u91cf":10,"\u53c2\u8003":19,"\u53c2\u89c1":[3,4],"\u53d6\u51b3\u4e8e\u662f\u5426\u5bfb\u627e\u5230gflags":1,"\u53d6\u51b3\u4e8e\u662f\u5426\u5bfb\u627e\u5230glog":1,"\u53d6\u51b3\u4e8e\u662f\u5426\u5bfb\u627e\u5230gtest":1,"\u53d6\u51b3\u4e8e\u662f\u5426\u627e\u5230swig":1,"\u53d8\u4e3a3\u4e2a\u65b0\u7684\u5b50\u6b65\u9aa4":10,"\u53d8\u4f1a\u62a5\u8fd9\u4e2a\u9519\u8bef":7,"\u53e5\u5b50\u8868\u793a\u7684\u8ba1\u7b97\u66f4\u65b0\u4e3a2\u6b65":10,"\u53ea\u662f\u5c06\u53e5\u5b50\u5229\u7528\u8fde\u7eed\u5411\u91cf\u8868\u793a\u66ff\u6362\u7a00\u758f":10,"\u53ea\u662f\u8bf4\u660e\u6570\u636e\u7684\u987a\u5e8f\u662f\u91cd\u8981\u7684":20,"\u53ea\u9700\u8981\u4f7f\u7528\u4e00\u884c\u4ee3\u7801\u5373\u53ef\u4ee5\u8bbe\u7f6e\u8bad\u7ec3\u5f15\u7528\u8fd9\u4e2adataprovid":20,"\u53ea\u9700\u8981\u5728":20,"\u53ea\u9700\u8981\u77e5\u9053\u8fd9\u53ea\u662f\u4e00\u4e2a\u6807\u8bb0\u5c5e\u6027\u7684\u65b9\u6cd5\u5c31\u53ef\u4ee5\u4e86":20,"\u53ef\u4ee5\u4e3a\u4e00\u4e2a\u6570\u636e\u6587\u4ef6\u8fd4\u56de\u591a\u6761\u8bad\u7ec3\u6837\u672c":20,"\u53ef\u4ee5\u4f20\u516510k":0,"\u53ef\u4ee5\u4f7f\u7528":0,"\u53ef\u4ee5\u4f7f\u7528\u8bad\u7ec3\u597d\u7684\u6a21\u578b\u8bc4\u4f30\u5e26\u6709label\u7684\u9a8c\u8bc1\u96c6":10,"\u53ef\u4ee5\u4f7f\u7528graphviz\u5bf9paddlepaddle\u7684\u7f51\u7edc\u6a21\u578b\u8fdb\u884c\u7ed8\u5236":13,"\u53ef\u4ee5\u4f7f\u7528paddl":13,"\u53ef\u4ee5\u4f7f\u7528python\u7684":23,"\u53ef\u4ee5\u53c2\u8003":10,"\u53ef\u4ee5\u5728\u4e00\u4e2a\u51fd\u6570\u91cc":20,"\u53ef\u4ee5\u5728cmake\u7684\u547d\u4ee4\u884c\u8bbe\u7f6e":1,"\u53ef\u4ee5\u5c06\u4e00\u6761\u6570\u636e\u8bbe\u7f6e\u6210\u591a\u4e2abatch":20,"\u53ef\u4ee5\u5c06paddlepaddle\u7684\u6a21\u578b\u548c\u914d\u7f6e\u6253\u5305\u6210\u4e00\u4e2a\u6587\u4ef6":13,"\u53ef\u4ee5\u5c06paddlepaddle\u7684\u8bad\u7ec3\u6a21\u578b\u4ee5proto":13,"\u53ef\u4ee5\u65b9\u4fbf\u5d4c\u5165\u5f0f\u5de5\u4f5c":1,"\u53ef\u4ee5\u6839\u636e\u4e0d\u540c\u7684\u6570\u636e\u914d\u7f6e\u4e0d\u540c\u7684\u8f93\u5165\u7c7b\u578b":20,"\u53ef\u4ee5\u901a\u8fc7show":10,"\u53ef\u7528\u5728\u6d4b\u8bd5\u6216\u8bad\u7ec3\u65f6\u6307\u5b9a\u521d\u59cb\u5316\u6a21\u578b":10,"\u53ef\u9009":20,"\u5404\u79cd\u53c2\u6570\u548c\u7ef4\u62a4":0,"\u540c\u65f6":[0,1,6,20],"\u540c\u65f6\u4f1a\u8ba1\u7b97\u5206\u7c7b\u51c6\u786e\u7387":10,"\u540c\u65f6\u5305\u542bcpu\u5f00\u53d1\u73af\u5883":6,"\u540c\u65f6\u5305\u542bcpu\u5f00\u53d1\u73af\u5883\u548c\u6e90\u7801":6,"\u540c\u65f6\u5305\u542bgpu\u5f00\u53d1\u73af\u5883":6,"\u540c\u65f6\u5305\u542bgpu\u5f00\u53d1\u73af\u5883\u548c\u6e90\u7801":6,"\u540c\u65f6\u6b22\u8fce\u8d21\u732e\u66f4\u591a\u7684\u5b89\u88c5\u5305":5,"\u540d\u79f0":10,"\u540e\u9762\u8ddf\u7740\u4e00\u7cfb\u5217\u7f16\u8bd1\u53c2\u6570":18,"\u5411\u91cf\u8868\u793a":10,"\u5426":1,"\u5426\u5219":19,"\u547d\u4ee4":0,"\u547d\u4ee4\u4e3a":6,"\u547d\u4ee4\u6307\u5b9a\u7684\u53c2\u6570\u4f1a\u4f20\u5165\u7f51\u7edc\u914d\u7f6e\u4e2d":10,"\u547d\u4ee4\u8fd0\u884c\u955c\u50cf":6,"\u547d\u4ee4\u9884\u5148\u4e0b\u8f7d\u955c\u50cf":6,"\u548c":[1,6,20],"\u548c\u4e09\u79cd\u5e8f\u5217\u6a21\u5f0f":20,"\u548c\u5dee\u8bc4":10,"\u548c\u6587\u672c\u4fe1\u606f\u7528tab\u95f4\u9694":10,"\u548c\u6d4b\u8bd5\u6587\u4ef6\u5217\u8868":19,"\u548c\u7528\u6237\u4f20\u5165\u7684\u53c2\u6570":20,"\u548c\u9884\u5904\u7406\u811a\u672c":10,"\u548ccudnn":7,"\u548cgpu\u7248\u672c":7,"\u548cinitalizer\u91cc\u5b9a\u4e49\u987a\u5e8f\u4e00\u81f4":10,"\u5668":10,"\u56db\u79cd\u6570\u636e\u7c7b\u578b\u662f":20,"\u56fe\u50cf\u5206\u7c7b":9,"\u5728":[1,7,20],"\u5728\u58f0\u660edataprovider\u7684\u65f6\u5019\u4f20\u5165\u4e86dictionary\u4f5c\u4e3a\u53c2\u6570":20,"\u5728\u5b8c\u6210\u4e86\u6570\u636e\u548c\u7f51\u7edc\u7ed3\u6784\u642d\u5efa\u4e4b\u540e":10,"\u5728\u672c\u95ee\u9898\u4e2d":10,"\u5728\u6a21\u578b\u914d\u7f6e\u4e2d\u5229\u7528":10,"\u5728\u6b64\u4e3a\u65b9\u4fbf\u5bf9\u6bd4\u4e0d\u540c\u7f51\u7edc\u7ed3\u6784":10,"\u5728\u6bcf\u4e2a\u7ef4\u5ea6\u4e0a\u53d6\u51fa\u5728\u8be5\u53e5\u8bdd\u65b0\u7684\u5411\u91cf\u96c6\u5408\u4e0a\u8be5\u7ef4\u5ea6\u7684\u6700\u5927\u503c\u4f5c\u4e3a\u6700\u540e\u7684\u53e5\u5b50\u8868\u793a\u5411\u91cf":10,"\u5728\u7a0b\u5e8f\u5f00\u59cb\u9636\u6bb5":23,"\u5728\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u8fdb\u884c\u6d4b\u8bd5":19,"\u5728\u8bad\u7ec3\u914d\u7f6e\u91cc":20,"\u5728\u8fd9\u4e2a\u51fd\u6570\u4e2d":20,"\u5728\u914d\u7f6e\u4e2d\u8bfb\u53d6\u4e86\u5b57\u5178":20,"\u5728cmake\u914d\u7f6e\u65f6\u53ef\u4ee5\u4f7f\u7528":1,"\u5728pydataprovider\u4e2d":20,"\u5728python\u73af\u5883\u4e0b\u9884\u6d4b\u7ed3\u679c":23,"\u57fa\u672c\u7684\u5904\u7406\u903b\u8f91\u4e5f\u548cmnist\u903b\u8f91\u4e00\u81f4":20,"\u57fa\u672c\u7684pydataprovider\u4f7f\u7528\u4ecb\u7ecd\u5b8c\u6bd5\u4e86":20,"\u591a\u4e2ainput\u4ee5list\u65b9\u5f0f\u8f93\u5165":10,"\u591a\u6b21\u8fd4\u56de\u53d8\u91cf":20,"\u591a\u7ebf\u7a0b\u4e0b\u8f7d\u8fc7\u7a0b\u4e2d":0,"\u591a\u7ebf\u7a0b\u6570\u636e\u8bfb\u53d6":20,"\u597d\u8bc4":10,"\u5982\u679c\u4e0d\u4e86\u89e3":20,"\u5982\u679c\u4e0d\u4f7f\u7528\u5219\u4f1a\u4f7f\u7528\u4e00\u4e2a\u7b80\u5316\u7248\u7684\u547d\u4ee4\u884c\u53c2\u6570\u89e3\u6790":1,"\u5982\u679c\u4e0d\u4f7f\u7528\u5219\u4f1a\u4f7f\u7528\u4e00\u4e2a\u7b80\u5316\u7248\u7684\u65e5\u5fd7\u5b9e\u73b0":1,"\u5982\u679c\u4e0d\u5207\u8bcd":10,"\u5982\u679c\u4e0d\u8bbe\u7f6e\u7684\u8bdd":20,"\u5982\u679c\u4f7f\u7528gpu\u7248\u672c\u7684paddlepaddl":7,"\u5982\u679c\u5728":7,"\u5982\u679c\u5728\u7b2c\u4e00\u6b21cmake\u4e4b\u540e\u60f3\u8981\u91cd\u65b0\u8bbe":1,"\u5982\u679c\u5c0f\u4e8e\u8fd9\u4e2a\u4e0b\u8f7d\u901f\u5ea6":0,"\u5982\u679c\u60a8\u4f7f\u7528":6,"\u5982\u679c\u60f3\u8981\u5728\u5916\u90e8\u673a\u5668\u8bbf\u95ee\u8fd9\u4e2acontain":6,"\u5982\u679c\u6709\u66f4\u590d\u6742\u7684\u4f7f\u7528":19,"\u5982\u679c\u7cfb\u7edf\u4f7f\u7528\u4e0d\u652f\u6301":7,"\u5982\u679c\u8fd0\u884cgpu\u7248\u672c\u7684paddlepaddl":6,"\u5982\u679ctest":19,"\u5b83\u5305\u542b\u7684\u53c2\u6570\u6709":20,"\u5b89\u88c5\u597d\u7684paddlepaddle\u811a\u672c\u5305\u62ec\u591a\u6761\u547d\u4ee4":13,"\u5b89\u88c5\u5b8c\u6210\u7684paddlepaddle\u4e3b\u4f53\u5305\u62ec\u4e09\u4e2a\u90e8\u5206":6,"\u5b89\u88c5\u5b8c\u6210paddlepaddle\u540e":7,"\u5b89\u88c5\u6559\u7a0b":10,"\u5b89\u88c5\u65b9\u6cd5\u8bf7\u53c2\u8003":6,"\u5b89\u88c5\u7f16\u8bd1\u4f9d\u8d56":3,"\u5b89\u88c5\u7f16\u8bd1paddlepaddle\u9700\u8981\u7684\u4f9d\u8d56":2,"\u5b89\u88c5docker\u9700\u8981\u60a8\u7684\u673a\u5668":6,"\u5b89\u88c5paddlepaddl":10,"\u5b89\u88c5paddlepaddle\u7684docker\u955c\u50cf":5,"\u5b8c\u6210\u591a\u673a\u8bad\u7ec3":13,"\u5b8c\u6574\u4ee3\u7801\u89c1":23,"\u5b9a\u4e49\u6587\u672c\u4fe1\u606f":10,"\u5b9e\u73b0\u4e86\u6253\u5f00\u6587\u672c\u6587\u4ef6":20,"\u5bc6\u7801\u4e5f\u662froot":6,"\u5bf9\u4e8e\u7528\u6237\u6765\u8bf4":20,"\u5bf9\u4e8e\u7ed9\u5b9a\u7684\u4e00\u6761\u6587\u672c":10,"\u5bf9\u4e8ecuda\u7684toolkit\u6709\u65ad\u70b9\u7eed\u4f20\u548c\u4f20\u8f93\u901f\u5ea6\u8fc7\u5c0f\u91cd\u542f\u4e0b\u8f7d\u7684":0,"\u5bf9\u4e8emnist\u800c\u8a00":20,"\u5bf9\u8be5\u8868\u793a\u8fdb\u884c\u975e\u7ebf\u6027\u53d8\u6362":10,"\u5bf9\u8c61":20,"\u5bf9\u8c61convert":23,"\u5c06\u4f1a\u6d4b\u8bd5\u914d\u7f6e\u6587\u4ef6\u4e2dtest":10,"\u5c06\u5b57\u5178\u5b58\u5165\u4e86set":20,"\u5c06\u5bbf\u4e3b\u673a\u76848022\u7aef\u53e3\u6620\u5c04\u5230container\u768422\u7aef\u53e3\u4e0a":6,"\u5c06\u6570\u636e\u7ec4\u5408\u6210batch\u8bad\u7ec3":20,"\u5c06\u6587\u4ef6\u7684\u7edd\u5bf9\u8def\u5f84\u6216\u76f8\u5bf9\u8def\u5f84":19,"\u5c06\u8bc4\u8bba\u5206\u4e3a\u597d\u8bc4":10,"\u5c06\u8be5\u53e5\u8bdd\u5305\u542b\u7684\u6240\u6709\u5355\u8bcd\u5411\u91cf\u6c42\u5e73\u5747\u5f97\u5230\u53e5\u5b50\u7684\u8868\u793a":10,"\u5c06ssh\u88c5\u5165\u7cfb\u7edf\u5185\u5e76\u5f00\u542f\u8fdc\u7a0b\u8bbf\u95ee":6,"\u5c31\u50cf\u8fd9\u4e2a\u6837\u4f8b\u4e00\u6837":20,"\u5c31\u662f\u5c06\u8fd9\u4e9b\u52a8\u6001\u5e93\u52a0\u5230\u73af\u5883\u53d8\u91cf\u91cc\u9762":7,"\u5c5e\u6027":20,"\u5dee\u8bc4":10,"\u5e38\u89c1\u7684\u8f93\u51fa\u683c\u5f0f\u4e3a":18,"\u5e76\u4e14\u4f7f\u7528\u5173\u952e\u8bcd":20,"\u5e76\u4e14\u5220\u9664container\u4e2d\u7684\u6570\u636e":6,"\u5e76\u4e14\u6807\u8bb0process\u51fd\u6570\u662f\u4e00\u4e2adataprovid":20,"\u5e76\u4f7f\u7528\u4e86dropout":10,"\u5e76\u572823\u884c\u8fd4\u56de\u7ed9paddlepaddle\u8fdb\u7a0b":20,"\u5e76\u5c06\u6bcf\u884c\u8f6c\u6362\u6210\u548c":20,"\u5e76\u63d0\u4f9b":6,"\u5e76\u63d0\u4f9b\u4e86\u7b80\u5355\u7684cache\u529f\u80fd":20,"\u5e76\u8bbe\u7f6e\u597d\u5bf9\u5e94\u7684\u73af\u5883\u53d8\u91cf":7,"\u5e76\u9010\u6e10\u5c55\u793a\u66f4\u52a0\u6df1\u5165\u7684\u529f\u80fd":10,"\u5e8f\u5217\u6a21\u578b\u6570\u636e\u63d0\u4f9b":19,"\u5e8f\u5217\u6a21\u578b\u662f\u6307\u6570\u636e\u7684\u67d0\u4e00\u7ef4\u5ea6\u662f\u4e00\u4e2a\u5e8f\u5217\u5f62\u5f0f":20,"\u5e93\u7684\u8bdd":7,"\u5f0f":20,"\u5f15\u7528\u7684dataprovider\u662f":20,"\u5f53\u524dlog_period\u4e2abatch\u6240\u6709\u6837\u672c\u7684\u5e73\u5747\u5206\u7c7b\u9519\u8bef\u7387":10,"\u5f53\u524dlog_period\u4e2abatch\u6240\u6709\u6837\u672c\u7684\u5e73\u5747cost":10,"\u5f53\u7136":19,"\u5f88":10,"\u5f97\u5230\u7ed3\u679c":7,"\u5feb\u901f\u5165\u95e8":11,"\u5ff5\u662f":20,"\u60a8\u53ef\u4ee5\u4f7f\u7528":6,"\u60a8\u5c31\u53ef\u4ee5\u8fdc\u7a0b\u7684\u4f7f\u7528paddlepaddle\u5566":6,"\u60a8\u9700\u8981\u5728\u673a\u5668\u4e2d\u5b89\u88c5\u597ddocker":6,"\u60a8\u9700\u8981\u8fdb\u5165\u955c\u50cf\u8fd0\u884cpaddlepaddl":6,"\u60c5\u611f\u5206\u6790":9,"\u60f3\u8981\u8fd0\u884cpaddlepaddl":6,"\u6210\u4e3a\u7ef4\u5ea6\u4e3ahidden":10,"\u6211\u4eec\u4ece\u63d0\u524d\u7ed9\u5b9a\u7684\u7c7b\u522b\u96c6\u5408\u4e2d\u9009\u62e9\u5176\u6240\u5c5e\u7c7b":10,"\u6211\u4eec\u4ee5\u6587\u672c\u5206\u7c7b\u95ee\u9898\u4f5c\u4e3a\u80cc\u666f":10,"\u6211\u4eec\u4f7f\u7528":10,"\u6211\u4eec\u5728\u6b64\u603b":10,"\u6211\u4eec\u5c06\u4ee5\u57fa\u672c\u7684\u903b\u8f91\u56de\u5f52\u7f51\u7edc\u4f5c\u4e3a\u8d77\u70b9":10,"\u6211\u4eec\u5c06\u5728\u540e\u9762\u4ecb\u7ecd\u8bad\u7ec3\u548c\u9884\u6d4b\u7684\u6d41\u7a0b\u7684\u811a\u672c":10,"\u6211\u4eec\u5c06\u8bad\u7ec3\u7684\u547d\u4ee4\u884c\u4fdd\u5b58\u5728\u4e86":10,"\u6211\u4eec\u63a8\u8350\u4f7f\u7528docker\u955c\u50cf\u6765\u90e8\u7f72\u73af\u5883":5,"\u6211\u4eec\u63d0\u4f9b\u4e866\u4e2adocker":6,"\u6211\u4eec\u63d0\u4f9b\u4e86\u4e00\u4e2a\u5de5\u5177\u7c7bdataproviderconvert":23,"\u6211\u4eec\u8fdb\u5165\u5230\u8bad\u7ec3\u90e8\u5206":10,"\u6211\u4eec\u91c7\u7528\u5355\u5c42lstm\u6a21\u578b":10,"\u6211\u4eec\u968f\u65f6\u603b\u7ed3\u4e86\u5404\u4e2a\u7f51\u7edc\u7684\u590d\u6742\u5ea6\u548c\u6548\u679c":10,"\u6216\u5176\u4ed6":10,"\u6216\u8005":6,"\u6216\u800510g\u8fd9\u6837\u7684\u5355\u4f4d":0,"\u6216\u8005\u6570\u636e\u5e93\u8fde\u63a5\u5730\u5740\u7b49\u7b49":19,"\u6216\u8005\u8bbe\u7f6e\u4e3anone":19,"\u6216\u8005\u9700\u8981\u66f4\u9ad8\u7684\u6548\u7387":19,"\u6216\u8005\u9ad8\u6027\u80fd\u7684":6,"\u6240\u4ee5":[20,23],"\u6240\u4ee5\u5728cpu\u7684\u8fd0\u7b97\u6027\u80fd\u4e0a\u5e76\u4e0d\u4f1a\u6709\u4e25\u91cd\u7684\u5f71\u54cd":6,"\u6240\u4ee5\u5982\u679c\u60f3\u8981\u5728\u540e\u53f0\u542f\u7528ssh":6,"\u6240\u4ee5\u5c06":20,"\u6240\u4ee5\u7528\u4e8e\u9884\u6d4b\u7684\u914d\u7f6e\u6587\u4ef6\u8981\u505a\u76f8\u5e94\u7684\u4fee\u6539":23,"\u6240\u4ee5\u8f93\u51fa\u7684value\u5305\u542b\u4e24\u4e2a\u5411\u91cf":23,"\u6240\u4ee5gpu\u5728\u8fd0\u7b97\u6027\u80fd\u4e0a\u4e5f\u4e0d\u4f1a\u6709\u4e25\u91cd\u7684\u5f71\u54cd":6,"\u6240\u4ee5init_hook\u5c3d\u91cf\u4f7f\u7528":20,"\u6240\u6709\u5b57\u7b26\u90fd\u5c06\u8f6c\u6362\u4e3a\u8fde\u7eed\u6574\u6570\u8868\u793a\u7684id\u4f20\u7ed9\u6a21\u578b":10,"\u6240\u6709\u6587\u4ef6\u5217\u8868":20,"\u6240\u6709\u7684":1,"\u6240\u6709\u7684\u4e0b\u8f7d\u7ebf\u7a0b\u5173\u95ed\u65f6":0,"\u6240\u6709\u914d\u7f6e\u5728":10,"\u6240\u8c13\u65f6\u95f4\u6b65\u4fe1\u606f":20,"\u6253\u5370\u7684\u65e5\u5fd7\u53d8\u591a":1,"\u6267\u884c":0,"\u6267\u884c\u5982\u4e0b\u547d\u4ee4\u5373\u53ef\u4ee5\u5173\u95ed\u8fd9\u4e2acontain":6,"\u6267\u884c\u65b9\u6cd5\u5982\u4e0b":6,"\u62a5\u9519":7,"\u62fc\u63a5\u6210\u4e00\u4e2a\u65b0\u7684\u5411\u91cf\u8868\u793a":10,"\u6307\u4ee4\u96c6\u7684cpu\u8fd0\u884cpaddlepaddl":7,"\u6307\u4ee4\u96c6\u7684x86":7,"\u6307\u5b9a\u521d\u59cb\u5316\u6a21\u578b\u8def\u5f84":10,"\u6307\u5b9a\u751f\u6210\u6570\u636e\u7684\u51fd\u6570":10,"\u6307\u5b9a\u8bad\u7ec3":10,"\u6307\u5b9abatch":10,"\u6307\u5b9aoutputs\u8f93\u51fa\u6982\u7387\u5c42":10,"\u6389\u7f16\u8bd1\u76ee\u5f55\u540e":1,"\u63a5\u4e0b\u6765\u4f7f\u7528":23,"\u63a5\u53e3\u4f7f\u7528\u591a\u7ebf\u7a0b\u8bfb\u53d6\u6570\u636e":20,"\u63a8\u8350\u4f7f\u7528\u5c06\u672c\u5730\u7f51\u5361":6,"\u63a8\u8350\u4f7f\u7528\u6700\u65b0\u7248\u672c\u7684cudnn":1,"\u63a8\u8350\u6e05\u7406":1,"\u63a8\u8350\u76f4\u63a5\u653e\u7f6e\u5230\u8bad\u7ec3\u76ee\u5f55":19,"\u63cf\u8ff0":1,"\u63cf\u8ff0\u4e86docker":0,"\u6548\u679c\u4e00\u81f4":20,"\u6548\u679c\u603b\u7ed3":10,"\u6559\u7a0b":10,"\u6570\u636e":20,"\u6570\u636e\u4f20\u8f93\u65e0\u9700label\u6570\u636e":10,"\u6570\u636e\u5904\u7406python\u6587\u4ef6\u540d":10,"\u6570\u636e\u5982\u4f55\u5b58\u50a8\u7b49\u7b49":20,"\u6570\u636e\u63d0\u4f9b":19,"\u6570\u636e\u6587\u4ef6\u5b58\u653e\u5728\u672c\u5730\u78c1\u76d8\u4e2d":19,"\u6570\u636e\u662f\u7ed9\u4e00\u6bb5\u82f1\u6587\u6587\u672c":20,"\u6570\u636e\u683c\u5f0f\u548c\u8be6\u7ec6\u6587\u6863\u8bf7\u53c2\u8003":10,"\u6587\u4ef6\u4e2d":10,"\u6587\u672c\u4e2d\u7684\u5355\u8bcd\u7528\u7a7a\u683c\u5206\u9694":10,"\u6587\u672c\u4fe1\u606f\u5c31\u662f\u4e00\u4e2a\u5e8f\u5217":20,"\u6587\u672c\u5206\u7c7b\u95ee\u9898":10,"\u6587\u672c\u5377\u79ef\u5206\u4e3a\u4e09\u4e2a\u6b65\u9aa4":10,"\u6587\u672c\u751f\u6210":9,"\u65b0\u5199layer":11,"\u65b9\u4fbf\u8c03\u8bd5\u4f7f\u7528":13,"\u65b9\u4fbf\u90e8\u7f72\u5206\u53d1":13,"\u65e0\u9700label\u76f8\u5173\u7684\u5c42":10,"\u65f6\u5e8f\u6a21\u578b\u5373\u4e3arnn\u6a21\u578b":10,"\u65f6\u5e8f\u6a21\u578b\u5747\u4f7f\u7528\u8be5\u811a\u672c":10,"\u662f":1,"\u662f\u4e00\u4e2abatch":20,"\u662f\u4e00\u4e2apython\u7684":20,"\u662f\u4e00\u4e2aswig\u5c01\u88c5\u7684paddlepaddle\u5305":6,"\u662f\u4e0d\u662f\u5f88\u7b80\u5355\u5462":20,"\u662f\u4e2adataprovider\u662f\u4e0d\u662f\u8981\u505ashuffl":20,"\u662f\u4ec0\u4e48\u4e5f\u6ca1\u5173\u7cfb":20,"\u662f\u4ece\u8bad\u7ec3\u914d\u7f6e\u4f20\u5165\u7684dict\u5bf9\u8c61":20,"\u662f\u521d\u59cb\u5316\u65f6\u8c03\u7528\u7684\u51fd\u6570":20,"\u662f\u540c\u4e00\u4e2a\u5bf9\u8c61":20,"\u662f\u5426\u4f7f\u7528\u53cc\u7cbe\u5ea6\u6d6e\u70b9\u6570":1,"\u662f\u5426\u4f7f\u7528\u8fd0\u884c\u65f6\u52a8\u6001\u52a0\u8f7dcuda\u52a8\u6001\u5e93":1,"\u662f\u5426\u4f7f\u7528gflags":1,"\u662f\u5426\u4f7f\u7528glog":1,"\u662f\u5426\u5185\u5d4cpython\u89e3\u91ca\u5668":1,"\u662f\u5426\u5bfb\u627e\u5230cuda\u5de5\u5177\u94fe":1,"\u662f\u5426\u5f00\u542f\u5355\u5143\u6d4b\u8bd5":1,"\u662f\u5426\u5f00\u542f\u8ba1\u65f6\u529f\u80fd\u5f00\u542f\u8ba1\u65f6\u529f\u80fd\u4f1a\u5bfc\u81f4\u8fd0\u884c\u7565\u6162":1,"\u662f\u5426\u5f00\u542fgpu\u529f\u80fd":0,"\u662f\u5426\u5f00\u542frdma\u652f\u6301":1,"\u662f\u5426\u7f16\u8bd1\u4e2d\u6587\u6587\u6863":1,"\u662f\u5426\u7f16\u8bd1\u542b\u6709avx\u6307\u4ee4\u96c6\u7684paddlepaddle\u4e8c\u8fdb\u5236":1,"\u662f\u5426\u7f16\u8bd1\u65f6\u8fdb\u884c\u4ee3\u7801\u98ce\u683c\u68c0\u67e5":1,"\u662f\u5426\u7f16\u8bd1\u82f1\u6587\u6587\u6863":1,"\u662f\u5426\u7f16\u8bd1gpu\u652f\u6301":1,"\u662f\u5426\u7f16\u8bd1python\u7684swig\u63a5\u53e3":1,"\u662f\u5728\u8fd0\u884c\u65f6\u6267\u884c\u7684":20,"\u662f\u6570\u636e\u7f13\u5b58\u7684\u7b56\u7565":20,"\u662f\u6570\u636e\u8f93\u5165\u683c\u5f0f":20,"\u662f\u8bbe\u7f6e\u8fd9\u4e2adataprovider\u8fd4\u56de\u4ec0\u4e48\u6837\u7684\u6570\u636e":20,"\u662f\u8bbe\u7f6edataprovider\u5728\u5185\u5b58\u4e2d\u6682\u5b58\u7684\u6570\u636e\u6761\u6570":20,"\u662fpaddlepaddle\u8d1f\u8d23\u63d0\u4f9b\u6570\u636e\u7684\u6a21\u5757":19,"\u662fpython\u7684\u4e00\u4e2a\u5173\u952e\u8bcd":20,"\u663e":10,"\u66f4\u591a\u7684\u5b89\u88c5\u5305paddlepaddle\u4f1a\u5728\u8fd1\u671f\u63d0\u4f9b":7,"\u66f4\u8be6\u7ec6\u7528\u4f8b\u8bf7\u53c2\u8003\u6587\u6863":10,"\u66f4\u8be6\u7ec6\u7684\u4ecb\u7ecd\u8bf7\u53c2\u8003\u5404\u4e2a\u547d\u4ee4\u7684\u547d\u4ee4\u884c\u53c2\u6570\u6587\u6863":13,"\u66f4\u8be6\u7ec6\u7684\u7f51\u7edc\u914d\u7f6e":10,"\u6700\u4f4e\u7ebf\u7a0b\u7684\u4e0b\u8f7d\u901f\u5ea6":0,"\u6700\u540e\u4f7f\u7528":23,"\u6709\u503c\u7684\u4f4d\u7f6e\u53ea\u80fd\u53d61":20,"\u6709\u503c\u7684\u90e8\u5206\u53ef\u4ee5\u662f\u4efb\u4f55\u6d6e\u70b9\u6570":20,"\u6709\u90e8\u5206\u53c2\u6570\u662fpaddle\u81ea\u52a8\u751f\u6210\u7684":20,"\u672c\u8282\u6211\u4eec\u5c06\u4e13\u6ce8\u4e8e\u7f51\u7edc\u7ed3\u6784\u7684\u4ecb\u7ecd":10,"\u6765\u63a5\u53d7\u4e0d\u4f7f\u7528\u7684":20,"\u6765\u81ea\u5b9a\u4e49\u4f20\u6570\u636e\u7684\u8fc7\u7a0b":19,"\u6765\u8bf4\u660e\u7b80\u5355\u7684pydataprovider\u5982\u4f55\u4f7f\u7528":20,"\u6765\u8fdb\u884c\u8bad\u7ec3":6,"\u6765\u914d\u7f6ecudnn\u7684\u5b89\u88c5\u8def\u5f84":1,"\u6784\u9020gradientmachin":23,"\u6790\u597d\u7684\u914d\u7f6e\u521b\u5efa\u795e\u7ecf\u7f51\u7edc":23,"\u6807\u7b7e\u662f0":20,"\u6837\u4f8b\u6570\u636e\u4e3a":20,"\u6837\u4f8b\u6570\u636e\u5982\u4e0b":20,"\u6837\u672c":20,"\u6839\u636e\u4e0a\u4e00\u6b65\u89e3":23,"\u683c\u5f0f\u5982\u4e0b":10,"\u68d2":10,"\u6a21\u578b\u5b58\u50a8\u8def\u5f84":10,"\u6a21\u578b\u8bad\u7ec3\u4f1a\u770b\u5230\u8fd9\u6837\u7684\u65e5\u5fd7":10,"\u6a21\u578b\u914d\u7f6e":11,"\u6b22\u8fce\u5927\u5bb6\u8d21\u732e\u5404\u4e2a\u53d1\u884c\u7248\u7684\u5b89\u88c5\u5305":7,"\u6b63\u6837\u672c":10,"\u6bcf\u4e00\u4e2a\u4efb\u52a1\u6d41\u7a0b\u90fd\u53ef\u4ee5\u5206\u4e3a\u5982\u4e0b5\u4e2a\u57fa\u7840\u90e8\u5206":10,"\u6bcf\u4e00\u884c":20,"\u6bcf\u4e2a\u5c42\u90fd\u6709\u4e00\u4e2a\u6216\u591a\u4e2ainput":10,"\u6bcf\u4e2apass\u7684\u7b2c0\u4e2abatch\u5230\u5f53\u524dbatch\u6240\u6709\u6837\u672c\u7684\u5e73\u5747\u5206\u7c7b\u9519\u8bef\u7387":10,"\u6bcf\u4e2apass\u7684\u7b2c0\u4e2abatch\u5230\u5f53\u524dbatch\u6240\u6709\u6837\u672c\u7684\u5e73\u5747cost":10,"\u6bcf\u6b21\u90fd\u4f1a\u4ecepython\u7aef\u8bfb\u53d6\u6570\u636e":20,"\u6bcf\u884c\u4fdd\u5b58\u4e00\u6761\u6837\u672c":10,"\u6bcf\u9694\u591a\u5c11batch\u6253\u5370\u4e00\u6b21\u65e5\u5fd7":10,"\u6bd4\u5982\u901a\u8fc7\u7528\u6237\u5bf9\u7535\u5b50\u5546\u52a1\u7f51\u7ad9\u8bc4\u8bba":10,"\u6bd4\u8f83\u53ef\u80fd\u7684\u547d\u4ee4\u5982\u4e0b":7,"\u6ca1\u6709\u8bbe\u7f6e":7,"\u6ce8\u610f":[0,5,20],"\u6d41\u7a0b\u5982\u4e0b":10,"\u6d4b\u8bd5\u6570\u636e":10,"\u6d4b\u8bd5\u7684\u65f6\u5019\u9ed8\u8ba4\u4e0dshuffl":20,"\u6d4b\u8bd5\u811a\u672c\u5982\u4e0b":10,"\u6e90\u7801":10,"\u6e90\u7801\u548c\u8fd0\u884cdemo\u7684\u5fc5\u8981\u4f9d\u8d56":6,"\u6e90\u7801\u6839\u76ee\u5f55":0,"\u6fc0\u6d3b\u51fd\u6570\u7c7b\u578b":10,"\u7136\u540e\u6267\u884c\u5982\u4e0b\u547d\u4ee4\u5373\u53ef\u5b8c\u6210\u5b89\u88c5":7,"\u7136\u540e\u8fd0\u884c\u8fd9\u4e2acontainer\u5373\u53ef":6,"\u751f\u6210\u5404\u4e2a\u5e73\u53f0\u7684makefil":1,"\u75280\u548c1\u8868\u793a":20,"\u7528\u4e86\u4e24\u4e2a\u6708\u4e4b\u540e\u8fd9\u4e2a\u663e\u793a\u5668\u5c4f\u5e55\u788e\u4e86":10,"\u7528\u6237\u4e5f\u53ef\u4ee5\u5728c":19,"\u7528\u6237\u53ef\u4ee5\u4f7f\u7528python\u7684":19,"\u7528\u6237\u53ef\u4ee5\u6839\u636e\u8bad\u7ec3log\u9009\u62e9test\u7ed3\u679c\u6700\u597d\u7684\u6a21\u578b\u6765\u9884\u6d4b":10,"\u7528\u6237\u540d\u4e3a":6,"\u7528\u6237\u5728dataprovider\u4e2d\u9700\u8981\u5b9e\u73b0\u5982\u4f55\u8bbf\u95ee\u5176\u4e2d\u6bcf\u4e00\u4e2a\u6587\u4ef6":19,"\u7528\u6237\u5b9a\u4e49\u7684\u53c2\u6570\u4f7f\u7528args\u5728\u8bad\u7ec3\u914d\u7f6e\u4e2d\u8bbe\u7f6e":20,"\u7528\u6237\u63a5\u53e3":11,"\u7528\u6237\u9700\u8981\u5148\u5c06paddlepaddle\u5b89\u88c5\u5305\u4e0b\u8f7d\u5230\u672c\u5730":7,"\u7528\u6765\u505a\u9884\u6d4b\u548c\u7b80\u5355\u7684\u5b9a\u5236\u5316":6,"\u7531\u4e8e\u6570\u636e\u662f\u4e24\u6761":23,"\u7531\u4e8edocker\u662f\u57fa\u4e8e\u5bb9\u5668\u7684\u8f7b\u91cf\u5316\u865a\u62df\u65b9\u6848":6,"\u7531\u4e8epaddlepaddle\u7684docker\u955c\u50cf\u5e76\u4e0d\u5305\u542b\u4efb\u4f55\u9884\u5b9a\u4e49\u7684\u8fd0\u884c\u547d\u4ee4":6,"\u7531\u6613\u5230\u96be\u5c55\u793a4\u79cd\u4e0d\u540c\u7684\u7f51\u7edc\u914d\u7f6e":10,"\u7684":10,"\u7684\u5b89\u88c5\u6587\u6863":6,"\u7684\u6587\u6863":20,"\u7684\u65f6\u5019\u5982\u679c\u62a5\u4e00\u4e9b\u4f9d\u8d56\u672a\u627e\u5230\u7684\u9519\u8bef\u662f\u6b63\u5e38\u7684":7,"\u7684\u662f":20,"\u7684\u673a\u5668\u4e0a\u8fdb\u884c":0,"\u7684\u7f51\u6865\u6765\u8fdb\u884c\u7f51\u7edc\u901a\u4fe1":6,"\u7684demo\u5b66\u4e60\u5982\u4f55\u8fdb\u884c\u591a\u673a\u8bad\u7ec3":10,"\u7684docker\u53ef\u80fd\u7f3a\u4e4f":0,"\u7684matrix":23,"\u7684python\u5305\u662fpaddlepaddle\u7684\u8bad\u7ec3\u4e3b\u8981\u7a0b\u5e8f":6,"\u7684python\u5305\u6765\u505a\u914d\u7f6e\u6587\u4ef6\u89e3\u6790\u7b49\u5de5\u4f5c":6,"\u7684python\u9884\u6d4b\u8fc7\u7a0b":10,"\u76ee\u524dpaddlepaddle\u7684\u5b89\u88c5\u5305\u53ea\u652f\u6301":7,"\u76ee\u524dpaddlepaddle\u7684\u5b89\u88c5\u5305\u8fd8\u5904\u5728pr":5,"\u76ee\u5f55":10,"\u76ee\u5f55\u4e0b":0,"\u76f4\u63a5\u63d0\u53d6\u51fa\u795e\u7ecf\u7f51\u7edcoutput\u5c42\u7684\u8f93\u51fa\u7ed3\u679c":23,"\u76f8\u5173\u547d\u4ee4\u4e3a":6,"\u76f8\u5173\u7684\u6982":20,"\u76f8\u5bf9\u4e8epaddlepaddle\u7a0b\u5e8f\u8fd0\u884c\u65f6\u7684\u8def\u5f84":19,"\u77e5\u9053\u5982\u4f55\u4ece":20,"\u793a":10,"\u7a0b\u5e8f\u6216\u8005\u81ea\u5b9a\u4e49\u4e00\u4e2a\u542b\u6709\u542f\u52a8\u811a\u672c\u7684imag":6,"\u7aef\u81ea\u5b9a\u4e49\u4e00\u4e2a":19,"\u7b2c\u4e00\u4e2a\u53c2\u6570\u662f":20,"\u7b2c\u4e00\u4e2apass\u4f1a\u4ecepython\u7aef\u8bfb\u53d6\u6570\u636e":20,"\u7b2c\u4e00\u6bb5\u6570\u636e\u4e3a\u8fd9\u5f20\u56fe\u7247\u7684label":20,"\u7b2c\u4e8c\u4e2a\u53c2\u6570\u662ffilenam":20,"\u7b2c\u4e8c\u6bb5\u6570\u636e\u4e3a\u8fd9\u4e2a\u56fe\u7247\u7684\u50cf\u7d20\u503c":20,"\u7b80\u5355\u4f18\u5316":0,"\u7b80\u5355\u7684\u4f7f\u7528":19,"\u7b80\u5355\u7684\u4f7f\u7528\u573a\u666f":19,"\u7b80\u5355\u7684\u4f7f\u7528\u6837\u4f8b\u4e3a":0,"\u7b80\u5355\u7684\u542b\u6709ssh\u7684dockerfile\u5982\u4e0b":6,"\u7b80\u5355\u7684pydataprovider\u6837\u4f8b\u5c31\u8bf4\u660e\u5b8c\u6bd5\u4e86":20,"\u7c7b\u522bid":10,"\u7c7b\u522bid\u7684\u6570\u636e\u7c7b\u578b":10,"\u7c7b\u578b\u6765\u8bbe\u7f6e":20,"\u7ec3":13,"\u7ed3\u4e0a\u8ff0\u7f51\u7edc\u7ed3\u6784\u5728amazon":10,"\u7ef4\u5ea6\u4e3aword":10,"\u7ef4\u5ea6\u662f\u7c7b\u522b\u4e2a\u6570":10,"\u7ef4\u5ea6\u662f\u8bcd\u5178\u5927\u5c0f":10,"\u7f13\u5b58\u8bad\u7ec3\u6570\u636e\u5230\u5185\u5b58":20,"\u7f16\u8bd1\u53c2\u6570\u9009\u9879\u6587\u4ef6":18,"\u7f16\u8bd1\u6587\u6863":7,"\u7f16\u8bd1\u9009\u9879":1,"\u7f16\u8bd1\u9009\u9879\u5217\u8868\u5982\u4e0b":1,"\u7f16\u8bd1paddlepaddl":7,"\u7f16\u8bd1paddlepaddle\u7684gpu\u7248\u672c\u5e76\u4e0d\u9700\u8981\u4e00\u5b9a\u5728\u5177\u6709gpu":0,"\u7f51\u7edc\u540d\u79f0":10,"\u7f51\u7edc\u914d\u7f6e":10,"\u7f6e\u8fd9\u4e9b\u53d8\u91cf":1,"\u800c":6,"\u800c\u4e09\u79cd\u5e8f\u5217\u6a21\u5f0f\u4e3a":20,"\u800c\u4e0d\u4f7f\u7528docker":6,"\u800c\u4e0d\u7528\u5173\u5fc3\u6570\u636e\u5982\u4f55\u4f20\u8f93\u7ed9paddlepaddl":20,"\u800c\u4e14\u9884\u6d4b\u7f51\u7edc\u901a\u5e38\u76f4\u63a5\u8f93\u51fa\u6700\u540e\u4e00\u5c42\u7684\u7ed3\u679c\u800c\u4e0d\u662f\u50cf\u8bad\u7ec3\u65f6\u4e00\u6837\u4ee5cost":23,"\u800c\u5728":[1,20],"\u800c\u6ca1\u6709\u6d4b\u8bd5\u6570\u636e":20,"\u800c\u7279\u5f81\u5373\u4e3a":20,"\u800c\u8fd9\u4e2a\u4e00\u822c\u8bf4\u660epaddlepaddle\u5df2\u7ecf\u5b89\u88c5\u5b8c\u6bd5\u4e86":7,"\u800c\u975e\u9759\u6001\u52a0\u8f7dcuda\u52a8\u6001\u5e93":1,"\u800cgpu\u7684\u9a71\u52a8\u548c\u8bbe\u5907\u5168\u90e8\u6620\u5c04\u5230\u4e86\u5bb9\u5668\u5185":6,"\u800cpaddlepaddle\u8fdb\u7a0b\u5e2e\u52a9\u7528\u6237\u505a\u4e86":20,"\u811a\u672c":6,"\u811a\u672c\u53ef\u4ee5\u542f\u52a8paddlepaddle\u7684\u8bad\u7ec3\u8fdb\u7a0b\u548cpserv":6,"\u811a\u672c\u548c":6,"\u811a\u672c\u7c7b\u4f3c\u4e8e":6,"\u81ea\u5b9a\u4e49\u4e00\u4e2adataprovid":19,"\u81f3\u5c11\u5177\u67093":6,"\u81f3\u6b64":[6,20],"\u83b7\u53d6\u5229\u7528one":10,"\u83b7\u53d6\u6bcf\u4e2a\u5355\u8bcd\u5de6\u53f3\u5404k\u4e2a\u8fd1\u90bb":10,"\u83b7\u53d6\u8be5\u6761\u6837\u672c\u7c7b\u522bid":10,"\u8868\u793a\u6574\u6570\u6807\u7b7e":20,"\u8868\u793a\u662f\u5426\u5141\u8bb8paddle\u6682\u5b58\u7565\u5fae\u591a\u4f59pool_size\u7684\u6570\u636e":20,"\u8868\u793a\u7a00\u758f\u7684\u5411\u91cf":20,"\u8868\u793a\u7a00\u758f\u7684\u96f6\u4e00\u5411\u91cf":20,"\u8868\u793a\u7a20\u5bc6\u7684\u6d6e\u70b9\u6570\u5411\u91cf":20,"\u8868\u793a\u8fc7\u4e8620\u4e2abatch":10,"\u8868\u793a\u8fc7\u4e862560\u4e2a\u6837\u672c":10,"\u8868\u793a\u8fd9\u4e2adataprovider\u662f\u8bad\u7ec3\u7528\u7684dataprovider\u6216\u8005\u6d4b\u8bd5\u7528\u7684":20,"\u89e3\u51b3\u529e\u6cd5\u662f\u5c06cuda":7,"\u89e3\u51b3\u65b9\u6cd5\u5f88\u7b80\u5355":7,"\u89e3\u6790\u8bad\u7ec3\u65f6\u7684\u914d\u7f6e\u6587\u4ef6":23,"\u89e3\u91ca":10,"\u8ba9\u795e\u7ecf\u7f51\u7edc\u53ef\u4ee5\u8fdb\u884c\u8bad\u7ec3":19,"\u8bad\u7ec3":6,"\u8bad\u7ec3\u6587\u4ef6\u5217\u8868":19,"\u8bad\u7ec3\u65f6\u6240\u9700\u8bbe\u7f6e\u7684\u4e3b\u8981\u53c2\u6570\u5982\u4e0b":10,"\u8bad\u7ec3\u7684\u65f6\u5019\u9ed8\u8ba4shuffl":20,"\u8bad\u7ec3\u811a\u672c":10,"\u8bad\u7ec3\u811a\u672c\u5728":10,"\u8bad\u7ec3\u8f6e\u6b21":10,"\u8bb2\u6570\u636e\u53d1\u9001\u7ed9paddlepaddl":20,"\u8bbe\u7f6e\u4e0b\u5217\u7f16\u8bd1\u9009\u9879\u65f6":1,"\u8bbe\u7f6e\u6210":20,"\u8bbe\u7f6e\u6210\u4e86\u5e8f\u5217":20,"\u8bbe\u7f6e\u8f93\u5165\u7c7b\u578b":20,"\u8bc4\u4f30\u4ea7\u54c1\u7684\u8d28\u91cf":10,"\u8bcd\u6027\u6807\u6ce8":9,"\u8be5\u5c42\u795e\u7ecf\u5143\u4e2a\u6570":10,"\u8be5\u6570\u636e":20,"\u8be5\u6784\u5efa\u811a\u672c\u5145\u5206\u8003\u8651\u4e86\u7f51\u7edc\u4e0d\u7a33\u5b9a\u7684\u60c5\u51b5":0,"\u8be5\u6a21\u578b\u4f9d\u7136\u662f\u4f7f\u7528\u903b\u8f91\u56de\u5f52\u5206\u7c7b\u7f51\u7edc\u7684\u6846\u67b6":10,"\u8be5\u76ee\u5f55\u4e0b\u6709\u4e24\u4e2a\u6587\u4ef6":0,"\u8be5\u811a\u672c\u7684\u4f7f\u7528\u65b9\u6cd5\u662f":0,"\u8be5image\u57fa\u4e8eubuntu":0,"\u8be5image\u7684\u6784\u5efa\u5728dock":0,"\u8be6\u60c5\u8bf7\u53c2\u8003":23,"\u8be6\u7ec6\u7684\u53c2\u6570\u89e3\u91ca\u5982\u4e0b\u9762\u8868\u683c":10,"\u8be6\u7ec6\u7684\u547d\u4ee4\u884c\u53c2\u6570\u8bf7\u53c2\u8003":23,"\u8be6\u7ec6\u7684cmake\u4f7f\u7528\u65b9\u6cd5\u53ef\u4ee5\u53c2\u8003":1,"\u8bf4\u660e":1,"\u8bf4\u660e\u547d\u4ee4\u884c\u53c2\u6570":7,"\u8bf7\u53c2\u8003":[6,7,20],"\u8bf7\u53c2\u8003\u4e0b\u8282refer":20,"\u8bf7\u53c2\u8003\u4e0b\u8ff0\u6587\u7ae0":19,"\u8bf7\u5b89\u88c5cuda":7,"\u8bfb\u5165\u89e3\u6790\u8bad\u7ec3\u914d\u7f6e":23,"\u8bfb\u53d6\u6570\u636e":20,"\u8c03\u7528":1,"\u8d1f\u6837\u672c":10,"\u8d1f\u8d23\u591a\u673a\u8bad\u7ec3\u4e2d\u7684\u53c2\u6570\u805a\u5408\u5de5\u4f5c":13,"\u8d1f\u9762\u60c5\u7eea\u4e24\u7c7b":20,"\u8def\u5f84\u53d8\u91cf\u4e3a":1,"\u8f93\u5165n\u4e2a\u5355\u8bcd":10,"\u8f93\u51fa\u4e3an\u4e2aword":10,"\u8fd0\u884c":[6,7],"\u8fd0\u884c\u4f7f\u7528\u7684cudnn\u5c3d\u91cf\u662f\u540c\u4e00\u4e2a\u7248\u672c":1,"\u8fd0\u884c\u8fd9\u4e2acontain":6,"\u8fd0\u884cpaddlepaddle\u7684gpu\u7248\u672c\u4e00\u5b9a\u8981\u5728\u5177\u6709cuda\u7684\u673a\u5668\u4e0a\u8fd0\u884c":0,"\u8fd4\u56de\u6570\u636e\u5728paddlepaddle\u4e2d\u662f\u4ec5\u4ec5\u8fd4\u56de\u4e00\u6761\u5b8c\u6574\u7684\u8bad\u7ec3\u6837\u672c":20,"\u8fd4\u56de\u7684\u987a\u5e8f\u9700\u8981\u548c":20,"\u8fd4\u56debatch_size\u7684\u5927\u5c0f":20,"\u8fd9":10,"\u8fd93\u4e2a\u5b50\u6b65\u9aa4\u53ef\u914d\u7f6e\u4e3a":10,"\u8fd9\u4e2a\u51fd\u6570\u4ee5\u4e00\u6761\u6570\u636e\u4e3a\u53c2\u6570":20,"\u8fd9\u4e2a\u51fd\u6570\u4f1a\u5728":20,"\u8fd9\u4e2a\u51fd\u6570\u5728\u521d\u59cb\u5316\u7684\u65f6\u5019\u4f1a\u88ab\u8c03\u7528":20,"\u8fd9\u4e2a\u51fd\u6570\u7684\u53c2\u6570\u662f":20,"\u8fd9\u4e2a\u521d\u59cb\u5316\u51fd\u6570\u5177\u6709\u5982\u4e0b\u53c2\u6570":20,"\u8fd9\u4e2a\u53c2\u6570\u5728\u8fd9\u4e2a\u6837\u4f8b\u91cc\u6ca1\u6709\u4f7f\u7528":20,"\u8fd9\u4e2a\u53c2\u6570\u88abpaddlepaddle\u8fdb\u7a0b\u4f20\u5165":20,"\u8fd9\u4e2a\u548c\u5728":20,"\u8fd9\u4e2a\u58f0\u660e\u57fa\u672c\u4e0a\u548cmnist\u7684\u6837\u4f8b\u4e00\u81f4":20,"\u8fd9\u4e2a\u5b57\u5178\u53ef\u4ee5\u5728":20,"\u8fd9\u4e2a\u5bf9\u8c61\u548cprocess\u7684\u7b2c\u4e00\u4e2a\u53c2\u6570\u4e00\u81f4":20,"\u8fd9\u4e2a\u5de5\u5177\u7c7b\u63a5\u6536\u548cpydataprovider2\u4e00\u6837\u7684\u8f93\u5165\u6570\u636e":23,"\u8fd9\u4e2a\u5e8f\u5217\u6a21\u578b\u6bd4\u8f83\u590d\u6742":20,"\u8fd9\u4e2a\u63a5\u53e3\u5e76\u4e0d\u7528\u6237\u53cb\u597d":23,"\u8fd9\u4e2a\u663e\u793a\u5668\u5f88\u68d2":10,"\u8fd9\u4e2a\u6a21\u5757\u4e2d\u7684":20,"\u8fd9\u4e2a\u8bbe\u7f6e\u4e3a":20,"\u8fd9\u4e2a\u8f6f\u4ef6\u5305\u6587\u6863\u76f8\u5bf9\u5b8c\u5584":23,"\u8fd9\u4e9b\u53c2\u6570\u5305\u62ecpaddle\u5b9a\u4e49\u7684\u53c2\u6570":20,"\u8fd9\u4e9b\u53d8":1,"\u8fd9\u4e9b\u53d8\u91cf\u53ea\u5728\u7b2c\u4e00\u6b21cmake\u7684\u65f6\u5019\u6709\u6548":1,"\u8fd9\u4e9b\u53d8\u91cf\u5747\u53ef\u4ee5\u4f7f\u7528":1,"\u8fd9\u4e9b\u6d41\u7a0b\u4e2d\u7684\u6570\u636e\u4e0b\u8f7d":10,"\u8fd9\u6837\u505a\u53ef\u4ee5\u907f\u514d\u5f88\u591a\u6b7b\u9501\u95ee\u9898":20,"\u8fd9\u884c\u7684\u4f5c\u7528\u662f\u8bbe\u7f6edataprovider\u7684\u4e00\u4e9b\u5c5e\u6027":20,"\u8fd9\u91cc":20,"\u8fd9\u91cc\u4e3e\u4f8b\u7684\u6570\u636e\u662f\u82f1\u6587\u60c5\u611f\u5206\u7c7b\u7684\u6570\u636e":20,"\u8fd9\u91cc\u4ee5":10,"\u8fd9\u91cc\u4ee5mnist\u624b\u5199\u8bc6\u522b\u4e3a\u4f8b":20,"\u8fd9\u91cc\u53ef\u4ee5\u53c2\u8003paddle\u7684":18,"\u8fd9\u91cc\u6211\u4eec\u4f7f\u7528\u7b80\u5355\u7684\u6587\u672c\u6587\u4ef6\u8868\u793amnist\u56fe\u7247":20,"\u8fd9\u91cc\u6307\u5b9a\u8bcd\u5178":10,"\u8fd9\u91cc\u6ca1\u6709\u4ecb\u7ecd\u591a\u673a\u5206\u5e03\u5f0f\u8bad\u7ec3":10,"\u8fd9\u91cc\u7684":20,"\u8fd9\u91cc\u7684\u8f93\u5165\u7279\u5f81\u662f\u8bcdid\u7684\u5e8f\u5217":20,"\u8fd9\u91cc\u8981\u6ce8\u610f\u9884\u6d4b\u6570\u636e\u901a\u5e38":23,"\u8fd9\u91cc\u8bbe\u7f6e\u7684\u662f\u8fd4\u56de\u4e00\u4e2a":20,"\u8fd9\u91cc\u8bf4\u660e\u4e86\u8bad\u7ec3\u6570\u636e\u662f":20,"\u8fd9\u91cc\u91c7\u7528adam\u4f18\u5316\u65b9\u6cd5":10,"\u8fdb\u5165\u8be5\u6e90\u7801\u76ee\u5f55":0,"\u8fdb\u5165docker":6,"\u8fdc\u7a0b\u8bbf\u95ee":6,"\u8fde\u63a5\u8bf7\u53c2\u8003":10,"\u9009\u9879":1,"\u901a\u8fc7\u7f16\u8bd1\u65f6\u6307\u5b9a\u8def\u5f84\u6765\u5b9e\u73b0\u5f15\u7528\u5404\u79cdbla":1,"\u903b\u8f91\u56de\u5f52":10,"\u90a3\u4e48":20,"\u90a3\u4e48\u5728\u8bad\u7ec3\u8fc7\u7a0b\u4e2d":19,"\u90a3\u4e48\u5bf9\u5e94\u7684dataprovider\u65e2\u4e3a":20,"\u90a3\u4e48\u8fd9\u4e2a\u4e0b\u8f7d\u7ebf\u7a0b\u5c06\u4f1a\u5173\u95ed":0,"\u90a3\u4e48\u9700\u8981\u4ece\u6e90\u7801":7,"\u914d\u7f6e\u4e86":20,"\u914d\u7f6e\u53c2\u6570\u914d\u7f6e\u7ed9dataprovider\u7684":20,"\u914d\u7f6e\u6587\u4ef6":10,"\u91cc\u4f1a\u7ee7\u7eed\u5b89\u88c5":7,"\u91cc\u63d0\u4f9b\u4e86\u6570\u636e\u4e0b\u8f7d\u811a\u672c":10,"\u91cc\u9762\u8bfb\u53d6":20,"\u91cf\u4e5f\u53ef\u4ee5\u901a\u8fc7\u8c03\u7528cmake\u547d\u4ee4\u524d\u901a\u8fc7\u73af\u5883\u53d8\u91cf\u6307\u5b9a":1,"\u9488\u5bf9\u672c\u95ee\u9898":10,"\u94fe\u63a5\u4f55\u79cdblas\u7b49\u7b49":1,"\u9519\u8bef\u7387":10,"\u95f4\u9694":20,"\u9664\u4e86":20,"\u9664\u8fc7data\u5c42":10,"\u9700\u8981\u53c2\u8003":6,"\u9700\u8981\u6ce8\u610f":20,"\u9700\u8981\u6ce8\u610f\u7684\u662f":[1,7],"\u9884\u6d4b\u6570\u636e\u6307\u5b9atest":10,"\u9884\u6d4b\u7ed3\u679c\u4ee5\u6587\u672c\u7684\u5f62\u5f0f\u4fdd\u5b58\u5728":10,"\u9884\u6d4b\u811a\u672c":10,"\u9884\u6d4bid":10,"\u989d\u5916\u7684\u53c2\u6570":10,"\u9996\u5148\u5217\u4e3e\u903b\u8f91\u56de\u5f52\u7f51\u7edc":10,"\u9996\u5148\u6211\u4eec\u5c06\u8fd9\u4e2a\u6570\u636e\u6587\u4ef6":20,"\u9996\u5148\u8bf7\u53c2\u8003":10,"\u9aa43":10,"\u9ed8\u8ba4\u4e00\u4e2apass\u4fdd\u5b58\u4e00\u6b21\u6a21\u578b":10,"\u9ed8\u8ba4\u503c":1,"\u9ed8\u8ba4\u5355\u4f4d\u662fbyte":0,"\u9ed8\u8ba4\u60c5\u51b5\u4e0b\u4e00\u6761\u6570\u636e":20,"adamax\u7b49":10,"amazon\u7535\u5b50\u4ea7\u54c1\u8bc4\u8bba\u6570\u636e":10,"api\u9884\u6d4b":10,"argument\u4f20\u5165":20,"argument\u5f62\u5f0f\u4f20\u5165":20,"atlas\u5e93\u7684\u8def\u5f84":1,"batches\u8bbe\u7f6e\u6bcf\u9694\u591a\u5c11batch\u4fdd\u5b58\u4e00\u6b21\u6a21\u578b":10,"bool\u53c2\u6570":20,"case":[10,22],"cd\u5230\u542b\u6709dockerfile\u7684\u8def\u5f84\u4e2d":6,"cmake\u53ef\u4ee5\u5c06cmake\u9879\u76ee\u6587\u4ef6":1,"cmake\u662f\u4e00\u4e2a\u8de8\u5e73\u53f0\u7684\u7f16\u8bd1\u811a\u672c":1,"cmake\u7684\u5b98\u65b9\u6587\u6863":1,"cmake\u7f16\u8bd1\u65f6\u4f1a\u9996\u5148\u5728\u7cfb\u7edf\u8def\u5f84":1,"container\u540e":6,"cuda\u76f8\u5173\u7684driver\u548c\u8bbe\u5907\u6620\u5c04\u8fdbcontainer\u4e2d":6,"d\u547d\u4ee4\u5373\u53ef":1,"d\u547d\u4ee4\u6307\u5b9a":1,"dataprovider\u521b\u5efa\u7684\u65f6\u5019\u6267\u884c":20,"dataprovider\u53ef\u4ee5\u662f":20,"dataprovider\u63d0\u4f9b\u4e86\u4e24\u79cd\u7b80\u5355\u7684cache\u7b56\u7565":20,"dataprovider\u7684\u5177\u4f53\u7528\u6cd5\u548c\u5982\u4f55\u5b9e\u73b0\u4e00\u4e2a\u65b0\u7684dataprovid":19,"dim\u7684\u65b0\u7684\u5411\u91cf":10,"dim\u7ef4\u5ea6\u5411\u91cf":10,"docker\u662f\u4e00\u4e2a\u57fa\u4e8e\u5bb9\u5668\u7684\u8f7b\u91cf\u7ea7\u865a\u62df\u73af\u5883":6,"docker\u7684\u5b98\u65b9\u6587\u6863":6,"dockerfile\u548cbuild":0,"dockerfile\u662fdock":0,"dockerfile\u7684\u6587\u6863":6,"dockerfile\u7684\u6700\u4f73\u5b9e\u8df5":6,"driver\u6dfb\u52a0\u5230ld_library_path\u4e2d":7,"elec\u6d4b\u8bd5\u96c6":10,"embedding\u6a21\u578b\u9700\u8981\u7a0d\u5fae\u6539\u53d8\u6570\u636e\u63d0\u4f9b\u7684\u811a\u672c":10,"export":[1,6,7],"f\u4ee3\u8868\u4e00\u4e2a\u6d6e\u70b9\u6570":20,"float":20,"gpu\u53cc\u7f13\u5b58":20,"gpu\u7248\u672c\u4e8c\u8fdb\u5236":1,"gru\u6a21\u578b":10,"gru\u6a21\u578b\u914d\u7f6e":10,"i\u4ee3\u8868\u4e00\u4e2a\u6574\u6570":20,"id\u4e3a0\u7684\u6982\u7387":10,"id\u4e3a1\u7684\u6982\u7387":10,"image\u6784\u5efa\u6e90\u7801\u653e\u7f6e\u5728":0,"image\u7684\u4e3b\u8981\u63cf\u8ff0\u6587\u4ef6":0,"image\u7684\u4e3b\u8981\u6784\u5efa\u6b65\u9aa4":0,"image\u7684\u6784\u5efa\u6b65\u9aa4":0,"import":[10,20,23],"include\u4e0b\u9700\u8981\u5305\u542bcbla":1,"include\u4e0b\u9700\u8981\u5305\u542bmkl":1,"init_hook\u53ef\u4ee5\u4f20\u5165\u4e00\u4e2a\u51fd\u6570":20,"int":[10,20],"layer\u4f5c\u4e3a\u8f93\u51fa":23,"layer\u6587\u6863":10,"ld_library_path\u7b49\u7b49":7,"ld_library_path\u91cc\u9762\u627e\u4e0d\u5230\u8fd9\u4e9b\u52a8\u6001":7,"lib\u4e0b\u9700\u8981\u5305\u542bcblas\u548catlas\u4e24\u4e2a\u5e93":1,"lib\u4e0b\u9700\u8981\u5305\u542bcblas\u5e93":1,"lib\u4e0b\u9700\u8981\u5305\u542bopenblas\u5e93":1,"lib\u76ee\u5f55\u4e0b\u9700\u8981\u5305\u542b":1,"list\u4e0d\u8bbe\u7f6e":19,"list\u4e2d":19,"list\u4e2d\u7684\u4e00\u884c":20,"list\u4e3a\u7eaf\u6587\u672c\u6587\u4ef6":19,"list\u4e5f\u53ef\u4ee5\u653e\u7f6ehdfs\u6587\u4ef6\u8def\u5f84":19,"list\u5373\u4e3a":20,"list\u548ctest":19,"list\u5747\u4e3a\u672c\u5730\u7684\u4e24\u4e2a\u6587\u4ef6":19,"list\u6307\u5b9a\u7684\u6570\u636e":10,"list\u7684\u4f4d\u7f6e":10,"list\u82e5\u5e72\u6570\u636e\u6587\u4ef6\u8def\u5f84\u7684\u67d0\u4e00\u4e2a\u8def\u5f84":20,"lstm\u6a21\u578b\u7b49":10,"lstm\u6a21\u578b\u914d\u7f6e":10,"make\u548cmak":2,"mkl\u7684\u8def\u5f84":1,"mkl_sequential\u548cmkl_intel_lp64\u4e09\u4e2a\u5e93":1,"mnist\u662f\u4e00\u4e2a\u5305\u542b\u6709":20,"movielens\u6570\u636e\u96c6":9,"movielens\u8bc4\u5206\u56de\u5f52":9,"osx\u6216\u8005\u662fwindows\u673a\u5668":6,"osx\u7684\u5b89\u88c5\u6587\u6863":6,"paddle\u5b9a\u4e49\u7684\u53c2\u6570\u5305\u62ec":20,"paddlepaddle\u4f7f\u7528\u8fd0\u884c\u65f6\u52a8\u6001\u8fde\u63a5cuda\u7684so":7,"paddlepaddle\u4fdd\u7559\u6dfb\u52a0\u53c2\u6570\u7684\u6743\u529b":20,"paddlepaddle\u53ef\u4ee5\u4f7f\u7528":1,"paddlepaddle\u53ef\u4ee5\u8bfb\u53d6python\u5199\u7684\u4f20\u8f93\u6570\u636e\u811a\u672c":10,"paddlepaddle\u5728\u8fd0\u884c\u65f6\u627e\u4e0d\u5230\u5bf9\u5e94\u7684config\u6587\u4ef6":7,"paddlepaddle\u63a8\u8350\u4f7f\u7528docker\u8fdb\u884cpaddlepaddle\u7684\u90e8\u7f72\u548c":6,"paddlepaddle\u63d0\u4f9b\u4e86docker\u7684\u4f7f\u7528\u955c\u50cf":6,"paddlepaddle\u63d0\u4f9b\u6570\u4e2a\u9884\u7f16\u8bd1\u7684\u4e8c\u8fdb\u5236\u6765\u8fdb\u884c\u5b89\u88c5":5,"paddlepaddle\u63d0\u4f9b\u7684\u955c\u50cf\u5e76\u4e0d\u5305\u542b\u4efb\u4f55\u547d\u4ee4\u8fd0\u884c":6,"paddlepaddle\u7684\u6570\u636e\u5305\u62ec\u56db\u79cd\u4e3b\u8981\u7c7b\u578b":20,"paddlepaddle\u7684\u7f16\u8bd1\u9009\u9879\u53ef\u4ee5\u5728\u8c03\u7528cmake\u7684\u65f6\u5019\u8bbe\u7f6e":1,"paddlepaddle\u7684\u7f16\u8bd1\u9009\u9879\u662f\u53ef\u4ee5\u63a7\u5236paddlepaddle\u751f\u6210cpu":1,"paddlepaddle\u7684cpu\u4e8c\u8fdb\u5236":6,"paddlepaddle\u7684dock":0,"paddlepaddle\u7684gpu\u4e8c\u8fdb\u5236":6,"paddlepaddle\u7684python\u9884\u6d4b\u63a5\u53e3":22,"paddlepaddle\u7684ubuntu\u5b89\u88c5\u5305\u5206\u4e3a\u4e24\u4e2a\u7248\u672c":7,"paddlepaddle\u76ee\u524d\u4f7f\u7528swig\u5bf9\u5176\u5e38\u7528\u7684\u9884\u6d4b\u63a5\u53e3\u8fdb\u884c\u4e86\u5c01\u88c5":23,"paddlepaddle\u76ee\u524d\u652f\u6301ubuntu":7,"paddlepaddle\u8fd0\u884c\u65f6\u5982\u679c\u6ca1\u6709\u5bfb\u627e\u5230cuda\u7684driv":7,"paddlepaddle\u9700\u8981\u7528\u6237\u5728\u7f51\u7edc\u914d\u7f6e":19,"period\u8bbe\u7f6e\u6253\u5370\u53c2\u6570\u4fe1\u606f\u7b49":10,"process\u51fd\u6570":20,"process\u51fd\u6570\u662f\u5b9e\u73b0\u6570\u636e\u8f93\u5165\u7684\u4e3b\u51fd\u6570":20,"process\u51fd\u6570\u8c03\u7528\u591a\u6b21":20,"pserver\u4e3apaddlepaddle\u7684paramet":13,"pserver\u7684\u547d\u4ee4\u884c\u53c2\u6570":13,"pserver\u7ec4\u5408\u4f7f\u7528":13,"py\u6587\u4ef6\u7ed9\u51fa\u4e86\u5b8c\u6574\u4f8b\u5b50":10,"pydataprovider2\u6587\u6863":23,"pydataprovider2\u7684\u4f7f\u7528":19,"pydataprovider\u662fpaddlepaddle\u4f7f\u7528python\u63d0\u4f9b\u6570\u636e\u7684\u63a8\u8350\u63a5\u53e3":20,"python\u5305":6,"python\u7684":6,"python\u7684swig\u63a5\u53e3\u53ef\u4ee5\u65b9\u4fbf\u8fdb\u884c\u9884\u6d4b\u548c\u5b9a\u5236\u5316\u8bad\u7ec3":1,"release\u7684\u72b6\u6001":5,"return":10,"rnn\u914d\u7f6e":11,"server\u8fdb\u7a0b":13,"sh\u662fdocker":0,"shuffle\u8bad\u7ec3\u6570\u636e":20,"softmax\u8f93\u51fa":10,"string\u7684\u683c\u5f0f\u6253\u5370\u51fa\u6765":13,"swig_paddle\u63a5\u53d7\u7684\u539f\u59cb\u6570\u636e\u662fc":23,"train\u5373\u4e3apaddlepaddle\u7684\u8bad\u7ec3\u8fdb\u7a0b":13,"train\u5b8c\u6210\u5355\u673a\u591a\u663e\u5361\u591a\u7ebf\u7a0b\u7684\u8bad":13,"train\u7684\u547d\u4ee4\u884c\u53c2\u6570":13,"ubuntu\u7684deb\u5b89\u88c5\u5305\u7b49":5,"v2\u4e4b\u540e\u7684\u4efb\u4f55\u4e00\u4e2acudnn\u7248\u672c\u6765\u7f16\u8bd1\u8fd0\u884c":1,"value\u5373\u4e3asoftmax\u5c42\u7684\u8f93\u51fa":23,"var":6,"vector\u8868\u793a\u7684\u6bcf\u4e2a\u5355\u8bcd":10,"version\u53ef\u4ee5\u6253\u5370\u51fapaddle\u7684\u7248\u672c\u4fe1\u606f\u548c\u7f16\u8bd1\u7684\u9009\u9879":18,"version\u53ef\u4ee5\u6253\u5370\u51fapaddlepaddle\u7684\u7248\u672c\u548c\u7f16\u8bd1\u65f6\u4fe1\u606f":13,"version\u7684\u547d\u4ee4\u884c\u53c2\u6570":13,"yield\u6587\u672c\u4fe1\u606f\u548c\u7c7b\u522bid":10,__main__:23,__name__:23,abov:20,act:10,act_typ:10,activ:10,adadelta:10,adagrad:10,adam:10,adamoptim:10,afi:20,all:20,allow:10,alreadi:7,also:10,append:20,apt:[6,7],arg:[0,10,20],around:20,arrai:23,assert:23,atla:1,atlas_root:1,avg:10,avgcost:10,avgpool:10,avx:7,bag:10,baidu:[6,7],batch:10,batch_siz:10,binari:10,bla:1,bool:10,both:10,bow:10,build:[0,6],cach:[10,19],cache_pass_in_mem:[10,20],cachetyp:[10,20],calc_batch_s:20,call:10,can:10,can_over_batch_s:20,categori:10,cento:7,check:7,chines:9,chpasswd:6,classif:10,classification_cost:10,classification_error_evalu:10,close:20,cmake:1,cmd:6,cnn:10,code:[0,20,23],com:[6,7],comment:10,compil:18,conf:23,config:[7,10],config_arg:10,config_pars:23,connect:10,contain:[10,20],context_len:10,context_start:10,convert:[10,20,23],couldn:7,cpp:[7,10],cpu:[6,7,20],createfromconfigproto:23,cross:10,cuda_so:6,cudastat:7,cudasuccess:7,cudnn:1,cudnn_root:1,cudnnv5:1,current:[10,20],currentcost:10,currentev:10,dalla:20,data:7,data_config:23,data_initialz:10,data_lay:10,dataprovid:10,dataprovider_bow:10,dataprovider_emb:10,dataproviderconvert:23,dataset:10,deb:7,debian:7,decor:20,def:[10,20,23],defin:[10,20],define_py_data_sources2:[10,20],delar:10,demo:[6,10],dense_vector:[20,23],describ:10,descript:22,detail:22,dev:6,devel:6,devic:6,devices:6,dict:[10,20],dict_fil:10,dictionai:10,dictionari:[10,20],dictrionari:10,differ:10,dim:10,dimens:10,dir:10,doc:23,documentari:20,dpkg:7,driver:7,dso_handl:7,dtype:23,dump_config:13,dure:[10,20],dynam:20,each:[10,20],each_pixel_str:20,each_word:20,echo:6,either:10,els:10,emb:10,embed:9,embedding_lay:10,entropi:10,enumer:10,error:[7,10],etc:6,eval:10,exampl:10,experi:5,expose:6,f0831:7,fail:7,fals:10,fc_layer:10,featur:[10,20],festiv:20,file:[10,20],file_list:20,file_nam:10,filenam:20,fill:10,find:7,first:10,float32:23,fly:10,forwardtest:23,framework:10,from:[6,10,20,23],fulli:10,gener:[10,20],gentoo:7,get:[6,7,10,20],get_config_arg:10,get_data:10,github:7,give:20,given:10,globe:20,gpu:[6,7],gradient_clipping_threshold:10,gradientmachin:23,gru:10,gru_siz:10,help:23,hint:23,hl_cuda_devic:7,hl_dso_load:7,host:6,hot:10,hous:20,http:7,ignor:20,imag:6,imagenet:9,inarg:23,includ:10,init:10,init_hook:[10,19],init_model_path:10,initi:[10,20],initpaddl:23,input:[10,20],input_typ:[10,19],instal:2,insuffici:7,intal:5,integ:[10,20],integer_sequ:20,integer_valu:[10,20],integer_value_sequ:10,invok:20,is_predict:10,is_train:20,isinst:23,job:10,kernel:6,kwarg:[10,20],l2regular:10,label:[10,20],label_dim:10,lake:20,later:10,latest:[0,6],layer:10,ld_library_path:7,learning_method:10,learning_r:10,len:[10,20],lib64:[6,7],lib:1,libcuda:6,libnvidia:6,librari:7,list:[10,19,20],load_data_arg:23,loadparamet:23,local:[1,7],log_period:10,logger:20,look:[10,20],loss:10,lowest_dl_speed:0,lstm:10,lstm_size:10,mac:6,mai:5,main:23,maintainer:6,make:[7,20],make_diagram:13,maxid:10,maxid_lay:10,mean:10,memori:10,merge_model:13,method:20,mkdir:6,mkl:1,mkl_core:1,mkl_root:1,mnist:20,mnist_model:23,mnist_provid:20,mnist_train:20,model_config:23,modul:[10,20],momentum:10,movi:20,must:7,name:[6,10],necessari:10,need:10,neg:[10,20],net:6,no_cache:20,no_sequence:20,none:[10,20,23],note:[5,7],nullptr:7,num:10,num_pass:10,nvidia:6,obj:[10,20],object:[10,20],off:[0,1,18],omit:10,on_init:20,onli:10,open:[10,20],openbla:1,openblas_root:1,openssh:6,opt:1,other:10,outlin:22,output:10,outsid:20,packag:5,paddl:[0,6,7,10,13],paddle_gpu:0,paddle_ssh:6,paddle_ssh_machin:6,paddledev:6,paddlepaddl:[6,7,18,23],paramet:10,parse_config:23,pass:[10,20],path:[7,10],period:10,permitrootlogin:6,pixel:20,pixels_float:20,pixels_str:20,place:20,pleas:7,pool_siz:20,pooling_lay:10,pooling_typ:10,posit:[10,20],pre:5,pred:10,predict_output_dir:10,predict_sampl:23,preprocess:10,print:23,process:[10,20],process_pr:10,properli:10,pull:6,put:10,py_paddl:[6,23],pydataprovid:19,pydataprovider2:[10,20,23],pydataproviderwrapp:10,python:10,quick_start:10,rang:10,rank:10,rare:20,read:[10,20],refer:19,reference_cblas_root:1,reffer:1,regular:10,releas:[5,7],repres:10,represent:10,resnet:9,result:10,rmsprop:10,roce:6,root:6,run:6,runtim:[7,20],same:[10,20],sampl:[10,20],save:[10,20],save_dir:10,saw:20,sbin:6,script:0,second:10,sed:6,see:10,sentenc:20,sentiment:20,sentimental_provid:20,separ:10,seq_typ:20,sequel:20,sequence:20,sequence_conv_pool:10,sequencetyp:20,server:6,set:[10,20],setup:10,should_shuffl:20,simple_gru:10,simple_lstm:10,size:[10,20],smooth:5,softmax:10,softmaxactiv:10,sourc:10,spars:10,sparse_binary_vector:[10,20],sparse_float_vector:20,specifi:[7,10],split:[10,20],src_root:23,ssh:6,sshd:6,sshd_config:6,stat:10,state:5,still:5,stop:6,store:10,string:20,strip:10,structur:10,stun:20,sub_sequence:20,sure:7,swig_paddl:23,tag:[0,7],take:20,tbd:21,team:6,test:[10,19],test_data:23,test_list:[10,20],text:[10,20],text_conv:10,them:10,thi:[10,20],thing:20,train:7,train_list:[10,20],trainer:[10,20,23],trainer_config:[10,19,20,23],trainer_config_help:[10,20],trainerintern:10,trainermain:7,travel:20,trn:10,tst:10,two:10,txt:[10,20],type:[10,20],ubuntu:7,unk_idx:10,updat:6,use:[10,22],use_gpu:[10,23],usepam:6,user:10,usr:[1,6,7],valid:7,valu:[10,20,23],version:[6,7],via:7,want:20,what:10,when:20,which:10,whole:20,wilder:20,window:6,with_avx:[1,18],with_doc:1,with_doc_cn:1,with_doubl:18,with_double:1,with_dso:1,with_gflag:18,with_gflags:1,with_glog:[1,18],with_gpu:[0,1,18],with_metric_learn:18,with_predict_sdk:18,with_python:[1,18],with_rdma:[1,18],with_style_check:1,with_swig_py:1,with_testing:1,with_tim:18,with_timer:1,wonder:20,word:9,word_dict:10,word_dim:10,word_id:20,word_vector:10,xarg:6,yield:[10,20],you:[7,20],your:5,your_host_machine:6},titles:["\u6784\u5efaPaddlePaddle Docker Image","\u8bbe\u7f6ePaddlePaddle\u7684\u7f16\u8bd1\u9009\u9879","\u4f7f\u7528cmake\u7f16\u8bd1PaddlePaddle","\u5b89\u88c5\u7f16\u8bd1PaddlePaddle\u9700\u8981\u7684\u4f9d\u8d56","make\u548cmake install","\u7f16\u8bd1\u4e0e\u5b89\u88c5","\u5b89\u88c5PaddlePaddle\u7684Docker\u955c\u50cf","\u4f7f\u7528deb\u5305\u5728Ubuntu\u4e0a\u5b89\u88c5PaddlePaddle","\u96c6\u7fa4\u8bad\u7ec3","\u4f7f\u7528\u793a\u4f8b","PaddlePaddle\u5feb\u901f\u5165\u95e8\u6559\u7a0b","PaddlePaddle\u6587\u6863","&lt;no title&gt;","\u547d\u4ee4\u884c\u53c2\u6570","&lt;no title&gt;","&lt;no title&gt;","paddle pserver\u7684\u547d\u4ee4\u884c\u53c2\u6570","paddle train\u7684\u547d\u4ee4\u884c\u53c2\u6570","paddle version\u7684\u547d\u4ee4\u884c\u53c2\u6570","PaddlePaddle\u7684\u6570\u636e\u63d0\u4f9b(DataProvider)\u4ecb\u7ecd","PyDataProvider2\u7684\u4f7f\u7528","\u81ea\u5b9a\u4e49\u4e00\u4e2aDataProvider","\u7528\u6237\u63a5\u53e3","PaddlePaddle\u7684Python\u9884\u6d4b\u63a5\u53e3"],titleterms:{"\u4e0b\u8f7d\u548c\u8fd0\u884cdocker\u955c\u50cf":6,"\u4ecb\u7ecd":19,"\u4f18\u5316\u7b97\u6cd5":10,"\u4f7f\u7528\u6307\u5357":11,"\u4f7f\u7528\u6982\u8ff0":10,"\u4f7f\u7528\u793a\u4f8b":9,"\u4f7f\u7528\u811a\u672c\u6784\u5efapaddlepaddl":0,"\u4f7f\u7528cmake\u7f16\u8bd1paddlepaddl":2,"\u4f7f\u7528deb\u5305\u5728ubuntu\u4e0a\u5b89\u88c5paddlepaddl":7,"\u5377\u79ef\u6a21\u578b":10,"\u53c2\u8003":20,"\u53ef\u80fd\u9047\u5230\u7684\u95ee\u9898":7,"\u547d\u4ee4\u884c\u53c2\u6570":[10,13,22],"\u56fe\u50cf":9,"\u5b89\u88c5":10,"\u5b89\u88c5\u7f16\u8bd1paddlepaddle\u9700\u8981\u7684\u4f9d\u8d56":3,"\u5b89\u88c5paddlepaddle\u7684docker\u955c\u50cf":6,"\u5e38\u7528\u6a21\u578b":9,"\u5e8f\u5217\u6a21\u578b\u6570\u636e\u63d0\u4f9b":20,"\u5f00\u53d1\u6307\u5357":11,"\u6027\u80fd\u95ee\u9898":6,"\u603b\u4f53\u6548\u679c\u603b\u7ed3":10,"\u63a8\u8350":9,"\u6570\u636e\u5411\u6a21\u578b\u4f20\u9001":10,"\u6570\u636e\u63d0\u4f9b":22,"\u6570\u636e\u683c\u5f0f\u51c6\u5907":10,"\u65f6\u5e8f\u6a21\u578b":10,"\u6784\u5efapaddlepaddl":0,"\u6ce8\u610f\u4e8b\u9879":6,"\u7528\u6237\u63a5\u53e3":22,"\u7b80\u5355\u7684\u4f7f\u7528\u573a\u666f":20,"\u7b97\u6cd5\u6559\u7a0b":11,"\u7f16\u8bd1\u4e0e\u5b89\u88c5":5,"\u7f51\u7edc\u7ed3\u6784":10,"\u81ea\u5b9a\u4e49\u4e00\u4e2adataprovid":21,"\u81ea\u7136\u8bed\u8a00\u5904\u7406":9,"\u8bad\u7ec3\u6a21\u578b":10,"\u8bbe\u7f6epaddlepaddle\u7684\u7f16\u8bd1\u9009\u9879":1,"\u8bcd\u5411\u91cf\u6a21\u578b":10,"\u8f93\u51fa\u65e5\u5fd7":10,"\u8fdc\u7a0b\u8bbf\u95ee\u95ee\u9898\u548c\u4e8c\u6b21\u5f00\u53d1":6,"\u903b\u8f91\u56de\u5f52\u6a21\u578b":10,"\u914d\u7f6e\u4e2d\u7684\u6570\u636e\u52a0\u8f7d\u5b9a\u4e49":10,"\u9644\u5f55":10,"\u96c6\u7fa4\u8bad\u7ec3":8,"\u9884\u6d4b":[10,22],"blas\u76f8\u5173\u7684\u7f16\u8bd1\u9009\u9879":1,"bool\u578b\u7684\u7f16\u8bd1\u9009\u9879":1,"config\u6587\u4ef6\u627e\u4e0d\u5230":7,"cudnn\u76f8\u5173\u7684\u7f16\u8bd1\u9009\u9879":1,"driver\u627e\u4e0d\u5230":7,"make\u548cmak":4,"paddlepaddle\u5feb\u901f\u5165\u95e8\u6559\u7a0b":10,"paddlepaddle\u63d0\u4f9b\u7684docker\u955c\u50cf\u7248\u672c":6,"paddlepaddle\u6587\u6863":11,"paddlepaddle\u7684\u6570\u636e\u63d0\u4f9b":19,"paddlepaddle\u7684\u7f16\u8bd1\u9009\u9879":1,"paddlepaddle\u7684bool\u578b\u7f16\u8bd1\u9009\u9879":1,"paddlepaddle\u7684cblas\u7f16\u8bd1\u9009\u9879":1,"paddlepaddle\u7684python\u9884\u6d4b\u63a5\u53e3":23,"pserver\u7684\u547d\u4ee4\u884c\u53c2\u6570":16,"pydataprovider2\u7684\u4f7f\u7528":20,"python\u6570\u636e\u52a0\u8f7d\u811a\u672c":10,"so\u627e\u4e0d\u5230":7,"train\u7684\u547d\u4ee4\u884c\u53c2\u6570":17,"version\u7684\u547d\u4ee4\u884c\u53c2\u6570":18,algorithm:10,appendix:10,architectur:10,argument:10,cach:20,command:10,configur:10,convolut:10,cuda:[1,7],data:10,dataprovid:19,docker:0,image:0,init_hook:20,input_typ:20,instal:4,install:10,libcudart:7,libcudnn:7,line:10,log:10,logist:10,model:10,network:10,optimiz:10,overview:10,paddl:[16,17,18],predict:10,prepar:10,provid:[10,20],refer:20,regress:10,script:10,sequenc:10,summari:10,time:10,train:10,transfer:10,vector:10,word:10}})
+Search.setIndex({envversion:49,filenames:["build/docker/build_docker_image","build_and_install/cmake/compile_options","build_and_install/cmake/index","build_and_install/cmake/install_deps","build_and_install/cmake/make_and_install","build_and_install/index","build_and_install/install/docker_install","build_and_install/install/ubuntu_install","cluster/index","demo/index","demo/quick_start/index","index","ui/cmd/dump_config","ui/cmd/index","ui/cmd/make_diagram","ui/cmd/merge_model","ui/cmd/paddle_pserver","ui/cmd/paddle_train","ui/cmd/paddle_version","ui/data_provider/index","ui/data_provider/pydataprovider2","ui/data_provider/write_new_dataprovider","ui/index","ui/predict/swig_py_paddle"],objects:{},objnames:{},objtypes:{},terms:{"0000x":10,"000\u5f20\u7070\u5ea6\u56fe\u7247\u7684\u6570\u5b57\u5206\u7c7b\u6570\u636e\u96c6":20,"00186201e":23,"04\u7248\u672c\u4f7f\u7528deb\u5305\u5b89\u88c5":7,"08823112e":23,"0b0":7,"10\u4ee5\u4e0a\u7684linux":6,"10\u7ef4\u7684\u6574\u6570\u503c":20,"10gbe":6,"10m":0,"12194102e":23,"12\u7248\u672c\u6d4b\u8bd5\u901a\u8fc7":0,"12\u7248\u672c\u7684\u60c5\u51b5\u4e0b\u5e76\u6ca1\u6709\u6d4b\u8bd5":0,"15501715e":23,"15mb":10,"16mb":10,"1\u7684\u8bdd":20,"252kb":10,"25639710e":23,"25k":10,"27787406e":23,"28\u7684\u50cf\u7d20\u7070\u5ea6\u503c":20,"28\u7684\u7a20\u5bc6\u5411\u91cf\u548c\u4e00\u4e2a":20,"2\u8fdb\u884c\u8fdb\u4e00\u6b65\u6f14\u5316":10,"32777140e":23,"36540484e":23,"40gbe":6,"43630644e":23,"48565123e":23,"48684503e":23,"49316648e":23,"50k":0,"51111044e":23,"53018653e":23,"56gbe":6,"5\u5230\u672c\u5730\u73af\u5883\u4e2d":7,"70634608e":23,"72296313e":23,"85625684e":23,"93137714e":23,"96644767e":23,"99982715e":23,"9\u7684\u6570\u5b57":20,"\u4e00\u4e2a\u6587\u4ef6":20,"\u4e00\u4e2alogging\u5bf9\u8c61":20,"\u4e00\u4e2apass\u8868\u793a\u8fc7\u4e00\u904d\u6240\u6709\u8bad\u7ec3\u6837\u672c":10,"\u4e00\u6761":20,"\u4e00\u81f4\u7684\u7279\u5f81":20,"\u4e00\u822c\u60c5\u51b5\u4e0b":19,"\u4e00\u822c\u63a8\u8350\u8bbe\u7f6e\u6210true":20,"\u4e00\u884c\u4e3a\u4e00\u4e2a\u6837\u672c":10,"\u4e00\u884c\u5bf9\u5e94\u4e00\u4e2a\u6570\u636e\u6587\u4ef6":19,"\u4e0a\u7684\u6548\u679c":10,"\u4e0b\u8f7d\u8fdb\u7a0b\u4f1a\u91cd\u542f":0,"\u4e0b\u8ff0\u5185\u5bb9\u5c06\u5206\u4e3a\u5982\u4e0b\u51e0\u4e2a\u7c7b\u522b\u63cf\u8ff0":6,"\u4e0b\u975e\u5e38\u5c11\u7684\u53d8\u91cf\u5f15\u7528":20,"\u4e0b\u9762dataprovid":10,"\u4e0d\u4e00\u5b9a\u548c\u65f6\u95f4\u6709\u5173\u7cfb":20,"\u4e0d\u4f1a\u6267\u884c\u6d4b\u8bd5\u64cd\u4f5c":19,"\u4e0d\u5305\u542blabel":23,"\u4e0d\u540c\u7684\u6570\u636e\u7c7b\u578b\u548c\u5e8f\u5217\u6a21\u5f0f\u8fd4\u56de\u7684\u683c\u5f0f\u4e0d\u540c":20,"\u4e0d\u540c\u7684\u7a33\u5b9a\u7248\u672c":6,"\u4e0d\u7f13\u5b58\u4efb\u4f55\u6570\u636e":20,"\u4e0e\u8bad\u7ec3\u7f51\u7edc\u914d\u7f6e\u4e0d\u540c\u7684\u662f":10,"\u4e14\u5e8f\u5217\u7684\u6bcf\u4e00\u4e2a\u5143\u7d20\u8fd8\u662f\u4e00\u4e2a\u65f6\u95f4\u5e8f\u5217":20,"\u4e24\u4e2a\u6587\u6863":6,"\u4e24\u7c7b":10,"\u4e25\u91cd\u7684\u95ee\u9898":20,"\u4e2a":10,"\u4e2ayield":20,"\u4e2d":10,"\u4e2d\u5b9a\u4e49\u4f7f\u7528\u54ea\u79cddataprovider\u53ca\u5176\u53c2\u6570":19,"\u4e2d\u5b9a\u4e49\u7684\u987a\u5e8f\u4e00\u81f4":20,"\u4e2d\u5bfb\u627e\u8fd9\u4e9bblas\u7684\u5b9e\u73b0":1,"\u4e2d\u7684":20,"\u4e2d\u7684\u4e8c\u8fdb\u5236\u4f7f\u7528\u4e86":6,"\u4e2d\u7684set":20,"\u4e2d\u914d\u7f6e":20,"\u4e3a":20,"\u4e3a\u4e86\u8fd0\u884cpaddlepaddle\u7684docker\u955c\u50cf":6,"\u4e3a\u4f8b\u8fdb\u884c\u9884\u6d4b":10,"\u4e3b\u8981\u51fd\u6570\u662fprocess\u51fd\u6570":20,"\u4e3b\u8981\u5206\u4e3a\u4ee5\u4e0b\u51e0\u4e2a\u6b65\u9aa4":23,"\u4e3b\u8981\u5305\u62ec\u4e24\u90e8\u5206":10,"\u4e3b\u8981\u662f\u589e\u52a0\u4e86\u521d\u59cb\u5316\u673a\u5236":20,"\u4e3b\u8981\u6b65\u9aa4\u4e3a":23,"\u4e3b\u8981\u7531\u4e8e\u65e7\u7248\u672c":0,"\u4e3b\u8981\u7684\u8f6f\u4ef6\u5305\u4e3apy_paddl":23,"\u4e5f\u4f1a\u6254\u5230\u8fd9\u6761\u6570\u636e":20,"\u4e5f\u4f1a\u8bfb\u53d6\u76f8\u5173\u8def\u5f84\u53d8\u91cf\u6765\u8fdb\u884c\u641c\u7d22":1,"\u4e5f\u53ef\u4ee5":20,"\u4e5f\u53ef\u4ee5\u4f7f\u7528":20,"\u4e5f\u53ef\u4ee5\u548cpaddl":13,"\u4e5f\u53ef\u4ee5\u76f4\u63a5\u6267\u884c":6,"\u4e5f\u53ef\u4ee5\u901a\u8fc7\u5982\u4e0b\u65b9\u5f0f\u9884\u6d4b":10,"\u4e5f\u53ef\u4ee5\u901a\u8fc7save":10,"\u4e5f\u53ef\u4ee5\u9884\u6d4b\u6ca1\u6709label\u7684\u6d4b\u8bd5\u96c6":10,"\u4e5f\u5c31\u662f\u5c06\u8bcd\u5411\u91cf\u6a21\u578b\u989d\u6b65":10,"\u4e5f\u5c31\u662f\u76f4\u63a5\u5199\u5185\u5b58\u7684float\u6570\u7ec4":23,"\u4e5f\u9700\u8981\u4e24\u6b21\u968f\u673a\u9009\u62e9\u5230\u540c\u6837\u7684generator\u7684\u65f6\u5019":20,"\u4e86\u975e\u5e38\u65b9\u4fbf\u7684\u4e8c\u8fdb\u5236\u5206\u53d1\u624b\u6bb5":6,"\u4e8c\u6b21\u5f00\u53d1\u53ef\u4ee5":6,"\u4eba\u5458\u7b49\u7b49":0,"\u4ec5\u4ec5\u9700\u8981":20,"\u4ecb\u7ecdpaddlepaddle\u4f7f\u7528\u6d41\u7a0b\u548c\u5e38\u7528\u7684\u7f51\u7edc\u57fa\u7840\u5355\u5143\u7684\u914d\u7f6e\u65b9\u6cd5":10,"\u4ece\u6587\u4ef6\u4e2d\u8bfb\u53d6\u6bcf\u4e00\u6761\u6570\u636e":20,"\u4ece\u6587\u672c\u6587\u4ef6\u4e2d\u8bfb\u53d6":20,"\u4ece\u800c\u4e0d\u80fd\u5728\u8fd0\u884c\u7f16\u8bd1\u547d\u4ee4\u7684\u65f6\u5019\u63a5\u53d7\u53c2\u6570":0,"\u4ece\u800c\u751f\u6210\u591a\u4e2agener":20,"\u4ece\u800c\u9632\u6b62\u8fc7\u62df\u5408":19,"\u4ed6\u4eec\u662f":[13,20],"\u4ed6\u4eec\u7684\u4e0b\u8f7d\u5730\u5740\u662f":7,"\u4ed6\u53ef\u4ee5\u5c06\u67d0\u4e00\u4e2a\u51fd\u6570\u6807\u8bb0\u6210\u4e00\u4e2apydataprovid":20,"\u4ee4\u884c\u53c2\u6570\u6587\u6863":10,"\u4ee5\u592a\u7f51\u5361":6,"\u4ee5\u76f8\u5bf9\u8def\u5f84\u5f15\u7528":19,"\u4efb\u610f\u4e00\u79cdcblas\u5b9e\u73b0":1,"\u4f1a\u5c06latest\u66ff\u6362\u6210\u7a33\u5b9a\u7248\u672c\u7684\u7248\u672c\u53f7":6,"\u4f1a\u62a5\u5bfb\u627e\u4e0d\u5230\u8fd9\u4e9b\u52a8\u6001\u5e93":7,"\u4f1a\u6839\u636e\u547d\u4ee4\u884c\u53c2\u6570\u6307\u5b9a\u7684\u6d4b\u8bd5\u65b9\u5f0f":19,"\u4f1a\u6839\u636einput_types\u68c0\u67e5\u6570\u636e\u7684\u5408\u6cd5\u6027":20,"\u4f1a\u751f\u6210\u591a\u4e2agener":20,"\u4f1a\u9884\u5148\u8bfb\u53d6\u5168\u90e8\u6570\u636e\u5230\u5185\u5b58\u4e2d":20,"\u4f20\u5165\u547d\u4ee4\u884c\u53c2\u6570\u521d\u59cb\u5316":23,"\u4f20\u5165\u7684\u662f\u4e00\u4e2a\u51fd\u6570":20,"\u4f20\u5165\u7684\u914d\u7f6e\u53c2\u6570\u5305\u62ec":0,"\u4f20\u5165\u8fd9\u4e2a\u53d8\u91cf\u7684\u65b9\u5f0f\u4e3a":20,"\u4f46\u662f":[0,20],"\u4f46\u662f\u5728":20,"\u4f46\u662f\u5982\u679c\u4f7f\u7528\u4e86\u9ad8\u6027\u80fd\u7684\u7f51\u5361":6,"\u4f46\u662f\u65b9\u4fbf\u8c03\u8bd5\u548cbenchmark":1,"\u4f46\u662f\u6709\u65f6\u4e3a\u4e86\u8ba1\u7b97\u5747\u8861\u6027":20,"\u4f46\u9700\u8981\u6ce8\u610f\u7684\u662f\u7f16\u8bd1\u548c":1,"\u4f4e\u4e8edocker":0,"\u4f53\u53ef\u4ee5\u53c2\u8003":20,"\u4f7f\u5728python\u73af\u5883\u4e0b\u7684\u9884\u6d4b\u63a5\u53e3\u66f4\u52a0\u7b80\u5355":23,"\u4f7f\u7528":[1,6,23],"\u4f7f\u7528\u7684\u547d\u4ee4\u4e5f\u662f":1,"\u4f7f\u7528\u8be5\u63a5\u53e3\u7528\u6237\u53ef\u4ee5\u53ea\u5173\u6ce8\u5982\u4f55":20,"\u4f7f\u7528\u8be5dockerfile\u6784\u5efa\u51fa\u955c\u50cf":6,"\u4f7f\u7528\u8d77\u6765\u6216\u8bb8\u4f1a\u4e0d\u662f\u5f88\u987a\u7545":5,"\u4f7f\u7528\u8fd9\u4e2a\u5173\u952e\u8bcd":20,"\u4f7f\u7528deb\u5305\u5728ubuntu\u4e0a\u5b89\u88c5paddlepaddl":5,"\u4f7f\u7528dockerfile\u6784\u5efa\u4e00\u4e2a\u5168\u65b0\u7684dock":6,"\u4f7f\u7528mnist\u624b\u5199\u8bc6\u522b\u4f5c\u4e3a\u6837\u4f8b":23,"\u4f7f\u7528ssh\u8bbf\u95eepaddlepaddle\u955c\u50cf":6,"\u4f8b\u5982":[1,7,10,20],"\u4f8b\u5982\u6587\u4ef6\u540d\u662f":20,"\u4f8b\u5982rdma\u7f51\u5361":6,"\u4f8b\u5982sigmoid\u53d8\u6362":10,"\u4f9d\u6b21\u8fd4\u56de\u4e86\u6587\u4ef6\u4e2d\u7684\u6bcf\u6761\u6570\u636e":20,"\u4fe1\u606f":6,"\u5173\u95edcontain":6,"\u5176\u4e2d":[0,6,19,20,23],"\u5176\u4e2d\u6587\u672c\u8f93\u5165\u7c7b\u578b\u5b9a\u4e49\u4e3a\u6574\u6570\u65f6\u5e8f\u7c7b\u578binteg":10,"\u5176\u4e2d\u7b2c\u4e00\u884c\u662f\u5f15\u5165paddlepaddle\u7684pydataprovider2\u5305":20,"\u5176\u4ed6\u53c2\u6570\u5747\u4f7f\u7528kei":20,"\u5176\u4ed6\u53c2\u6570\u8bf7\u53c2\u8003":10,"\u5176\u4ed6\u53c2\u6570\u90fd\u4f7f\u7528kei":20,"\u5176\u4f5c\u7528\u662f\u5c06\u8bad\u7ec3\u6570\u636e\u4f20\u5165\u5185\u5b58\u6216\u8005\u663e\u5b58":19,"\u5176\u5b83\u90e8\u5206\u548c\u903b\u8f91\u56de\u5f52\u7f51\u7edc\u7ed3\u6784\u4e00\u81f4":10,"\u5176\u6570\u636e\u4f7f\u7528":20,"\u5176\u7b2c\u4e00\u884c\u8bf4\u660e\u4e86paddle\u7684\u7248\u672c":18,"\u5177":20,"\u5177\u4f53\u53ef\u4ee5\u8bbe\u7f6e\u6210\u4ec0\u4e48\u5176\u4ed6\u683c":20,"\u5177\u4f53\u6709\u54ea\u4e9b\u683c\u5f0f":20,"\u5177\u4f53\u8bf7\u53c2\u8003\u6ce8\u610f\u4e8b\u9879\u4e2d\u7684":6,"\u5177\u4f53dataprovider\u8fd8\u5177\u6709\u4ec0\u4e48\u529f\u80fd":20,"\u5177\u6709\u4e24\u4e2a\u53c2\u6570":20,"\u5177\u6709\u548c\u5bbf\u4e3b\u673a\u76f8\u8fd1\u7684\u8fd0\u884c\u6548\u7387":6,"\u5177\u6709\u7684\u5c5e\u6027\u6709":20,"\u5178\u578b\u7684\u8f93\u51fa\u7ed3\u679c\u4e3a":23,"\u5178\u578b\u7684\u9884\u6d4b\u4ee3\u7801\u5982\u4e0b":23,"\u5185\u5b58\u4e0d\u591f\u7528\u7684\u60c5\u51b5":19,"\u518d\u6307\u5b9a":1,"\u5199\u5165train":20,"\u5199\u5728train":19,"\u51c6\u5907\u6570\u636e":23,"\u51fd\u6570":20,"\u51fd\u6570\u4e2d":20,"\u51fd\u6570\u4e2d\u4f7f\u7528":20,"\u51fd\u6570\u4e2d\u7684":20,"\u51fd\u6570\u662f\u4f7f\u7528":20,"\u51fd\u6570\u6765\u4fdd\u8bc1\u517c\u5bb9\u6027":20,"\u51fd\u6570\u67e5\u8be2\u6587\u6863":23,"\u5206\u5e03\u5f0f\u8bad\u7ec3":10,"\u5206\u7c7b\u6210\u6b63\u9762\u60c5\u7eea\u548c":20,"\u5217\u8868\u5982\u4e0b":20,"\u5219\u5b57\u4e0e\u5b57\u4e4b\u95f4\u7528\u7a7a\u683c\u5206\u9694":10,"\u5219\u9700\u8981\u5148\u5c06":6,"\u5219\u9700\u8981\u8fdb\u884c\u4e00\u5b9a\u7684\u4e8c\u6b21\u5f00\u53d1":6,"\u521b\u5efa\u4e00\u4e2a":23,"\u521b\u5efagener":20,"\u5220\u9664contain":6,"\u5229\u7528\u5355\u8bcdid\u67e5\u627e\u5bf9\u5e94\u7684\u8be5\u5355\u8bcd\u7684\u8fde\u7eed\u8868\u793a\u5411\u91cf":10,"\u5229\u7528\u903b\u8f91\u56de\u5f52\u6a21\u578b\u5bf9\u8be5\u5411\u91cf\u8fdb\u884c\u5206\u7c7b":10,"\u522b":10,"\u5269\u4e0b\u7684pass\u4f1a\u76f4\u63a5\u4ece\u5185\u5b58\u91cc":20,"\u52a0\u4e86l2\u6b63\u5219\u548c\u68af\u5ea6\u622a\u65ad":10,"\u52a0\u8f7d\u6570\u636e":10,"\u5305":6,"\u5305\u548c":6,"\u5305\u62ec":10,"\u5305\u62ec\u7b80\u5355\u7684rnn\u6a21\u578b":10,"\u5305\u62ecdocker\u955c\u50cf":5,"\u5373":[6,10],"\u5373\u4e0d\u5728\u4e4e\u5185\u5b58\u6682\u5b58\u591a\u5c11\u6761\u6570\u636e":20,"\u5373\u4e0d\u662f\u4e00\u6761\u5e8f\u5217":20,"\u5373\u4ece\u5355\u8bcd\u5b57\u7b26\u4e32\u5230\u5355\u8bcdid\u7684\u5b57\u5178":20,"\u5373\u4f1a\u751f\u6210100\u4e2agener":20,"\u5373\u4f7f\u5728check\u4e2d\u6570\u636e\u4e0d\u5408\u6cd5":20,"\u5373\u4f7f\u5728process\u91cc\u9762\u53ea\u4f1a\u6709\u4e00":20,"\u5373\u5305\u542b\u65f6\u95f4\u6b65\u4fe1\u606f":20,"\u5373\u53ef":20,"\u5373\u53ef\u4ee5\u4f7f\u7528ssh\u8bbf\u95ee\u5bbf\u4e3b\u673a\u76848022\u7aef\u53e3":6,"\u5373\u53ef\u542f\u52a8\u548c\u8fdb\u5165paddlepaddle\u7684contain":6,"\u5373\u53ef\u5728\u672c\u5730\u7f16\u8bd1\u51fapaddlepaddle\u7684\u955c\u50cf":0,"\u5373\u53ef\u6253\u5370\u51fapaddlepaddle\u7684\u7248\u672c\u548c\u6784\u5efa":6,"\u5373\u5927\u90e8\u5206\u503c\u4e3a0":20,"\u5373\u5982\u679ctrain":20,"\u5373\u5bf9\u7b2c3\u6b65\u8fdb\u884c\u66ff\u6362":10,"\u5373\u662f\u4e00\u6761\u65f6\u95f4\u5e8f\u5217":20,"\u5373cpu\u7248\u672c":7,"\u5373train":20,"\u5377\u79ef\u7f51\u7edc\u662f\u4e00\u79cd\u7279\u6b8a\u7684\u4ece\u8bcd\u5411\u91cf\u8868\u793a\u5230\u53e5\u5b50\u8868\u793a\u7684\u65b9\u6cd5":10,"\u53c2\u6570":0,"\u53c2\u6570\u6570\u91cf":10,"\u53c2\u8003":19,"\u53c2\u89c1":[3,4],"\u53d6\u51b3\u4e8e\u662f\u5426\u5bfb\u627e\u5230gflags":1,"\u53d6\u51b3\u4e8e\u662f\u5426\u5bfb\u627e\u5230glog":1,"\u53d6\u51b3\u4e8e\u662f\u5426\u5bfb\u627e\u5230gtest":1,"\u53d6\u51b3\u4e8e\u662f\u5426\u627e\u5230swig":1,"\u53d8\u4e3a3\u4e2a\u65b0\u7684\u5b50\u6b65\u9aa4":10,"\u53d8\u4f1a\u62a5\u8fd9\u4e2a\u9519\u8bef":7,"\u53d8\u91cf":20,"\u53e5\u5b50\u8868\u793a\u7684\u8ba1\u7b97\u66f4\u65b0\u4e3a2\u6b65":10,"\u53ea\u662f\u5c06\u53e5\u5b50\u5229\u7528\u8fde\u7eed\u5411\u91cf\u8868\u793a\u66ff\u6362\u7a00\u758f":10,"\u53ea\u662f\u8bf4\u660e\u6570\u636e\u7684\u987a\u5e8f\u662f\u91cd\u8981\u7684":20,"\u53ea\u80fd\u591f\u8fd4\u56delist\u6216\u8005tupl":20,"\u53ea\u9700\u8981\u4f7f\u7528\u4e00\u884c\u4ee3\u7801\u5373\u53ef\u4ee5\u8bbe\u7f6e\u8bad\u7ec3\u5f15\u7528\u8fd9\u4e2adataprovid":20,"\u53ea\u9700\u8981\u5728":20,"\u53ea\u9700\u8981\u77e5\u9053\u8fd9\u53ea\u662f\u4e00\u4e2a\u6807\u8bb0\u5c5e\u6027\u7684\u65b9\u6cd5\u5c31\u53ef\u4ee5\u4e86":20,"\u53ef\u4ee5\u4e3a\u4e00\u4e2a\u6570\u636e\u6587\u4ef6\u8fd4\u56de\u591a\u6761\u8bad\u7ec3\u6837\u672c":20,"\u53ef\u4ee5\u4f20\u516510k":0,"\u53ef\u4ee5\u4f7f\u7528":0,"\u53ef\u4ee5\u4f7f\u7528\u8bad\u7ec3\u597d\u7684\u6a21\u578b\u8bc4\u4f30\u5e26\u6709label\u7684\u9a8c\u8bc1\u96c6":10,"\u53ef\u4ee5\u4f7f\u7528graphviz\u5bf9paddlepaddle\u7684\u7f51\u7edc\u6a21\u578b\u8fdb\u884c\u7ed8\u5236":13,"\u53ef\u4ee5\u4f7f\u7528paddl":13,"\u53ef\u4ee5\u4f7f\u7528python\u7684":23,"\u53ef\u4ee5\u53c2\u8003":10,"\u53ef\u4ee5\u5728\u4e00\u4e2a\u51fd\u6570\u91cc":20,"\u53ef\u4ee5\u5728cmake\u7684\u547d\u4ee4\u884c\u8bbe\u7f6e":1,"\u53ef\u4ee5\u5c06\u4e00\u6761\u6570\u636e\u8bbe\u7f6e\u6210\u591a\u4e2abatch":20,"\u53ef\u4ee5\u5c06paddlepaddle\u7684\u6a21\u578b\u548c\u914d\u7f6e\u6253\u5305\u6210\u4e00\u4e2a\u6587\u4ef6":13,"\u53ef\u4ee5\u5c06paddlepaddle\u7684\u8bad\u7ec3\u6a21\u578b\u4ee5proto":13,"\u53ef\u4ee5\u65b9\u4fbf\u5d4c\u5165\u5f0f\u5de5\u4f5c":1,"\u53ef\u4ee5\u6839\u636e\u4e0d\u540c\u7684\u6570\u636e\u914d\u7f6e\u4e0d\u540c\u7684\u8f93\u5165\u7c7b\u578b":20,"\u53ef\u4ee5\u8fd4\u56de\u4e00\u4e2adict":20,"\u53ef\u4ee5\u901a\u8fc7show":10,"\u53ef\u7528\u5728\u6d4b\u8bd5\u6216\u8bad\u7ec3\u65f6\u6307\u5b9a\u521d\u59cb\u5316\u6a21\u578b":10,"\u53ef\u80fd\u7684\u5185\u5b58\u6cc4\u9732\u95ee\u9898":19,"\u53ef\u9009":20,"\u5404\u79cd\u53c2\u6570\u548c\u7ef4\u62a4":0,"\u540c\u65f6":[0,1,6,20],"\u540c\u65f6\u4f1a\u8ba1\u7b97\u5206\u7c7b\u51c6\u786e\u7387":10,"\u540c\u65f6\u5305\u542bcpu\u5f00\u53d1\u73af\u5883":6,"\u540c\u65f6\u5305\u542bcpu\u5f00\u53d1\u73af\u5883\u548c\u6e90\u7801":6,"\u540c\u65f6\u5305\u542bgpu\u5f00\u53d1\u73af\u5883":6,"\u540c\u65f6\u5305\u542bgpu\u5f00\u53d1\u73af\u5883\u548c\u6e90\u7801":6,"\u540c\u65f6\u6b22\u8fce\u8d21\u732e\u66f4\u591a\u7684\u5b89\u88c5\u5305":5,"\u540d\u79f0":10,"\u540e\u9762\u8ddf\u7740\u4e00\u7cfb\u5217\u7f16\u8bd1\u53c2\u6570":18,"\u5411\u91cf\u8868\u793a":10,"\u5426":1,"\u5426\u5219":19,"\u547d\u4ee4":0,"\u547d\u4ee4\u4e3a":6,"\u547d\u4ee4\u6307\u5b9a\u7684\u53c2\u6570\u4f1a\u4f20\u5165\u7f51\u7edc\u914d\u7f6e\u4e2d":10,"\u547d\u4ee4\u8fd0\u884c\u955c\u50cf":6,"\u547d\u4ee4\u9884\u5148\u4e0b\u8f7d\u955c\u50cf":6,"\u548c":[1,6,20],"\u548c\u4e09\u79cd\u5e8f\u5217\u6a21\u5f0f":20,"\u548c\u5dee\u8bc4":10,"\u548c\u6587\u672c\u4fe1\u606f\u7528tab\u95f4\u9694":10,"\u548c\u6d4b\u8bd5\u6587\u4ef6\u5217\u8868":19,"\u548c\u7528\u6237\u4f20\u5165\u7684\u53c2\u6570":20,"\u548c\u9884\u5904\u7406\u811a\u672c":10,"\u548ccudnn":7,"\u548cgpu\u7248\u672c":7,"\u548cinitalizer\u91cc\u5b9a\u4e49\u987a\u5e8f\u4e00\u81f4":10,"\u5668":10,"\u56db\u79cd\u6570\u636e\u7c7b\u578b\u662f":20,"\u56fe\u50cf\u5206\u7c7b":9,"\u5728":[1,7,20],"\u5728\u58f0\u660edataprovider\u7684\u65f6\u5019\u4f20\u5165\u4e86dictionary\u4f5c\u4e3a\u53c2\u6570":20,"\u5728\u5b8c\u6210\u4e86\u6570\u636e\u548c\u7f51\u7edc\u7ed3\u6784\u642d\u5efa\u4e4b\u540e":10,"\u5728\u672c\u95ee\u9898\u4e2d":10,"\u5728\u6a21\u578b\u914d\u7f6e\u4e2d\u5229\u7528":10,"\u5728\u6b64\u4e3a\u65b9\u4fbf\u5bf9\u6bd4\u4e0d\u540c\u7f51\u7edc\u7ed3\u6784":10,"\u5728\u6bcf\u4e2a\u7ef4\u5ea6\u4e0a\u53d6\u51fa\u5728\u8be5\u53e5\u8bdd\u65b0\u7684\u5411\u91cf\u96c6\u5408\u4e0a\u8be5\u7ef4\u5ea6\u7684\u6700\u5927\u503c\u4f5c\u4e3a\u6700\u540e\u7684\u53e5\u5b50\u8868\u793a\u5411\u91cf":10,"\u5728\u7a0b\u5e8f\u5f00\u59cb\u9636\u6bb5":23,"\u5728\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u8fdb\u884c\u6d4b\u8bd5":19,"\u5728\u8bad\u7ec3\u914d\u7f6e\u91cc":20,"\u5728\u8fd9\u4e2a\u51fd\u6570\u4e2d":20,"\u5728\u914d\u7f6e\u4e2d\u8bfb\u53d6\u4e86\u5b57\u5178":20,"\u5728cmake\u914d\u7f6e\u65f6\u53ef\u4ee5\u4f7f\u7528":1,"\u5728pydataprovider\u4e2d":20,"\u5728python\u73af\u5883\u4e0b\u9884\u6d4b\u7ed3\u679c":23,"\u57fa\u672c\u4e0a\u4e0d\u80fd\u6574\u4f53\u4fee\u6b63":20,"\u57fa\u672c\u7684\u5904\u7406\u903b\u8f91\u4e5f\u548cmnist\u903b\u8f91\u4e00\u81f4":20,"\u57fa\u672c\u7684pydataprovider\u4f7f\u7528\u4ecb\u7ecd\u5b8c\u6bd5\u4e86":20,"\u591a\u4e2ainput\u4ee5list\u65b9\u5f0f\u8f93\u5165":10,"\u591a\u6b21\u8fd4\u56de\u53d8\u91cf":20,"\u591a\u7ebf\u7a0b\u4e0b\u8f7d\u8fc7\u7a0b\u4e2d":0,"\u591a\u7ebf\u7a0b\u6570\u636e\u8bfb\u53d6":20,"\u5927":20,"\u597d\u8bc4":10,"\u5982\u679c":20,"\u5982\u679c\u4e0d\u4e86\u89e3":20,"\u5982\u679c\u4e0d\u4f7f\u7528\u5219\u4f1a\u4f7f\u7528\u4e00\u4e2a\u7b80\u5316\u7248\u7684\u547d\u4ee4\u884c\u53c2\u6570\u89e3\u6790":1,"\u5982\u679c\u4e0d\u4f7f\u7528\u5219\u4f1a\u4f7f\u7528\u4e00\u4e2a\u7b80\u5316\u7248\u7684\u65e5\u5fd7\u5b9e\u73b0":1,"\u5982\u679c\u4e0d\u5207\u8bcd":10,"\u5982\u679c\u4e0d\u8bbe\u7f6e\u7684\u8bdd":20,"\u5982\u679c\u4f7f\u7528gpu\u7248\u672c\u7684paddlepaddl":7,"\u5982\u679c\u5728":7,"\u5982\u679c\u5728\u7b2c\u4e00\u6b21cmake\u4e4b\u540e\u60f3\u8981\u91cd\u65b0\u8bbe":1,"\u5982\u679c\u5728\u8bad\u7ec3\u65f6":20,"\u5982\u679c\u5c0f\u4e8e\u8fd9\u4e2a\u4e0b\u8f7d\u901f\u5ea6":0,"\u5982\u679c\u60a8\u4f7f\u7528":6,"\u5982\u679c\u60f3\u8981\u5728\u5916\u90e8\u673a\u5668\u8bbf\u95ee\u8fd9\u4e2acontain":6,"\u5982\u679c\u662ffalse\u7684\u8bdd":20,"\u5982\u679c\u662ftrue\u7684\u8bdd":20,"\u5982\u679c\u6709\u66f4\u590d\u6742\u7684\u4f7f\u7528":19,"\u5982\u679c\u7528\u6237\u4e0d\u6307\u5b9a\u8fd4\u56de\u6570\u636e\u7684\u5bf9\u5e94\u5173\u7cfb":20,"\u5982\u679c\u7cfb\u7edf\u4f7f\u7528\u4e0d\u652f\u6301":7,"\u5982\u679c\u8bbe\u7f6e\u6210true\u7684\u8bdd":20,"\u5982\u679c\u8fd0\u884cgpu\u7248\u672c\u7684paddlepaddl":6,"\u5982\u679ctest":19,"\u5b83\u5305\u542b\u7684\u53c2\u6570\u6709":20,"\u5b89\u88c5\u597d\u7684paddlepaddle\u811a\u672c\u5305\u62ec\u591a\u6761\u547d\u4ee4":13,"\u5b89\u88c5\u5b8c\u6210\u7684paddlepaddle\u4e3b\u4f53\u5305\u62ec\u4e09\u4e2a\u90e8\u5206":6,"\u5b89\u88c5\u5b8c\u6210paddlepaddle\u540e":7,"\u5b89\u88c5\u6559\u7a0b":10,"\u5b89\u88c5\u65b9\u6cd5\u8bf7\u53c2\u8003":6,"\u5b89\u88c5\u7f16\u8bd1\u4f9d\u8d56":3,"\u5b89\u88c5\u7f16\u8bd1paddlepaddle\u9700\u8981\u7684\u4f9d\u8d56":2,"\u5b89\u88c5docker\u9700\u8981\u60a8\u7684\u673a\u5668":6,"\u5b89\u88c5paddlepaddl":10,"\u5b89\u88c5paddlepaddle\u7684docker\u955c\u50cf":5,"\u5b8c\u6210\u591a\u673a\u8bad\u7ec3":13,"\u5b8c\u6574\u4ee3\u7801\u89c1":23,"\u5b9a\u4e49\u6587\u672c\u4fe1\u606f":10,"\u5b9e\u73b0\u4e86\u6253\u5f00\u6587\u672c\u6587\u4ef6":20,"\u5bc6\u7801\u4e5f\u662froot":6,"\u5bf9\u4e8e\u7528\u6237\u6765\u8bf4":20,"\u5bf9\u4e8e\u7ed9\u5b9a\u7684\u4e00\u6761\u6587\u672c":10,"\u5bf9\u4e8ecuda\u7684toolkit\u6709\u65ad\u70b9\u7eed\u4f20\u548c\u4f20\u8f93\u901f\u5ea6\u8fc7\u5c0f\u91cd\u542f\u4e0b\u8f7d\u7684":0,"\u5bf9\u4e8emnist\u800c\u8a00":20,"\u5bf9\u8be5\u8868\u793a\u8fdb\u884c\u975e\u7ebf\u6027\u53d8\u6362":10,"\u5bf9\u8c61":20,"\u5bf9\u8c61convert":23,"\u5c06\u4f1a\u6d4b\u8bd5\u914d\u7f6e\u6587\u4ef6\u4e2dtest":10,"\u5c06\u5b57\u5178\u5b58\u5165\u4e86set":20,"\u5c06\u5bbf\u4e3b\u673a\u76848022\u7aef\u53e3\u6620\u5c04\u5230container\u768422\u7aef\u53e3\u4e0a":6,"\u5c06\u6570\u636e\u7ec4\u5408\u6210batch\u8bad\u7ec3":20,"\u5c06\u6587\u4ef6\u7684\u7edd\u5bf9\u8def\u5f84\u6216\u76f8\u5bf9\u8def\u5f84":19,"\u5c06\u8bc4\u8bba\u5206\u4e3a\u597d\u8bc4":10,"\u5c06\u8be5\u53e5\u8bdd\u5305\u542b\u7684\u6240\u6709\u5355\u8bcd\u5411\u91cf\u6c42\u5e73\u5747\u5f97\u5230\u53e5\u5b50\u7684\u8868\u793a":10,"\u5c06ssh\u88c5\u5165\u7cfb\u7edf\u5185\u5e76\u5f00\u542f\u8fdc\u7a0b\u8bbf\u95ee":6,"\u5c31":20,"\u5c31\u50cf\u8fd9\u4e2a\u6837\u4f8b\u4e00\u6837":20,"\u5c31\u662f\u5c06\u8fd9\u4e9b\u52a8\u6001\u5e93\u52a0\u5230\u73af\u5883\u53d8\u91cf\u91cc\u9762":7,"\u5c5e\u6027":20,"\u5dee\u8bc4":10,"\u5e38\u89c1\u7684\u8f93\u51fa\u683c\u5f0f\u4e3a":18,"\u5e76\u4e14":20,"\u5e76\u4e14\u4f7f\u7528\u5173\u952e\u8bcd":20,"\u5e76\u4e14\u5220\u9664container\u4e2d\u7684\u6570\u636e":6,"\u5e76\u4e14\u5728\u5185\u5b58\u8db3\u591f":20,"\u5e76\u4e14\u6807\u8bb0process\u51fd\u6570\u662f\u4e00\u4e2adataprovid":20,"\u5e76\u4f7f\u7528\u4e86dropout":10,"\u5e76\u572823\u884c\u8fd4\u56de\u7ed9paddlepaddle\u8fdb\u7a0b":20,"\u5e76\u5c06\u6bcf\u884c\u8f6c\u6362\u6210\u548c":20,"\u5e76\u63d0\u4f9b":6,"\u5e76\u63d0\u4f9b\u4e86\u7b80\u5355\u7684cache\u529f\u80fd":20,"\u5e76\u8bbe\u7f6e\u597d\u5bf9\u5e94\u7684\u73af\u5883\u53d8\u91cf":7,"\u5e76\u9010\u6e10\u5c55\u793a\u66f4\u52a0\u6df1\u5165\u7684\u529f\u80fd":10,"\u5e8f\u5217\u6a21\u578b\u6570\u636e\u63d0\u4f9b":19,"\u5e8f\u5217\u6a21\u578b\u662f\u6307\u6570\u636e\u7684\u67d0\u4e00\u7ef4\u5ea6\u662f\u4e00\u4e2a\u5e8f\u5217\u5f62\u5f0f":20,"\u5e93\u7684\u8bdd":7,"\u5f0f":20,"\u5f15\u7528\u7684dataprovider\u662f":20,"\u5f53\u51fd\u6570\u8fd4\u56de\u7684\u65f6\u5019":20,"\u5f53\u524dlog_period\u4e2abatch\u6240\u6709\u6837\u672c\u7684\u5e73\u5747\u5206\u7c7b\u9519\u8bef\u7387":10,"\u5f53\u524dlog_period\u4e2abatch\u6240\u6709\u6837\u672c\u7684\u5e73\u5747cost":10,"\u5f53\u7136":19,"\u5f53\u8c03":20,"\u5f88":10,"\u5f97\u5230\u7ed3\u679c":7,"\u5feb\u901f\u5165\u95e8":11,"\u5ff5\u662f":20,"\u60a8\u53ef\u4ee5\u4f7f\u7528":6,"\u60a8\u5c31\u53ef\u4ee5\u8fdc\u7a0b\u7684\u4f7f\u7528paddlepaddle\u5566":6,"\u60a8\u9700\u8981\u5728\u673a\u5668\u4e2d\u5b89\u88c5\u597ddocker":6,"\u60a8\u9700\u8981\u8fdb\u5165\u955c\u50cf\u8fd0\u884cpaddlepaddl":6,"\u60c5\u611f\u5206\u6790":9,"\u60f3\u8981\u8fd0\u884cpaddlepaddl":6,"\u6210\u4e3a\u7ef4\u5ea6\u4e3ahidden":10,"\u6211\u4eec\u4ece\u63d0\u524d\u7ed9\u5b9a\u7684\u7c7b\u522b\u96c6\u5408\u4e2d\u9009\u62e9\u5176\u6240\u5c5e\u7c7b":10,"\u6211\u4eec\u4ee5\u6587\u672c\u5206\u7c7b\u95ee\u9898\u4f5c\u4e3a\u80cc\u666f":10,"\u6211\u4eec\u4f7f\u7528":10,"\u6211\u4eec\u5728\u6b64\u603b":10,"\u6211\u4eec\u5c06\u4ee5\u57fa\u672c\u7684\u903b\u8f91\u56de\u5f52\u7f51\u7edc\u4f5c\u4e3a\u8d77\u70b9":10,"\u6211\u4eec\u5c06\u5728\u540e\u9762\u4ecb\u7ecd\u8bad\u7ec3\u548c\u9884\u6d4b\u7684\u6d41\u7a0b\u7684\u811a\u672c":10,"\u6211\u4eec\u5c06\u8bad\u7ec3\u7684\u547d\u4ee4\u884c\u4fdd\u5b58\u5728\u4e86":10,"\u6211\u4eec\u63a8\u8350\u4f7f\u7528docker\u955c\u50cf\u6765\u90e8\u7f72\u73af\u5883":5,"\u6211\u4eec\u63d0\u4f9b\u4e866\u4e2adocker":6,"\u6211\u4eec\u63d0\u4f9b\u4e86\u4e00\u4e2a\u5de5\u5177\u7c7bdataproviderconvert":23,"\u6211\u4eec\u8fdb\u5165\u5230\u8bad\u7ec3\u90e8\u5206":10,"\u6211\u4eec\u91c7\u7528\u5355\u5c42lstm\u6a21\u578b":10,"\u6211\u4eec\u968f\u65f6\u603b\u7ed3\u4e86\u5404\u4e2a\u7f51\u7edc\u7684\u590d\u6742\u5ea6\u548c\u6548\u679c":10,"\u6216\u5176\u4ed6":10,"\u6216\u8005":6,"\u6216\u800510g\u8fd9\u6837\u7684\u5355\u4f4d":0,"\u6216\u8005\u5728python":20,"\u6216\u8005\u6570\u636e\u5e93\u8fde\u63a5\u5730\u5740\u7b49\u7b49":19,"\u6216\u8005\u8bbe\u7f6e\u4e3anone":19,"\u6216\u8005\u9700\u8981\u66f4\u9ad8\u7684\u6548\u7387":19,"\u6216\u8005\u9ad8\u6027\u80fd\u7684":6,"\u6240\u4ee5":[20,23],"\u6240\u4ee5\u5728cpu\u7684\u8fd0\u7b97\u6027\u80fd\u4e0a\u5e76\u4e0d\u4f1a\u6709\u4e25\u91cd\u7684\u5f71\u54cd":6,"\u6240\u4ee5\u5982\u679c\u5bf9\u4e8e\u5185\u5b58\u6bd4\u8f83\u5c0f\u7684\u673a\u5668":20,"\u6240\u4ee5\u5982\u679c\u60f3\u8981\u5728\u540e\u53f0\u542f\u7528ssh":6,"\u6240\u4ee5\u5c06":20,"\u6240\u4ee5\u63a8\u8350\u4f7f\u7528\u663e\u5f0f\u6307\u5b9a\u8fd4\u56de\u503c\u548c\u6570\u636e\u5bf9\u5e94\u5173\u7cfb":20,"\u6240\u4ee5\u6700\u4f73\u5b9e\u8df5\u63a8\u8350\u4e0d\u8981\u5c06\u6bcf\u4e00\u4e2a\u6837\u672c\u90fd\u653e\u5165train":20,"\u6240\u4ee5\u7528\u4e8e\u9884\u6d4b\u7684\u914d\u7f6e\u6587\u4ef6\u8981\u505a\u76f8\u5e94\u7684\u4fee\u6539":23,"\u6240\u4ee5\u8f93\u51fa\u7684value\u5305\u542b\u4e24\u4e2a\u5411\u91cf":23,"\u6240\u4ee5gpu\u5728\u8fd0\u7b97\u6027\u80fd\u4e0a\u4e5f\u4e0d\u4f1a\u6709\u4e25\u91cd\u7684\u5f71\u54cd":6,"\u6240\u4ee5init_hook\u5c3d\u91cf\u4f7f\u7528":20,"\u6240\u6709\u5b57\u7b26\u90fd\u5c06\u8f6c\u6362\u4e3a\u8fde\u7eed\u6574\u6570\u8868\u793a\u7684id\u4f20\u7ed9\u6a21\u578b":10,"\u6240\u6709\u6587\u4ef6\u5217\u8868":20,"\u6240\u6709\u7684":1,"\u6240\u6709\u7684\u4e0b\u8f7d\u7ebf\u7a0b\u5173\u95ed\u65f6":0,"\u6240\u6709\u914d\u7f6e\u5728":10,"\u6240\u8c13\u65f6\u95f4\u6b65\u4fe1\u606f":20,"\u624d\u4f1a\u91ca\u653e\u8be5\u6bb5\u5185\u5b58":20,"\u624d\u4f1astop":20,"\u6253\u5370\u7684\u65e5\u5fd7\u53d8\u591a":1,"\u6267\u884c":0,"\u6267\u884c\u5982\u4e0b\u547d\u4ee4\u5373\u53ef\u4ee5\u5173\u95ed\u8fd9\u4e2acontain":6,"\u6267\u884c\u65b9\u6cd5\u5982\u4e0b":6,"\u62a5\u9519":7,"\u62fc\u63a5\u6210\u4e00\u4e2a\u65b0\u7684\u5411\u91cf\u8868\u793a":10,"\u6307\u4ee4\u96c6\u7684cpu\u8fd0\u884cpaddlepaddl":7,"\u6307\u4ee4\u96c6\u7684x86":7,"\u6307\u5b9a\u521d\u59cb\u5316\u6a21\u578b\u8def\u5f84":10,"\u6307\u5b9a\u751f\u6210\u6570\u636e\u7684\u51fd\u6570":10,"\u6307\u5b9a\u8bad\u7ec3":10,"\u6307\u5b9abatch":10,"\u6307\u5b9aoutputs\u8f93\u51fa\u6982\u7387\u5c42":10,"\u6389\u7f16\u8bd1\u76ee\u5f55\u540e":1,"\u63a5\u4e0b\u6765\u4f7f\u7528":23,"\u63a5\u53e3\u4f7f\u7528\u591a\u7ebf\u7a0b\u8bfb\u53d6\u6570\u636e":20,"\u63a8\u8350\u4f7f\u7528\u5c06\u672c\u5730\u7f51\u5361":6,"\u63a8\u8350\u4f7f\u7528\u6700\u65b0\u7248\u672c\u7684cudnn":1,"\u63a8\u8350\u6e05\u7406":1,"\u63a8\u8350\u76f4\u63a5\u653e\u7f6e\u5230\u8bad\u7ec3\u76ee\u5f55":19,"\u63a8\u8350\u8bbe\u7f6e":20,"\u63cf\u8ff0":1,"\u63cf\u8ff0\u4e86docker":0,"\u6548\u679c\u4e00\u81f4":20,"\u6548\u679c\u603b\u7ed3":10,"\u6559\u7a0b":10,"\u6570\u636e":20,"\u6570\u636e\u4f20\u8f93\u65e0\u9700label\u6570\u636e":10,"\u6570\u636e\u5904\u7406python\u6587\u4ef6\u540d":10,"\u6570\u636e\u5982\u4f55\u5b58\u50a8\u7b49\u7b49":20,"\u6570\u636e\u63d0\u4f9b":19,"\u6570\u636e\u6587\u4ef6\u5b58\u653e\u5728\u672c\u5730\u78c1\u76d8\u4e2d":19,"\u6570\u636e\u662f\u7ed9\u4e00\u6bb5\u82f1\u6587\u6587\u672c":20,"\u6570\u636e\u683c\u5f0f\u548c\u8be6\u7ec6\u6587\u6863\u8bf7\u53c2\u8003":10,"\u6587\u4ef6":20,"\u6587\u4ef6\u4e2d":10,"\u6587\u672c\u4e2d\u7684\u5355\u8bcd\u7528\u7a7a\u683c\u5206\u9694":10,"\u6587\u672c\u4fe1\u606f\u5c31\u662f\u4e00\u4e2a\u5e8f\u5217":20,"\u6587\u672c\u5206\u7c7b\u95ee\u9898":10,"\u6587\u672c\u5377\u79ef\u5206\u4e3a\u4e09\u4e2a\u6b65\u9aa4":10,"\u6587\u672c\u751f\u6210":9,"\u65b0\u5199layer":11,"\u65b9\u4fbf\u8c03\u8bd5\u4f7f\u7528":13,"\u65b9\u4fbf\u90e8\u7f72\u5206\u53d1":13,"\u65e0\u9700label\u76f8\u5173\u7684\u5c42":10,"\u65f6\u5e8f\u6a21\u578b\u5373\u4e3arnn\u6a21\u578b":10,"\u65f6\u5e8f\u6a21\u578b\u5747\u4f7f\u7528\u8be5\u811a\u672c":10,"\u662f":1,"\u662f\u4e00\u4e2abatch":20,"\u662f\u4e00\u4e2apython\u7684":20,"\u662f\u4e00\u4e2aswig\u5c01\u88c5\u7684paddlepaddle\u5305":6,"\u662f\u4e0d\u662f\u5f88\u7b80\u5355\u5462":20,"\u662f\u4e2adataprovider\u662f\u4e0d\u662f\u8981\u505ashuffl":20,"\u662f\u4ec0\u4e48\u4e5f\u6ca1\u5173\u7cfb":20,"\u662f\u4ece\u8bad\u7ec3\u914d\u7f6e\u4f20\u5165\u7684dict\u5bf9\u8c61":20,"\u662f\u51e0\u4e4e\u4e0d\u5360\u5185\u5b58\u7684":20,"\u662f\u521d\u59cb\u5316\u65f6\u8c03\u7528\u7684\u51fd\u6570":20,"\u662f\u540c\u4e00\u4e2a\u5bf9\u8c61":20,"\u662f\u5426\u4f7f\u7528\u53cc\u7cbe\u5ea6\u6d6e\u70b9\u6570":1,"\u662f\u5426\u4f7f\u7528\u8fd0\u884c\u65f6\u52a8\u6001\u52a0\u8f7dcuda\u52a8\u6001\u5e93":1,"\u662f\u5426\u4f7f\u7528gflags":1,"\u662f\u5426\u4f7f\u7528glog":1,"\u662f\u5426\u5185\u5d4cpython\u89e3\u91ca\u5668":1,"\u662f\u5426\u5bfb\u627e\u5230cuda\u5de5\u5177\u94fe":1,"\u662f\u5426\u5f00\u542f\u5355\u5143\u6d4b\u8bd5":1,"\u662f\u5426\u5f00\u542f\u8ba1\u65f6\u529f\u80fd\u5f00\u542f\u8ba1\u65f6\u529f\u80fd\u4f1a\u5bfc\u81f4\u8fd0\u884c\u7565\u6162":1,"\u662f\u5426\u5f00\u542fgpu\u529f\u80fd":0,"\u662f\u5426\u5f00\u542frdma\u652f\u6301":1,"\u662f\u5426\u7f16\u8bd1\u4e2d\u6587\u6587\u6863":1,"\u662f\u5426\u7f16\u8bd1\u542b\u6709avx\u6307\u4ee4\u96c6\u7684paddlepaddle\u4e8c\u8fdb\u5236":1,"\u662f\u5426\u7f16\u8bd1\u65f6\u8fdb\u884c\u4ee3\u7801\u98ce\u683c\u68c0\u67e5":1,"\u662f\u5426\u7f16\u8bd1\u82f1\u6587\u6587\u6863":1,"\u662f\u5426\u7f16\u8bd1gpu\u652f\u6301":1,"\u662f\u5426\u7f16\u8bd1python\u7684swig\u63a5\u53e3":1,"\u662f\u5728\u8fd0\u884c\u65f6\u6267\u884c\u7684":20,"\u662f\u6570\u636e\u7f13\u5b58\u7684\u7b56\u7565":20,"\u662f\u6570\u636e\u8f93\u5165\u683c\u5f0f":20,"\u662f\u8bbe\u7f6e\u8fd9\u4e2adataprovider\u8fd4\u56de\u4ec0\u4e48\u6837\u7684\u6570\u636e":20,"\u662f\u8bbe\u7f6edataprovider\u5728\u5185\u5b58\u4e2d\u6682\u5b58\u7684\u6570\u636e\u6761\u6570":20,"\u662f\u8bbe\u7f6edataprovider\u5728\u5185\u5b58\u4e2d\u6700\u5c0f\u6682\u5b58\u7684\u6570\u636e\u6761\u6570":20,"\u662fpaddlepaddle\u8d1f\u8d23\u63d0\u4f9b\u6570\u636e\u7684\u6a21\u5757":19,"\u662fpython\u7684\u4e00\u4e2a\u5173\u952e\u8bcd":20,"\u663e":10,"\u66f4\u591a\u7684\u5b89\u88c5\u5305paddlepaddle\u4f1a\u5728\u8fd1\u671f\u63d0\u4f9b":7,"\u66f4\u8be6\u7ec6\u7528\u4f8b\u8bf7\u53c2\u8003\u6587\u6863":10,"\u66f4\u8be6\u7ec6\u7684\u4ecb\u7ecd\u8bf7\u53c2\u8003\u5404\u4e2a\u547d\u4ee4\u7684\u547d\u4ee4\u884c\u53c2\u6570\u6587\u6863":13,"\u66f4\u8be6\u7ec6\u7684\u7f51\u7edc\u914d\u7f6e":10,"\u6700\u4f4e\u7ebf\u7a0b\u7684\u4e0b\u8f7d\u901f\u5ea6":0,"\u6700\u540e\u4f7f\u7528":23,"\u6709100\u4e2a\u8bad\u7ec3\u6587\u4ef6":20,"\u6709\u503c\u7684\u4f4d\u7f6e\u53ea\u80fd\u53d61":20,"\u6709\u503c\u7684\u90e8\u5206\u53ef\u4ee5\u662f\u4efb\u4f55\u6d6e\u70b9\u6570":20,"\u6709\u90e8\u5206\u53c2\u6570\u662fpaddle\u81ea\u52a8\u751f\u6210\u7684":20,"\u672c\u8282\u6211\u4eec\u5c06\u4e13\u6ce8\u4e8e\u7f51\u7edc\u7ed3\u6784\u7684\u4ecb\u7ecd":10,"\u6765\u63a5\u53d7\u4e0d\u4f7f\u7528\u7684":20,"\u6765\u786e\u5b9a\u5bf9\u5e94\u5173\u7cfb":20,"\u6765\u81ea\u5b9a\u4e49\u4f20\u6570\u636e\u7684\u8fc7\u7a0b":19,"\u6765\u8bf4\u660e\u7b80\u5355\u7684pydataprovider\u5982\u4f55\u4f7f\u7528":20,"\u6765\u8fdb\u884c\u8bad\u7ec3":6,"\u6765\u914d\u7f6ecudnn\u7684\u5b89\u88c5\u8def\u5f84":1,"\u6784\u9020gradientmachin":23,"\u6790\u597d\u7684\u914d\u7f6e\u521b\u5efa\u795e\u7ecf\u7f51\u7edc":23,"\u6807\u7b7e\u662f0":20,"\u6837\u4f8b\u6570\u636e\u4e3a":20,"\u6837\u4f8b\u6570\u636e\u5982\u4e0b":20,"\u6837\u672c":20,"\u6839\u636e\u4e0a\u4e00\u6b65\u89e3":23,"\u6839\u636e\u6a21\u578b\u914d\u7f6e\u6587\u4ef6\u4e2d":20,"\u683c\u5f0f\u5982\u4e0b":10,"\u68d2":10,"\u6a21\u578b\u5b58\u50a8\u8def\u5f84":10,"\u6a21\u578b\u8bad\u7ec3\u4f1a\u770b\u5230\u8fd9\u6837\u7684\u65e5\u5fd7":10,"\u6a21\u578b\u914d\u7f6e":11,"\u6b22\u8fce\u5927\u5bb6\u8d21\u732e\u5404\u4e2a\u53d1\u884c\u7248\u7684\u5b89\u88c5\u5305":7,"\u6b63\u6837\u672c":10,"\u6bcf\u4e00\u4e2a\u4efb\u52a1\u6d41\u7a0b\u90fd\u53ef\u4ee5\u5206\u4e3a\u5982\u4e0b5\u4e2a\u57fa\u7840\u90e8\u5206":10,"\u6bcf\u4e00\u6761\u8bad\u7ec3\u6570\u636e\u90fd\u662f\u4e00\u4e2a\u6587\u4ef6":20,"\u6bcf\u4e00\u884c":20,"\u6bcf\u4e2a\u5c42\u90fd\u6709\u4e00\u4e2a\u6216\u591a\u4e2ainput":10,"\u6bcf\u4e2agenerator\u5728\u6ca1\u6709\u8c03\u7528\u7684\u65f6\u5019":20,"\u6bcf\u4e2apass\u7684\u7b2c0\u4e2abatch\u5230\u5f53\u524dbatch\u6240\u6709\u6837\u672c\u7684\u5e73\u5747\u5206\u7c7b\u9519\u8bef\u7387":10,"\u6bcf\u4e2apass\u7684\u7b2c0\u4e2abatch\u5230\u5f53\u524dbatch\u6240\u6709\u6837\u672c\u7684\u5e73\u5747cost":10,"\u6bcf\u6b21\u90fd\u4f1a\u4ecepython\u7aef\u8bfb\u53d6\u6570\u636e":20,"\u6bcf\u884c\u4fdd\u5b58\u4e00\u6761\u6837\u672c":10,"\u6bcf\u9694\u591a\u5c11batch\u6253\u5370\u4e00\u6b21\u65e5\u5fd7":10,"\u6bd4\u5982\u901a\u8fc7\u7528\u6237\u5bf9\u7535\u5b50\u5546\u52a1\u7f51\u7ad9\u8bc4\u8bba":10,"\u6bd4\u8f83\u53ef\u80fd\u7684\u547d\u4ee4\u5982\u4e0b":7,"\u6ca1\u6709\u4f5c\u7528":20,"\u6ca1\u6709\u8bbe\u7f6e":7,"\u6ce8\u610f":[0,5,20],"\u6d41\u7a0b\u5982\u4e0b":10,"\u6d4b\u8bd5\u6570\u636e":10,"\u6d4b\u8bd5\u7684\u65f6\u5019\u9ed8\u8ba4\u4e0dshuffl":20,"\u6d4b\u8bd5\u811a\u672c\u5982\u4e0b":10,"\u6e90\u7801":10,"\u6e90\u7801\u548c\u8fd0\u884cdemo\u7684\u5fc5\u8981\u4f9d\u8d56":6,"\u6e90\u7801\u6839\u76ee\u5f55":0,"\u6fc0\u6d3b\u51fd\u6570\u7c7b\u578b":10,"\u7136\u540e\u6267\u884c\u5982\u4e0b\u547d\u4ee4\u5373\u53ef\u5b8c\u6210\u5b89\u88c5":7,"\u7136\u540e\u8fd0\u884c\u8fd9\u4e2acontainer\u5373\u53ef":6,"\u751f\u6210\u5404\u4e2a\u5e73\u53f0\u7684makefil":1,"\u75280\u548c1\u8868\u793a":20,"\u7528\u4e86\u4e24\u4e2a\u6708\u4e4b\u540e\u8fd9\u4e2a\u663e\u793a\u5668\u5c4f\u5e55\u788e\u4e86":10,"\u7528\u6237\u4e5f\u53ef\u4ee5\u5728c":19,"\u7528\u6237\u4e5f\u53ef\u4ee5\u663e\u5f0f\u6307\u5b9a\u8fd4\u56de\u7684\u6570\u636e\u5bf9\u5e94\u5173\u7cfb":20,"\u7528\u6237\u53ef\u4ee5\u4f7f\u7528python\u7684":19,"\u7528\u6237\u53ef\u4ee5\u6839\u636e\u8bad\u7ec3log\u9009\u62e9test\u7ed3\u679c\u6700\u597d\u7684\u6a21\u578b\u6765\u9884\u6d4b":10,"\u7528\u6237\u540d\u4e3a":6,"\u7528\u6237\u5728dataprovider\u4e2d\u9700\u8981\u5b9e\u73b0\u5982\u4f55\u8bbf\u95ee\u5176\u4e2d\u6bcf\u4e00\u4e2a\u6587\u4ef6":19,"\u7528\u6237\u5b9a\u4e49\u7684\u53c2\u6570\u4f7f\u7528args\u5728\u8bad\u7ec3\u914d\u7f6e\u4e2d\u8bbe\u7f6e":20,"\u7528\u6237\u63a5\u53e3":11,"\u7528\u6237\u9700\u8981\u5148\u5c06paddlepaddle\u5b89\u88c5\u5305\u4e0b\u8f7d\u5230\u672c\u5730":7,"\u7528\u6765\u505a\u9884\u6d4b\u548c\u7b80\u5355\u7684\u5b9a\u5236\u5316":6,"\u7528\u8fc7\u4e00\u6b21\u7684\u65f6\u5019":20,"\u7531\u4e8e\u6570\u636e\u662f\u4e24\u6761":23,"\u7531\u4e8edocker\u662f\u57fa\u4e8e\u5bb9\u5668\u7684\u8f7b\u91cf\u5316\u865a\u62df\u65b9\u6848":6,"\u7531\u4e8epaddlepaddle\u7684docker\u955c\u50cf\u5e76\u4e0d\u5305\u542b\u4efb\u4f55\u9884\u5b9a\u4e49\u7684\u8fd0\u884c\u547d\u4ee4":6,"\u7531\u6613\u5230\u96be\u5c55\u793a4\u79cd\u4e0d\u540c\u7684\u7f51\u7edc\u914d\u7f6e":10,"\u7684":10,"\u7684\u540d\u5b57":20,"\u7684\u5b89\u88c5\u6587\u6863":6,"\u7684\u60c5\u51b5\u4e0b\u8d8a\u5927\u8d8a\u597d":20,"\u7684\u6587\u6863":20,"\u7684\u65f6\u5019\u5982\u679c\u62a5\u4e00\u4e9b\u4f9d\u8d56\u672a\u627e\u5230\u7684\u9519\u8bef\u662f\u6b63\u5e38\u7684":7,"\u7684\u662f":20,"\u7684\u673a\u5668\u4e0a\u8fdb\u884c":0,"\u7684\u7f51\u6865\u6765\u8fdb\u884c\u7f51\u7edc\u901a\u4fe1":6,"\u7684demo\u5b66\u4e60\u5982\u4f55\u8fdb\u884c\u591a\u673a\u8bad\u7ec3":10,"\u7684docker\u53ef\u80fd\u7f3a\u4e4f":0,"\u7684matrix":23,"\u7684python\u5305\u662fpaddlepaddle\u7684\u8bad\u7ec3\u4e3b\u8981\u7a0b\u5e8f":6,"\u7684python\u5305\u6765\u505a\u914d\u7f6e\u6587\u4ef6\u89e3\u6790\u7b49\u5de5\u4f5c":6,"\u7684python\u9884\u6d4b\u8fc7\u7a0b":10,"\u76ee\u524dpaddlepaddle\u7684\u5b89\u88c5\u5305\u53ea\u652f\u6301":7,"\u76ee\u524dpaddlepaddle\u7684\u5b89\u88c5\u5305\u8fd8\u5904\u5728pr":5,"\u76ee\u5f55":10,"\u76ee\u5f55\u4e0b":0,"\u76f4\u63a5\u63d0\u53d6\u51fa\u795e\u7ecf\u7f51\u7edcoutput\u5c42\u7684\u8f93\u51fa\u7ed3\u679c":23,"\u76f8\u5173\u547d\u4ee4\u4e3a":6,"\u76f8\u5173\u7684\u6982":20,"\u76f8\u5bf9\u4e8epaddlepaddle\u7a0b\u5e8f\u8fd0\u884c\u65f6\u7684\u8def\u5f84":19,"\u77e5\u9053\u5982\u4f55\u4ece":20,"\u793a":10,"\u7a0b\u5e8f\u6216\u8005\u81ea\u5b9a\u4e49\u4e00\u4e2a\u542b\u6709\u542f\u52a8\u811a\u672c\u7684imag":6,"\u7aef\u81ea\u5b9a\u4e49\u4e00\u4e2a":19,"\u7b2c\u4e00\u4e2a\u53c2\u6570\u662f":20,"\u7b2c\u4e00\u4e2apass\u4f1a\u4ecepython\u7aef\u8bfb\u53d6\u6570\u636e":20,"\u7b2c\u4e00\u6bb5\u6570\u636e\u4e3a\u8fd9\u5f20\u56fe\u7247\u7684label":20,"\u7b2c\u4e8c\u4e2a\u53c2\u6570\u662ffilenam":20,"\u7b2c\u4e8c\u6bb5\u6570\u636e\u4e3a\u8fd9\u4e2a\u56fe\u7247\u7684\u50cf\u7d20\u503c":20,"\u7b80\u5355\u4f18\u5316":0,"\u7b80\u5355\u7684\u4f7f\u7528":19,"\u7b80\u5355\u7684\u4f7f\u7528\u573a\u666f":19,"\u7b80\u5355\u7684\u4f7f\u7528\u6837\u4f8b\u4e3a":0,"\u7b80\u5355\u7684\u542b\u6709ssh\u7684dockerfile\u5982\u4e0b":6,"\u7b80\u5355\u7684pydataprovider\u6837\u4f8b\u5c31\u8bf4\u660e\u5b8c\u6bd5\u4e86":20,"\u7c7b\u522bid":10,"\u7c7b\u522bid\u7684\u6570\u636e\u7c7b\u578b":10,"\u7c7b\u578b\u6765\u8bbe\u7f6e":20,"\u7ec3":13,"\u7ed3\u4e0a\u8ff0\u7f51\u7edc\u7ed3\u6784\u5728amazon":10,"\u7ee7\u7eed\u8bad\u7ec3":20,"\u7ef4\u5ea6\u4e3aword":10,"\u7ef4\u5ea6\u662f\u7c7b\u522b\u4e2a\u6570":10,"\u7ef4\u5ea6\u662f\u8bcd\u5178\u5927\u5c0f":10,"\u7f13\u5b58\u8bad\u7ec3\u6570\u636e\u5230\u5185\u5b58":20,"\u7f16\u8bd1\u53c2\u6570\u9009\u9879\u6587\u4ef6":18,"\u7f16\u8bd1\u6587\u6863":7,"\u7f16\u8bd1\u9009\u9879":1,"\u7f16\u8bd1\u9009\u9879\u5217\u8868\u5982\u4e0b":1,"\u7f16\u8bd1paddlepaddl":7,"\u7f16\u8bd1paddlepaddle\u7684gpu\u7248\u672c\u5e76\u4e0d\u9700\u8981\u4e00\u5b9a\u5728\u5177\u6709gpu":0,"\u7f51\u7edc\u540d\u79f0":10,"\u7f51\u7edc\u914d\u7f6e":10,"\u7f6e\u8fd9\u4e9b\u53d8\u91cf":1,"\u800c":6,"\u800c\u4e09\u79cd\u5e8f\u5217\u6a21\u5f0f\u4e3a":20,"\u800c\u4e0d\u4f7f\u7528docker":6,"\u800c\u4e0d\u7528\u5173\u5fc3\u6570\u636e\u5982\u4f55\u4f20\u8f93\u7ed9paddlepaddl":20,"\u800c\u4e14\u9884\u6d4b\u7f51\u7edc\u901a\u5e38\u76f4\u63a5\u8f93\u51fa\u6700\u540e\u4e00\u5c42\u7684\u7ed3\u679c\u800c\u4e0d\u662f\u50cf\u8bad\u7ec3\u65f6\u4e00\u6837\u4ee5cost":23,"\u800c\u5728":[1,20],"\u800c\u5982\u679c\u6309\u987a\u5e8f\u8c03\u7528\u8fd9\u4e9bgenerator\u5c31\u4e0d\u4f1a\u51fa\u73b0\u8fd9\u4e2a\u95ee\u9898":20,"\u800c\u662f\u5c06\u6837\u672c\u7684\u5730\u5740\u653e\u5165\u53e6\u4e00\u4e2a\u6587\u672c":20,"\u800c\u6ca1\u6709\u6d4b\u8bd5\u6570\u636e":20,"\u800c\u7279\u5f81\u5373\u4e3a":20,"\u800c\u8fd9\u4e2a\u4e00\u822c\u8bf4\u660epaddlepaddle\u5df2\u7ecf\u5b89\u88c5\u5b8c\u6bd5\u4e86":7,"\u800c\u8fd9\u4e2a\u53d8\u91cf\u63a8\u8350\u5927\u4e8e\u8bad\u7ec3\u7684batch":20,"\u800c\u8fd9\u4e2acontext\u53ef\u80fd\u4f1a\u975e\u5e38":20,"\u800c\u975e\u9759\u6001\u52a0\u8f7dcuda\u52a8\u6001\u5e93":1,"\u800cgpu\u7684\u9a71\u52a8\u548c\u8bbe\u5907\u5168\u90e8\u6620\u5c04\u5230\u4e86\u5bb9\u5668\u5185":6,"\u800cpaddlepaddle\u8fdb\u7a0b\u5e2e\u52a9\u7528\u6237\u505a\u4e86":20,"\u811a\u672c":6,"\u811a\u672c\u53ef\u4ee5\u542f\u52a8paddlepaddle\u7684\u8bad\u7ec3\u8fdb\u7a0b\u548cpserv":6,"\u811a\u672c\u548c":6,"\u811a\u672c\u7c7b\u4f3c\u4e8e":6,"\u81ea\u5b9a\u4e49\u4e00\u4e2adataprovid":19,"\u81f3\u5c11\u5177\u67093":6,"\u81f3\u6b64":[6,20],"\u83b7\u53d6\u5229\u7528one":10,"\u83b7\u53d6\u6bcf\u4e2a\u5355\u8bcd\u5de6\u53f3\u5404k\u4e2a\u8fd1\u90bb":10,"\u83b7\u53d6\u8be5\u6761\u6837\u672c\u7c7b\u522bid":10,"\u8868\u793a\u6574\u6570\u6807\u7b7e":20,"\u8868\u793a\u662f\u5426\u5141\u8bb8paddle\u6682\u5b58\u7565\u5fae\u591a\u4f59pool_size\u7684\u6570\u636e":20,"\u8868\u793a\u7a00\u758f\u7684\u5411\u91cf":20,"\u8868\u793a\u7a00\u758f\u7684\u96f6\u4e00\u5411\u91cf":20,"\u8868\u793a\u7a20\u5bc6\u7684\u6d6e\u70b9\u6570\u5411\u91cf":20,"\u8868\u793a\u8fc7\u4e8620\u4e2abatch":10,"\u8868\u793a\u8fc7\u4e862560\u4e2a\u6837\u672c":10,"\u8868\u793a\u8fd9\u4e2adataprovider\u662f\u8bad\u7ec3\u7528\u7684dataprovider\u6216\u8005\u6d4b\u8bd5\u7528\u7684":20,"\u89e3\u51b3\u529e\u6cd5\u662f\u5c06cuda":7,"\u89e3\u51b3\u65b9\u6cd5\u5f88\u7b80\u5355":7,"\u89e3\u6790\u8bad\u7ec3\u65f6\u7684\u914d\u7f6e\u6587\u4ef6":23,"\u89e3\u91ca":10,"\u8ba9\u795e\u7ecf\u7f51\u7edc\u53ef\u4ee5\u8fdb\u884c\u8bad\u7ec3":19,"\u8bad\u7ec3":6,"\u8bad\u7ec3\u6570\u636e\u975e\u5e38\u591a\u7684\u60c5\u51b5\u4e0b":20,"\u8bad\u7ec3\u6587\u4ef6\u5217\u8868":19,"\u8bad\u7ec3\u65f6\u6240\u9700\u8bbe\u7f6e\u7684\u4e3b\u8981\u53c2\u6570\u5982\u4e0b":10,"\u8bad\u7ec3\u7684\u65f6\u5019\u9ed8\u8ba4shuffl":20,"\u8bad\u7ec3\u811a\u672c":10,"\u8bad\u7ec3\u811a\u672c\u5728":10,"\u8bad\u7ec3\u8f6e\u6b21":10,"\u8bb2\u6570\u636e\u53d1\u9001\u7ed9paddlepaddl":20,"\u8bbe\u7f6e\u4e0b\u5217\u7f16\u8bd1\u9009\u9879\u65f6":1,"\u8bbe\u7f6e\u6210":20,"\u8bbe\u7f6e\u6210\u4e86\u5e8f\u5217":20,"\u8bbe\u7f6e\u6210true\u7684\u8bdd":20,"\u8bbe\u7f6e\u8f93\u5165\u7c7b\u578b":20,"\u8bc4\u4f30\u4ea7\u54c1\u7684\u8d28\u91cf":10,"\u8bcd\u6027\u6807\u6ce8":9,"\u8be5\u5c42\u795e\u7ecf\u5143\u4e2a\u6570":10,"\u8be5\u6570\u636e":20,"\u8be5\u6784\u5efa\u811a\u672c\u5145\u5206\u8003\u8651\u4e86\u7f51\u7edc\u4e0d\u7a33\u5b9a\u7684\u60c5\u51b5":0,"\u8be5\u6a21\u578b\u4f9d\u7136\u662f\u4f7f\u7528\u903b\u8f91\u56de\u5f52\u5206\u7c7b\u7f51\u7edc\u7684\u6846\u67b6":10,"\u8be5\u76ee\u5f55\u4e0b\u6709\u4e24\u4e2a\u6587\u4ef6":0,"\u8be5\u811a\u672c\u7684\u4f7f\u7528\u65b9\u6cd5\u662f":0,"\u8be5image\u57fa\u4e8eubuntu":0,"\u8be5image\u7684\u6784\u5efa\u5728dock":0,"\u8be6\u60c5\u8bf7\u53c2\u8003":23,"\u8be6\u7ec6\u7684\u53c2\u6570\u89e3\u91ca\u5982\u4e0b\u9762\u8868\u683c":10,"\u8be6\u7ec6\u7684\u547d\u4ee4\u884c\u53c2\u6570\u8bf7\u53c2\u8003":23,"\u8be6\u7ec6\u7684cmake\u4f7f\u7528\u65b9\u6cd5\u53ef\u4ee5\u53c2\u8003":1,"\u8bf4\u660e":1,"\u8bf4\u660e\u547d\u4ee4\u884c\u53c2\u6570":7,"\u8bf7\u53c2\u8003":[6,7,20],"\u8bf7\u53c2\u8003\u4e0b\u8282refer":20,"\u8bf7\u53c2\u8003\u4e0b\u8ff0\u6587\u7ae0":19,"\u8bf7\u5b89\u88c5cuda":7,"\u8bfb\u5165\u89e3\u6790\u8bad\u7ec3\u914d\u7f6e":23,"\u8bfb\u53d6\u6570\u636e":20,"\u8c03\u7528":1,"\u8c03\u7528\u4e00\u6b21":20,"\u8c03\u7528\u7b2c\u4e8c\u6b21\u7684\u65f6\u5019":20,"\u8d1f\u6837\u672c":10,"\u8d1f\u8d23\u591a\u673a\u8bad\u7ec3\u4e2d\u7684\u53c2\u6570\u805a\u5408\u5de5\u4f5c":13,"\u8d1f\u9762\u60c5\u7eea\u4e24\u7c7b":20,"\u8def\u5f84\u53d8\u91cf\u4e3a":1,"\u8f93\u5165n\u4e2a\u5355\u8bcd":10,"\u8f93\u51fa\u4e3an\u4e2aword":10,"\u8fd0\u884c":[6,7],"\u8fd0\u884c\u4f7f\u7528\u7684cudnn\u5c3d\u91cf\u662f\u540c\u4e00\u4e2a\u7248\u672c":1,"\u8fd0\u884c\u8fd9\u4e2acontain":6,"\u8fd0\u884cpaddlepaddle\u7684gpu\u7248\u672c\u4e00\u5b9a\u8981\u5728\u5177\u6709cuda\u7684\u673a\u5668\u4e0a\u8fd0\u884c":0,"\u8fd4\u56de0":20,"\u8fd4\u56de\u4e00\u4e2alist\u6216\u8005tupl":20,"\u8fd4\u56de\u6570\u636e\u5728paddlepaddle\u4e2d\u662f\u4ec5\u4ec5\u8fd4\u56de\u4e00\u6761\u5b8c\u6574\u7684\u8bad\u7ec3\u6837\u672c":20,"\u8fd4\u56de\u7684\u987a\u5e8f\u9700\u8981\u548c":20,"\u8fd4\u56debatch_size\u7684\u5927\u5c0f":20,"\u8fd9":10,"\u8fd93\u4e2a\u5b50\u6b65\u9aa4\u53ef\u914d\u7f6e\u4e3a":10,"\u8fd9\u4e2a\u4e5f\u662fpaddlepaddle\u6240\u80fd\u591f\u4fdd\u8bc1\u7684shuffle\u7c92\u5ea6":20,"\u8fd9\u4e2a\u51fd\u6570\u4ee5\u4e00\u6761\u6570\u636e\u4e3a\u53c2\u6570":20,"\u8fd9\u4e2a\u51fd\u6570\u4f1a\u5728":20,"\u8fd9\u4e2a\u51fd\u6570\u5728\u521d\u59cb\u5316\u7684\u65f6\u5019\u4f1a\u88ab\u8c03\u7528":20,"\u8fd9\u4e2a\u51fd\u6570\u7684\u53c2\u6570\u662f":20,"\u8fd9\u4e2a\u521d\u59cb\u5316\u51fd\u6570\u5177\u6709\u5982\u4e0b\u53c2\u6570":20,"\u8fd9\u4e2a\u53c2\u6570\u5728\u8fd9\u4e2a\u6837\u4f8b\u91cc\u6ca1\u6709\u4f7f\u7528":20,"\u8fd9\u4e2a\u53c2\u6570\u88abpaddlepaddle\u8fdb\u7a0b\u4f20\u5165":20,"\u8fd9\u4e2a\u548c\u5728":20,"\u8fd9\u4e2a\u58f0\u660e\u57fa\u672c\u4e0a\u548cmnist\u7684\u6837\u4f8b\u4e00\u81f4":20,"\u8fd9\u4e2a\u5b57\u5178\u53ef\u4ee5\u5728":20,"\u8fd9\u4e2a\u5bf9\u5e94\u5173\u7cfb\u53ef\u80fd\u4e0d\u6b63\u786e":20,"\u8fd9\u4e2a\u5bf9\u8c61\u548cprocess\u7684\u7b2c\u4e00\u4e2a\u53c2\u6570\u4e00\u81f4":20,"\u8fd9\u4e2a\u5de5\u5177\u7c7b\u63a5\u6536\u548cpydataprovider2\u4e00\u6837\u7684\u8f93\u5165\u6570\u636e":23,"\u8fd9\u4e2a\u5e8f\u5217\u6a21\u578b\u6bd4\u8f83\u590d\u6742":20,"\u8fd9\u4e2a\u63a5\u53e3\u5e76\u4e0d\u7528\u6237\u53cb\u597d":23,"\u8fd9\u4e2a\u663e\u793a\u5668\u5f88\u68d2":10,"\u8fd9\u4e2a\u672c\u8eab\u4e0d\u662f\u4e00\u4e2a\u5f88":20,"\u8fd9\u4e2a\u6a21\u5757\u4e2d\u7684":20,"\u8fd9\u4e2a\u8bbe\u7f6e\u4e3a":20,"\u8fd9\u4e2a\u8f6f\u4ef6\u5305\u6587\u6863\u76f8\u5bf9\u5b8c\u5584":23,"\u8fd9\u4e2a\u95ee\u9898\u662fpydataprovider\u8bfb\u6570\u636e\u65f6\u5019\u7684\u903b\u8f91\u95ee\u9898":20,"\u8fd9\u4e9b\u53c2\u6570\u5305\u62ecpaddle\u5b9a\u4e49\u7684\u53c2\u6570":20,"\u8fd9\u4e9b\u53d8":1,"\u8fd9\u4e9b\u53d8\u91cf\u53ea\u5728\u7b2c\u4e00\u6b21cmake\u7684\u65f6\u5019\u6709\u6548":1,"\u8fd9\u4e9b\u53d8\u91cf\u5747\u53ef\u4ee5\u4f7f\u7528":1,"\u8fd9\u4e9b\u6d41\u7a0b\u4e2d\u7684\u6570\u636e\u4e0b\u8f7d":10,"\u8fd9\u6837\u505a\u53ef\u4ee5\u907f\u514d\u5f88\u591a\u6b7b\u9501\u95ee\u9898":20,"\u8fd9\u884c\u7684\u4f5c\u7528\u662f\u8bbe\u7f6edataprovider\u7684\u4e00\u4e9b\u5c5e\u6027":20,"\u8fd9\u91cc":20,"\u8fd9\u91cc\u4e3e\u4f8b\u7684\u6570\u636e\u662f\u82f1\u6587\u60c5\u611f\u5206\u7c7b\u7684\u6570\u636e":20,"\u8fd9\u91cc\u4ee5":10,"\u8fd9\u91cc\u4ee5mnist\u624b\u5199\u8bc6\u522b\u4e3a\u4f8b":20,"\u8fd9\u91cc\u53ef\u4ee5\u53c2\u8003paddle\u7684":18,"\u8fd9\u91cc\u6211\u4eec\u4f7f\u7528\u7b80\u5355\u7684\u6587\u672c\u6587\u4ef6\u8868\u793amnist\u56fe\u7247":20,"\u8fd9\u91cc\u6307\u5b9a\u8bcd\u5178":10,"\u8fd9\u91cc\u6ca1\u6709\u4ecb\u7ecd\u591a\u673a\u5206\u5e03\u5f0f\u8bad\u7ec3":10,"\u8fd9\u91cc\u7684":20,"\u8fd9\u91cc\u7684\u8f93\u5165\u7279\u5f81\u662f\u8bcdid\u7684\u5e8f\u5217":20,"\u8fd9\u91cc\u8981\u6ce8\u610f\u9884\u6d4b\u6570\u636e\u901a\u5e38":23,"\u8fd9\u91cc\u8bbe\u7f6e\u7684\u662f\u8fd4\u56de\u4e00\u4e2a":20,"\u8fd9\u91cc\u8bf4\u660e\u4e86\u8bad\u7ec3\u6570\u636e\u662f":20,"\u8fd9\u91cc\u91c7\u7528adam\u4f18\u5316\u65b9\u6cd5":10,"\u8fdb\u5165\u8be5\u6e90\u7801\u76ee\u5f55":0,"\u8fdb\u5165docker":6,"\u8fdc\u7a0b\u8bbf\u95ee":6,"\u8fde\u63a5\u8bf7\u53c2\u8003":10,"\u9009\u9879":1,"\u901a\u8fc7\u7f16\u8bd1\u65f6\u6307\u5b9a\u8def\u5f84\u6765\u5b9e\u73b0\u5f15\u7528\u5404\u79cdbla":1,"\u903b\u8f91\u56de\u5f52":10,"\u90a3\u4e48":20,"\u90a3\u4e48\u5728\u8bad\u7ec3\u8fc7\u7a0b\u4e2d":19,"\u90a3\u4e48\u5bf9\u5e94\u7684dataprovider\u65e2\u4e3a":20,"\u90a3\u4e48\u8fd9\u4e2a\u4e0b\u8f7d\u7ebf\u7a0b\u5c06\u4f1a\u5173\u95ed":0,"\u90a3\u4e48\u9700\u8981\u4ece\u6e90\u7801":7,"\u90a3\u4e48paddlepaddle\u4f1a\u7c97\u7565\u7684\u6839\u636elayer\u7684\u58f0\u660e\u987a\u5e8f":20,"\u90fd\u4f20\u9012\u7ed9process\u51fd\u6570":20,"\u914d\u7f6e\u4e86":20,"\u914d\u7f6e\u53c2\u6570\u914d\u7f6e\u7ed9dataprovider\u7684":20,"\u914d\u7f6e\u6587\u4ef6":10,"\u91cc\u4f1a\u7ee7\u7eed\u5b89\u88c5":7,"\u91cc\u63d0\u4f9b\u4e86\u6570\u636e\u4e0b\u8f7d\u811a\u672c":10,"\u91cc\u9762\u8bfb\u53d6":20,"\u91cf\u4e5f\u53ef\u4ee5\u901a\u8fc7\u8c03\u7528cmake\u547d\u4ee4\u524d\u901a\u8fc7\u73af\u5883\u53d8\u91cf\u6307\u5b9a":1,"\u9488\u5bf9\u672c\u95ee\u9898":10,"\u94fe\u63a5\u4f55\u79cdblas\u7b49\u7b49":1,"\u9519\u8bef\u7387":10,"\u95f4\u9694":20,"\u9664\u4e86":20,"\u9664\u8fc7data\u5c42":10,"\u9700\u8981\u53c2\u8003":6,"\u9700\u8981\u6ce8\u610f":20,"\u9700\u8981\u6ce8\u610f\u7684\u662f":[1,7],"\u9884\u6d4b\u6570\u636e\u6307\u5b9atest":10,"\u9884\u6d4b\u7ed3\u679c\u4ee5\u6587\u672c\u7684\u5f62\u5f0f\u4fdd\u5b58\u5728":10,"\u9884\u6d4b\u811a\u672c":10,"\u9884\u6d4bid":10,"\u989d\u5916\u7684\u53c2\u6570":10,"\u9996\u5148\u5217\u4e3e\u903b\u8f91\u56de\u5f52\u7f51\u7edc":10,"\u9996\u5148\u6211\u4eec\u5c06\u8fd9\u4e2a\u6570\u636e\u6587\u4ef6":20,"\u9996\u5148\u8bf7\u53c2\u8003":10,"\u9aa43":10,"\u9ed8\u8ba4\u4e00\u4e2apass\u4fdd\u5b58\u4e00\u6b21\u6a21\u578b":10,"\u9ed8\u8ba4\u503c":1,"\u9ed8\u8ba4\u5355\u4f4d\u662fbyte":0,"\u9ed8\u8ba4\u60c5\u51b5\u4e0b\u4e00\u6761\u6570\u636e":20,"adamax\u7b49":10,"amazon\u7535\u5b50\u4ea7\u54c1\u8bc4\u8bba\u6570\u636e":10,"api\u9884\u6d4b":10,"argument\u4f20\u5165":20,"argument\u5f62\u5f0f\u4f20\u5165":20,"atlas\u5e93\u7684\u8def\u5f84":1,"batches\u8bbe\u7f6e\u6bcf\u9694\u591a\u5c11batch\u4fdd\u5b58\u4e00\u6b21\u6a21\u578b":10,"bool\u53c2\u6570":20,"case":[10,22],"cd\u5230\u542b\u6709dockerfile\u7684\u8def\u5f84\u4e2d":6,"check\u662ffalse\u7684\u8bdd":20,"cmake\u53ef\u4ee5\u5c06cmake\u9879\u76ee\u6587\u4ef6":1,"cmake\u662f\u4e00\u4e2a\u8de8\u5e73\u53f0\u7684\u7f16\u8bd1\u811a\u672c":1,"cmake\u7684\u5b98\u65b9\u6587\u6863":1,"cmake\u7f16\u8bd1\u65f6\u4f1a\u9996\u5148\u5728\u7cfb\u7edf\u8def\u5f84":1,"container\u540e":6,"cuda\u76f8\u5173\u7684driver\u548c\u8bbe\u5907\u6620\u5c04\u8fdbcontainer\u4e2d":6,"d\u547d\u4ee4\u5373\u53ef":1,"d\u547d\u4ee4\u6307\u5b9a":1,"dataprovider\u521b\u5efa\u7684\u65f6\u5019\u6267\u884c":20,"dataprovider\u53ef\u4ee5\u662f":20,"dataprovider\u63d0\u4f9b\u4e86\u4e24\u79cd\u7b80\u5355\u7684cache\u7b56\u7565":20,"dataprovider\u7684\u5177\u4f53\u7528\u6cd5\u548c\u5982\u4f55\u5b9e\u73b0\u4e00\u4e2a\u65b0\u7684dataprovid":19,"dim\u7684\u65b0\u7684\u5411\u91cf":10,"dim\u7ef4\u5ea6\u5411\u91cf":10,"docker\u662f\u4e00\u4e2a\u57fa\u4e8e\u5bb9\u5668\u7684\u8f7b\u91cf\u7ea7\u865a\u62df\u73af\u5883":6,"docker\u7684\u5b98\u65b9\u6587\u6863":6,"dockerfile\u548cbuild":0,"dockerfile\u662fdock":0,"dockerfile\u7684\u6587\u6863":6,"dockerfile\u7684\u6700\u4f73\u5b9e\u8df5":6,"driver\u6dfb\u52a0\u5230ld_library_path\u4e2d":7,"elec\u6d4b\u8bd5\u96c6":10,"embedding\u6a21\u578b\u9700\u8981\u7a0d\u5fae\u6539\u53d8\u6570\u636e\u63d0\u4f9b\u7684\u811a\u672c":10,"export":[1,6,7],"f\u4ee3\u8868\u4e00\u4e2a\u6d6e\u70b9\u6570":20,"float":20,"generator\u4fbf\u4f1a\u5b58\u4e0b\u5f53\u524d\u7684\u4e0a\u4e0b\u6587":20,"generator\u7684\u4e0a\u4e0b\u6587\u4e2d\u5c3d\u91cf\u7559":20,"generator\u81f3\u5c11\u8c03\u7528\u4e24\u6b21\u624d\u4f1a\u77e5\u9053\u662f\u5426\u505c\u6b62":20,"gpu\u53cc\u7f13\u5b58":20,"gpu\u7248\u672c\u4e8c\u8fdb\u5236":1,"gru\u6a21\u578b":10,"gru\u6a21\u578b\u914d\u7f6e":10,"i\u4ee3\u8868\u4e00\u4e2a\u6574\u6570":20,"id\u4e3a0\u7684\u6982\u7387":10,"id\u4e3a1\u7684\u6982\u7387":10,"image\u6784\u5efa\u6e90\u7801\u653e\u7f6e\u5728":0,"image\u7684\u4e3b\u8981\u63cf\u8ff0\u6587\u4ef6":0,"image\u7684\u4e3b\u8981\u6784\u5efa\u6b65\u9aa4":0,"image\u7684\u6784\u5efa\u6b65\u9aa4":0,"import":[10,20,23],"include\u4e0b\u9700\u8981\u5305\u542bcbla":1,"include\u4e0b\u9700\u8981\u5305\u542bmkl":1,"init_hook\u53ef\u4ee5\u4f20\u5165\u4e00\u4e2a\u51fd\u6570":20,"int":[10,20],"key\u662fdata_layer\u7684\u540d\u5b57":20,"layer\u4f5c\u4e3a\u8f93\u51fa":23,"layer\u6587\u6863":10,"ld_library_path\u7b49\u7b49":7,"ld_library_path\u91cc\u9762\u627e\u4e0d\u5230\u8fd9\u4e9b\u52a8\u6001":7,"lib\u4e0b\u9700\u8981\u5305\u542bcblas\u548catlas\u4e24\u4e2a\u5e93":1,"lib\u4e0b\u9700\u8981\u5305\u542bcblas\u5e93":1,"lib\u4e0b\u9700\u8981\u5305\u542bopenblas\u5e93":1,"lib\u76ee\u5f55\u4e0b\u9700\u8981\u5305\u542b":1,"list\u4e0d\u8bbe\u7f6e":19,"list\u4e2d":[19,20],"list\u4e2d\u7684\u4e00\u884c":20,"list\u4e2d\u7684\u6bcf\u4e00\u884c":20,"list\u4e3a\u7eaf\u6587\u672c\u6587\u4ef6":19,"list\u4e5f\u53ef\u4ee5\u653e\u7f6ehdfs\u6587\u4ef6\u8def\u5f84":19,"list\u5199\u5165\u90a3\u4e2a\u6587\u672c\u6587\u4ef6\u7684\u5730\u5740":20,"list\u5373\u4e3a":20,"list\u548ctest":19,"list\u5747\u4e3a\u672c\u5730\u7684\u4e24\u4e2a\u6587\u4ef6":19,"list\u6307\u5b9a\u7684\u6570\u636e":10,"list\u7684\u4f4d\u7f6e":10,"list\u82e5\u5e72\u6570\u636e\u6587\u4ef6\u8def\u5f84\u7684\u67d0\u4e00\u4e2a\u8def\u5f84":20,"lstm\u6a21\u578b\u7b49":10,"lstm\u6a21\u578b\u914d\u7f6e":10,"make\u548cmak":2,"mkl\u7684\u8def\u5f84":1,"mkl_sequential\u548cmkl_intel_lp64\u4e09\u4e2a\u5e93":1,"mnist\u662f\u4e00\u4e2a\u5305\u542b\u6709":20,"movielens\u6570\u636e\u96c6":9,"movielens\u8bc4\u5206\u56de\u5f52":9,"osx\u6216\u8005\u662fwindows\u673a\u5668":6,"osx\u7684\u5b89\u88c5\u6587\u6863":6,"paddle\u5b9a\u4e49\u7684\u53c2\u6570\u5305\u62ec":20,"paddlepaddle\u4f7f\u7528\u8fd0\u884c\u65f6\u52a8\u6001\u8fde\u63a5cuda\u7684so":7,"paddlepaddle\u4fdd\u7559\u6dfb\u52a0\u53c2\u6570\u7684\u6743\u529b":20,"paddlepaddle\u53ef\u4ee5\u4f7f\u7528":1,"paddlepaddle\u53ef\u4ee5\u8bfb\u53d6python\u5199\u7684\u4f20\u8f93\u6570\u636e\u811a\u672c":10,"paddlepaddle\u5728\u8fd0\u884c\u65f6\u627e\u4e0d\u5230\u5bf9\u5e94\u7684config\u6587\u4ef6":7,"paddlepaddle\u5c06train":20,"paddlepaddle\u63a8\u8350\u4f7f\u7528docker\u8fdb\u884cpaddlepaddle\u7684\u90e8\u7f72\u548c":6,"paddlepaddle\u63d0\u4f9b\u4e86docker\u7684\u4f7f\u7528\u955c\u50cf":6,"paddlepaddle\u63d0\u4f9b\u6570\u4e2a\u9884\u7f16\u8bd1\u7684\u4e8c\u8fdb\u5236\u6765\u8fdb\u884c\u5b89\u88c5":5,"paddlepaddle\u63d0\u4f9b\u7684\u955c\u50cf\u5e76\u4e0d\u5305\u542b\u4efb\u4f55\u547d\u4ee4\u8fd0\u884c":6,"paddlepaddle\u7684\u6570\u636e\u5305\u62ec\u56db\u79cd\u4e3b\u8981\u7c7b\u578b":20,"paddlepaddle\u7684\u7f16\u8bd1\u9009\u9879\u53ef\u4ee5\u5728\u8c03\u7528cmake\u7684\u65f6\u5019\u8bbe\u7f6e":1,"paddlepaddle\u7684\u7f16\u8bd1\u9009\u9879\u662f\u53ef\u4ee5\u63a7\u5236paddlepaddle\u751f\u6210cpu":1,"paddlepaddle\u7684cpu\u4e8c\u8fdb\u5236":6,"paddlepaddle\u7684dock":0,"paddlepaddle\u7684gpu\u4e8c\u8fdb\u5236":6,"paddlepaddle\u7684python\u9884\u6d4b\u63a5\u53e3":22,"paddlepaddle\u7684ubuntu\u5b89\u88c5\u5305\u5206\u4e3a\u4e24\u4e2a\u7248\u672c":7,"paddlepaddle\u76ee\u524d\u4f7f\u7528swig\u5bf9\u5176\u5e38\u7528\u7684\u9884\u6d4b\u63a5\u53e3\u8fdb\u884c\u4e86\u5c01\u88c5":23,"paddlepaddle\u76ee\u524d\u652f\u6301ubuntu":7,"paddlepaddle\u8fd0\u884c\u65f6\u5982\u679c\u6ca1\u6709\u5bfb\u627e\u5230cuda\u7684driv":7,"paddlepaddle\u9700\u8981\u7528\u6237\u5728\u7f51\u7edc\u914d\u7f6e":19,"period\u8bbe\u7f6e\u6253\u5370\u53c2\u6570\u4fe1\u606f\u7b49":10,"process\u51fd\u6570":20,"process\u51fd\u6570\u662f\u5b9e\u73b0\u6570\u636e\u8f93\u5165\u7684\u4e3b\u51fd\u6570":20,"process\u51fd\u6570\u8c03\u7528\u591a\u6b21":20,"pserver\u4e3apaddlepaddle\u7684paramet":13,"pserver\u7684\u547d\u4ee4\u884c\u53c2\u6570":13,"pserver\u7ec4\u5408\u4f7f\u7528":13,"py\u6587\u4ef6\u7ed9\u51fa\u4e86\u5b8c\u6574\u4f8b\u5b50":10,"pydataprovider2\u4f1a\u5c3d\u91cf\u4f7f\u7528\u5185\u5b58":20,"pydataprovider2\u6587\u6863":23,"pydataprovider2\u7684\u4f7f\u7528":19,"pydataprovider\u662fpaddlepaddle\u4f7f\u7528python\u63d0\u4f9b\u6570\u636e\u7684\u63a8\u8350\u63a5\u53e3":20,"python\u5305":6,"python\u53ef\u4ee5\u89e3\u9664\u6389\u5185\u90e8\u53d8\u91cf\u7684\u5f15\u7528":20,"python\u7684":6,"python\u7684swig\u63a5\u53e3\u53ef\u4ee5\u65b9\u4fbf\u8fdb\u884c\u9884\u6d4b\u548c\u5b9a\u5236\u5316\u8bad\u7ec3":1,"release\u7684\u72b6\u6001":5,"return":[10,20],"rnn\u914d\u7f6e":11,"server\u8fdb\u7a0b":13,"sh\u662fdocker":0,"shuffle\u8bad\u7ec3\u6570\u636e":20,"softmax\u8f93\u51fa":10,"string\u7684\u683c\u5f0f\u6253\u5370\u51fa\u6765":13,"swig_paddle\u63a5\u53d7\u7684\u539f\u59cb\u6570\u636e\u662fc":23,"train\u5373\u4e3apaddlepaddle\u7684\u8bad\u7ec3\u8fdb\u7a0b":13,"train\u5b8c\u6210\u5355\u673a\u591a\u663e\u5361\u591a\u7ebf\u7a0b\u7684\u8bad":13,"train\u7684\u547d\u4ee4\u884c\u53c2\u6570":13,"ubuntu\u7684deb\u5b89\u88c5\u5305\u7b49":5,"v2\u4e4b\u540e\u7684\u4efb\u4f55\u4e00\u4e2acudnn\u7248\u672c\u6765\u7f16\u8bd1\u8fd0\u884c":1,"value\u5373\u4e3asoftmax\u5c42\u7684\u8f93\u51fa":23,"value\u662f\u7279\u5f81\u503c":20,"var":6,"vector\u8868\u793a\u7684\u6bcf\u4e2a\u5355\u8bcd":10,"version\u53ef\u4ee5\u6253\u5370\u51fapaddle\u7684\u7248\u672c\u4fe1\u606f\u548c\u7f16\u8bd1\u7684\u9009\u9879":18,"version\u53ef\u4ee5\u6253\u5370\u51fapaddlepaddle\u7684\u7248\u672c\u548c\u7f16\u8bd1\u65f6\u4fe1\u606f":13,"version\u7684\u547d\u4ee4\u884c\u53c2\u6570":13,"yield\u6587\u672c\u4fe1\u606f\u548c\u7c7b\u522bid":10,__main__:23,__name__:23,abov:20,act:10,act_typ:10,activ:10,adadelta:10,adagrad:10,adam:10,adamoptim:10,afi:20,all:20,allow:10,alreadi:7,also:10,append:20,apt:[6,7],arg:[0,10,20],around:20,arrai:23,assert:23,atla:1,atlas_root:1,avg:10,avgcost:10,avgpool:10,avx:7,bag:10,baidu:[6,7],batch:10,batch_siz:10,binari:10,bla:1,bool:10,both:10,bow:10,build:[0,6],cach:[10,19],cache_pass_in_mem:[10,20],cachetyp:[10,20],calc_batch_s:20,call:10,can:10,can_over_batch_s:20,categori:10,cento:7,check:[7,20],check_fail_continu:20,chines:9,chpasswd:6,classif:10,classification_cost:10,classification_error_evalu:10,close:20,cmake:1,cmd:6,cnn:10,code:[0,20,23],com:[6,7],comment:10,compil:18,conf:23,config:[7,10],config_arg:10,config_pars:23,connect:10,contain:[10,20],context:20,context_len:10,context_start:10,convert:[10,20,23],couldn:7,cpp:[7,10],cpu:[6,7,20],createfromconfigproto:23,cross:10,cuda_so:6,cudastat:7,cudasuccess:7,cudnn:1,cudnn_root:1,cudnnv5:1,current:[10,20],currentcost:10,currentev:10,dalla:20,data:7,data_config:23,data_initialz:10,data_lay:[10,20],dataprovid:10,dataprovider_bow:10,dataprovider_emb:10,dataproviderconvert:23,dataset:10,deb:7,debian:7,decor:20,def:[10,20,23],defin:[10,20],define_py_data_sources2:[10,20],delar:10,demo:[6,10],dense_vector:[20,23],describ:10,descript:22,detail:22,dev:6,devel:6,devic:6,devices:6,dict:[10,20],dict_fil:10,dictionai:10,dictionari:[10,20],dictrionari:10,differ:10,dim:10,dimens:10,dir:10,doc:23,documentari:20,dpkg:7,driver:7,dso_handl:7,dtype:23,dump_config:13,dure:[10,20],dynam:20,each:[10,20],each_pixel_str:20,each_word:20,echo:6,either:10,els:10,emb:10,embed:9,embedding_lay:10,entropi:10,enumer:10,error:[7,10],etc:6,eval:10,exampl:10,experi:5,expose:6,f0831:7,fail:7,fals:10,fc_layer:10,featur:[10,20],festiv:20,file:[10,20],file_list:20,file_nam:10,filenam:20,fill:10,find:7,first:10,float32:23,fly:10,forwardtest:23,framework:10,from:[6,10,20,23],fulli:10,func:20,gener:[10,20],gentoo:7,get:[6,7,10,20],get_config_arg:10,get_data:10,github:7,give:20,given:10,globe:20,gpu:[6,7],gradient_clipping_threshold:10,gradientmachin:23,gru:10,gru_siz:10,help:23,hint:23,hl_cuda_devic:7,hl_dso_load:7,host:6,hot:10,hous:20,http:7,ignor:20,imag:6,imagenet:9,img:20,inarg:23,includ:10,init:10,init_hook:[10,19],init_model_path:10,initi:[10,20],initpaddl:23,input:[10,20],input_typ:[10,19],instal:2,insuffici:7,intal:5,integ:[10,20],integer_sequ:20,integer_valu:[10,20],integer_value_sequ:10,invok:20,is_predict:10,is_train:20,isinst:23,iterat:20,job:10,kernel:6,kwarg:[10,20],l2regular:10,label:[10,20],label_dim:10,lake:20,later:10,latest:[0,6],layer:10,ld_library_path:7,learning_method:10,learning_r:10,len:[10,20],lib64:[6,7],lib:1,libcuda:6,libnvidia:6,librari:7,list:[10,19,20],load_data_arg:23,loadparamet:23,local:[1,7],log_period:10,logger:20,look:[10,20],loss:10,lowest_dl_speed:0,lstm:10,lstm_size:10,mac:6,mai:5,main:23,maintainer:6,make:[7,20],make_diagram:13,maxid:10,maxid_lay:10,mean:10,memori:10,merge_model:13,method:20,min_pool_s:20,mkdir:6,mkl:1,mkl_core:1,mkl_root:1,mnist:20,mnist_model:23,mnist_provid:20,mnist_train:20,model_config:23,modul:[10,20],momentum:10,movi:20,must:7,name:[6,10,20],necessari:10,need:10,neg:[10,20],net:6,next:20,no_cache:20,no_sequence:20,none:[10,20,23],note:[5,7],nullptr:7,num:10,num_pass:10,nvidia:6,obj:[10,20],object:[10,20],off:[0,1,18],omit:10,on_init:20,onli:10,open:[10,20],openbla:1,openblas_root:1,openssh:6,opt:1,other:10,outlin:22,output:10,outsid:20,packag:5,paddl:[0,6,7,10,13],paddle_gpu:0,paddle_ssh:6,paddle_ssh_machin:6,paddledev:6,paddlepaddl:[6,7,18,23],paramet:10,parse_config:23,pass:[10,20],path:[7,10],period:10,permitrootlogin:6,pixel:20,pixels_float:20,pixels_str:20,place:20,pleas:7,pool_siz:20,pooling_lay:10,pooling_typ:10,posit:[10,20],pre:5,pred:10,predict_output_dir:10,predict_sampl:23,preprocess:10,print:23,process:[10,20],process_pr:10,properli:10,pull:6,put:10,py_paddl:[6,23],pydataprovid:19,pydataprovider2:[10,20,23],pydataproviderwrapp:10,python:10,quick_start:10,rang:10,rank:10,rare:20,read:[10,20],real_process:20,refer:19,reference_cblas_root:1,reffer:1,regular:10,releas:[5,7],repres:10,represent:10,resnet:9,result:[10,20],rmsprop:10,roce:6,root:6,run:6,runtim:[7,20],same:[10,20],sampl:[10,20],save:[10,20],save_dir:10,saw:20,sbin:6,script:0,second:10,sed:6,see:10,sentenc:20,sentiment:20,sentimental_provid:20,separ:10,seq_typ:20,sequel:20,sequence:20,sequence_conv_pool:10,sequencetyp:20,server:6,set:[10,20],setup:10,should_shuffl:20,simple_gru:10,simple_lstm:10,size:[10,20],smooth:5,softmax:10,softmaxactiv:10,sourc:10,spars:10,sparse_binary_vector:[10,20],sparse_float_vector:20,specifi:[7,10],split:[10,20],src_root:23,ssh:6,sshd:6,sshd_config:6,stat:10,state:5,still:5,stop:6,store:10,string:20,strip:10,structur:10,stun:20,sub_sequence:20,sure:7,swig_paddl:23,tag:[0,7],take:20,tbd:21,team:6,test:[10,19],test_data:23,test_list:[10,20],text:[10,20],text_conv:10,them:10,thi:[10,20],thing:20,tmp:20,train:7,train_list:[10,20],trainer:[10,20,23],trainer_config:[10,19,20,23],trainer_config_help:[10,20],trainerintern:10,trainermain:7,travel:20,trn:10,tst:10,two:10,txt:[10,20],type:[10,20],ubuntu:7,unk_idx:10,updat:6,use:[10,22],use_dynamic_ord:20,use_gpu:[10,23],usepam:6,user:10,usr:[1,6,7],valid:7,valu:[10,20,23],version:[6,7],via:7,want:20,what:10,when:20,which:10,whole:20,wilder:20,window:6,with_avx:[1,18],with_doc:1,with_doc_cn:1,with_doubl:18,with_double:1,with_dso:1,with_gflag:18,with_gflags:1,with_glog:[1,18],with_gpu:[0,1,18],with_metric_learn:18,with_predict_sdk:18,with_python:[1,18],with_rdma:[1,18],with_style_check:1,with_swig_py:1,with_testing:1,with_tim:18,with_timer:1,wonder:20,word:9,word_dict:10,word_dim:10,word_id:20,word_vector:10,xarg:6,yield:[10,20],you:[7,20],your:5,your_host_machine:6},titles:["\u6784\u5efaPaddlePaddle Docker Image","\u8bbe\u7f6ePaddlePaddle\u7684\u7f16\u8bd1\u9009\u9879","\u4f7f\u7528cmake\u7f16\u8bd1PaddlePaddle","\u5b89\u88c5\u7f16\u8bd1PaddlePaddle\u9700\u8981\u7684\u4f9d\u8d56","make\u548cmake install","\u7f16\u8bd1\u4e0e\u5b89\u88c5","\u5b89\u88c5PaddlePaddle\u7684Docker\u955c\u50cf","\u4f7f\u7528deb\u5305\u5728Ubuntu\u4e0a\u5b89\u88c5PaddlePaddle","\u96c6\u7fa4\u8bad\u7ec3","\u4f7f\u7528\u793a\u4f8b","PaddlePaddle\u5feb\u901f\u5165\u95e8\u6559\u7a0b","PaddlePaddle\u6587\u6863","&lt;no title&gt;","\u547d\u4ee4\u884c\u53c2\u6570","&lt;no title&gt;","&lt;no title&gt;","paddle pserver\u7684\u547d\u4ee4\u884c\u53c2\u6570","paddle train\u7684\u547d\u4ee4\u884c\u53c2\u6570","paddle version\u7684\u547d\u4ee4\u884c\u53c2\u6570","PaddlePaddle\u7684\u6570\u636e\u63d0\u4f9b(DataProvider)\u4ecb\u7ecd","PyDataProvider2\u7684\u4f7f\u7528","\u81ea\u5b9a\u4e49\u4e00\u4e2aDataProvider","\u7528\u6237\u63a5\u53e3","PaddlePaddle\u7684Python\u9884\u6d4b\u63a5\u53e3"],titleterms:{"\u4e0b\u8f7d\u548c\u8fd0\u884cdocker\u955c\u50cf":6,"\u4ecb\u7ecd":19,"\u4f18\u5316\u7b97\u6cd5":10,"\u4f7f\u7528\u6307\u5357":11,"\u4f7f\u7528\u6982\u8ff0":10,"\u4f7f\u7528\u793a\u4f8b":9,"\u4f7f\u7528\u811a\u672c\u6784\u5efapaddlepaddl":0,"\u4f7f\u7528cmake\u7f16\u8bd1paddlepaddl":2,"\u4f7f\u7528deb\u5305\u5728ubuntu\u4e0a\u5b89\u88c5paddlepaddl":7,"\u5185\u5b58\u4e0d\u591f\u7528\u7684\u60c5\u51b5":20,"\u5377\u79ef\u6a21\u578b":10,"\u53c2\u8003":20,"\u53ef\u80fd\u7684\u5185\u5b58\u6cc4\u9732\u95ee\u9898":20,"\u53ef\u80fd\u9047\u5230\u7684\u95ee\u9898":7,"\u547d\u4ee4\u884c\u53c2\u6570":[10,13,22],"\u56fe\u50cf":9,"\u5b89\u88c5":10,"\u5b89\u88c5\u7f16\u8bd1paddlepaddle\u9700\u8981\u7684\u4f9d\u8d56":3,"\u5b89\u88c5paddlepaddle\u7684docker\u955c\u50cf":6,"\u5e38\u7528\u6a21\u578b":9,"\u5e8f\u5217\u6a21\u578b\u6570\u636e\u63d0\u4f9b":20,"\u5f00\u53d1\u6307\u5357":11,"\u6027\u80fd\u95ee\u9898":6,"\u603b\u4f53\u6548\u679c\u603b\u7ed3":10,"\u63a8\u8350":9,"\u6570\u636e\u5411\u6a21\u578b\u4f20\u9001":10,"\u6570\u636e\u63d0\u4f9b":22,"\u6570\u636e\u683c\u5f0f\u51c6\u5907":10,"\u65f6\u5e8f\u6a21\u578b":10,"\u6784\u5efapaddlepaddl":0,"\u6ce8\u610f\u4e8b\u9879":[6,20],"\u7528\u6237\u63a5\u53e3":22,"\u7b80\u5355\u7684\u4f7f\u7528\u573a\u666f":20,"\u7b97\u6cd5\u6559\u7a0b":11,"\u7f16\u8bd1\u4e0e\u5b89\u88c5":5,"\u7f51\u7edc\u7ed3\u6784":10,"\u81ea\u5b9a\u4e49\u4e00\u4e2adataprovid":21,"\u81ea\u7136\u8bed\u8a00\u5904\u7406":9,"\u8bad\u7ec3\u6a21\u578b":10,"\u8bbe\u7f6epaddlepaddle\u7684\u7f16\u8bd1\u9009\u9879":1,"\u8bcd\u5411\u91cf\u6a21\u578b":10,"\u8f93\u51fa\u65e5\u5fd7":10,"\u8fdc\u7a0b\u8bbf\u95ee\u95ee\u9898\u548c\u4e8c\u6b21\u5f00\u53d1":6,"\u903b\u8f91\u56de\u5f52\u6a21\u578b":10,"\u914d\u7f6e\u4e2d\u7684\u6570\u636e\u52a0\u8f7d\u5b9a\u4e49":10,"\u9644\u5f55":10,"\u96c6\u7fa4\u8bad\u7ec3":8,"\u9884\u6d4b":[10,22],"blas\u76f8\u5173\u7684\u7f16\u8bd1\u9009\u9879":1,"bool\u578b\u7684\u7f16\u8bd1\u9009\u9879":1,"config\u6587\u4ef6\u627e\u4e0d\u5230":7,"cudnn\u76f8\u5173\u7684\u7f16\u8bd1\u9009\u9879":1,"driver\u627e\u4e0d\u5230":7,"make\u548cmak":4,"paddlepaddle\u5feb\u901f\u5165\u95e8\u6559\u7a0b":10,"paddlepaddle\u63d0\u4f9b\u7684docker\u955c\u50cf\u7248\u672c":6,"paddlepaddle\u6587\u6863":11,"paddlepaddle\u7684\u6570\u636e\u63d0\u4f9b":19,"paddlepaddle\u7684\u7f16\u8bd1\u9009\u9879":1,"paddlepaddle\u7684bool\u578b\u7f16\u8bd1\u9009\u9879":1,"paddlepaddle\u7684cblas\u7f16\u8bd1\u9009\u9879":1,"paddlepaddle\u7684python\u9884\u6d4b\u63a5\u53e3":23,"pserver\u7684\u547d\u4ee4\u884c\u53c2\u6570":16,"pydataprovider2\u7684\u4f7f\u7528":20,"python\u6570\u636e\u52a0\u8f7d\u811a\u672c":10,"so\u627e\u4e0d\u5230":7,"train\u7684\u547d\u4ee4\u884c\u53c2\u6570":17,"version\u7684\u547d\u4ee4\u884c\u53c2\u6570":18,algorithm:10,appendix:10,architectur:10,argument:10,cach:20,command:10,configur:10,convolut:10,cuda:[1,7],data:10,dataprovid:19,docker:0,image:0,init_hook:20,input_typ:20,instal:4,install:10,libcudart:7,libcudnn:7,line:10,log:10,logist:10,model:10,network:10,optimiz:10,overview:10,paddl:[16,17,18],predict:10,prepar:10,provid:[10,20],refer:20,regress:10,script:10,sequenc:10,summari:10,time:10,train:10,transfer:10,vector:10,word:10}})
\ No newline at end of file
--- a/doc_cn/ui/data_provider/index.html
+++ b/doc_cn/ui/data_provider/index.html
@@ -61,6 +61,11 @@
 <li class="toctree-l3"><a class="reference internal" href="pydataprovider2.html#cache">cache</a></li>
 </ul>
 </li>
+<li class="toctree-l2"><a class="reference internal" href="pydataprovider2.html#id4">注意事项</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="pydataprovider2.html#id5">可能的内存泄露问题</a></li>
+<li class="toctree-l3"><a class="reference internal" href="pydataprovider2.html#id6">内存不够用的情况</a></li>
+</ul>
+</li>
 </ul>
 </li>
 <li class="toctree-l1"><a class="reference internal" href="write_new_dataprovider.html">自定义一个DataProvider</a></li>

--- a/doc_cn/ui/data_provider/pydataprovider2.html
+++ b/doc_cn/ui/data_provider/pydataprovider2.html
@@ -140,10 +140,66 @@ process函数调用多次 <code class="code docutils literal"><span class="pre">
                        <span class="n">test_list</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
                        <span class="n">module</span><span class="o">=</span><span class="s1">&#39;mnist_provider&#39;</span><span class="p">,</span>
                        <span class="n">obj</span><span class="o">=</span><span class="s1">&#39;process&#39;</span><span class="p">)</span>
+<span class="n">img</span> <span class="o">=</span> <span class="n">data_layer</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;pixel&#39;</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="mi">784</span><span class="p">)</span>
+<span class="n">label</span> <span class="o">=</span> <span class="n">data_layer</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;label&#39;</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="mi">10</span><span class="p">)</span>
 </pre></div>
 </div>
 <p>这里说明了训练数据是 &#8216;train.list&#8217;，而没有测试数据。引用的DataProvider是 &#8216;mnist_provider&#8217;
 这个模块中的 &#8216;process&#8217; 函数。</p>
+<p>同时，根据模型配置文件中 <code class="code docutils literal"><span class="pre">data_layer</span></code> 的名字，用户也可以显式指定返回的数据对应关系。例如:</p>
+<div class="highlight-default"><table class="highlighttable"><tr><td class="linenos"><div class="linenodiv"><pre> 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</pre></div></td><td class="code"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">paddle.trainer.PyDataProvider2</span> <span class="k">import</span> <span class="o">*</span>
+<span class="c1"># Define a py data provider</span>
+<span class="nd">@provider</span><span class="p">(</span><span class="n">input_types</span><span class="o">=</span><span class="p">[</span>
+    <span class="n">dense_vector</span><span class="p">(</span><span class="mi">28</span> <span class="o">*</span> <span class="mi">28</span><span class="p">),</span>
+    <span class="n">integer_value</span><span class="p">(</span><span class="mi">10</span><span class="p">)</span>
+<span class="p">])</span>
+<span class="k">def</span> <span class="nf">process</span><span class="p">(</span><span class="n">settings</span><span class="p">,</span> <span class="n">filename</span><span class="p">):</span>  <span class="c1"># settings is not used currently.</span>
+    <span class="n">f</span> <span class="o">=</span> <span class="nb">open</span><span class="p">(</span><span class="n">filename</span><span class="p">,</span> <span class="s1">&#39;r&#39;</span><span class="p">)</span>  <span class="c1"># open one of training file</span>
+    <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">f</span><span class="p">:</span>  <span class="c1"># read each line</span>
+        <span class="n">label</span><span class="p">,</span> <span class="n">pixel</span> <span class="o">=</span> <span class="n">line</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s1">&#39;;&#39;</span><span class="p">)</span>
+        <span class="c1"># get features and label</span>
+        <span class="n">pixels_str</span> <span class="o">=</span> <span class="n">pixel</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s1">&#39; &#39;</span><span class="p">)</span>
+        <span class="n">pixels_float</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">each_pixel_str</span> <span class="ow">in</span> <span class="n">pixels_str</span><span class="p">:</span>
+            <span class="n">pixels_float</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="nb">float</span><span class="p">(</span><span class="n">each_pixel_str</span><span class="p">))</span>
+        <span class="c1"># give data to paddle.</span>
+        <span class="k">yield</span> <span class="p">{</span> <span class="s2">&quot;pixel&quot;</span><span class="p">:</span> <span class="n">pixels_float</span><span class="p">,</span> <span class="s1">&#39;label&#39;</span><span class="p">:</span> <span class="nb">int</span><span class="p">(</span><span class="n">label</span><span class="p">)</span> <span class="p">}</span>
+    <span class="n">f</span><span class="o">.</span><span class="n">close</span><span class="p">()</span>  <span class="c1"># close file</span>
+</pre></div>
+</td></tr></table></div>
+<p>如果用户不指定返回数据的对应关系，那么PaddlePaddle会粗略的根据layer的声明顺序，
+来确定对应关系。这个对应关系可能不正确。所以推荐使用显式指定返回值和数据对应关系。</p>
 <p>至此，简单的PyDataProvider样例就说明完毕了。对于用户来说，讲数据发送给PaddlePaddle，仅仅需要
 知道如何从 <strong>一个文件</strong> 里面读取 <strong>一条</strong> 样本。而PaddlePaddle进程帮助用户做了</p>
 <ul class="simple">
@@ -256,11 +312,13 @@ DataProvider创建的时候执行。这个初始化函数具有如下参数:</p>
 <h2>参考(Reference)<a class="headerlink" href="#reference" title="Permalink to this headline">¶</a></h2>
 <div class="section" id="provider">
 <h3>&#64;provider<a class="headerlink" href="#provider" title="Permalink to this headline">¶</a></h3>
-<p><a class="reference external" href="mailto:'&#37;&#52;&#48;provider">'<span>&#64;</span>provider</a>&#8216;是一个Python的 <a class="reference external" href="http://www.learnpython.org/en/Decorators">Decorator</a> ，他可以将某一个函数标记成一个PyDataProvider。它包含的参数有:</p>
+<p><code class="code docutils literal"><span class="pre">&#64;provider</span></code> 是一个Python的 <a class="reference external" href="http://www.learnpython.org/en/Decorators">Decorator</a> ，他可以将某一个函数标记成一个PyDataProvider。它包含的参数有:</p>
 <ul class="simple">
 <li><a class="reference internal" href="#input-types">input_types</a> 是数据输入格式。具体有哪些格式，参考 <a class="reference internal" href="#input-types">input_types</a> 。</li>
 <li>should_shuffle 是个DataProvider是不是要做shuffle，如果不设置的话，训练的时候默认shuffle，
-测试的时候默认不shuffle</li>
+测试的时候默认不shuffle。</li>
+<li>min_pool_size 是设置DataProvider在内存中最小暂存的数据条数。这个也是PaddlePaddle所能够保证的shuffle粒度。
+设置成-1的话，会预先读取全部数据到内存中。</li>
 <li>pool_size 是设置DataProvider在内存中暂存的数据条数。设置成-1的话，即不在乎内存暂存多少条数据。</li>
 <li>can_over_batch_size 表示是否允许Paddle暂存略微多余pool_size的数据。这样做可以避免很多死锁问题。
 一般推荐设置成True</li>
@@ -268,6 +326,11 @@ DataProvider创建的时候执行。这个初始化函数具有如下参数:</p>
 是一个batch size，但是有时为了计算均衡性，可以将一条数据设置成多个batch size</li>
 <li>cache 是数据缓存的策略，参考 <a class="reference internal" href="#cache">cache</a></li>
 <li>init_hook 是初始化时调用的函数，参考 <a class="reference internal" href="#init-hook">init_hook</a></li>
+<li>use_dynamic_order 如果是true的话，可以返回一个dict，key是data_layer的名字，value是特征值。同时，也可以
+返回一个list或者tuple。如果是false的话，只能够返回list或者tuple</li>
+<li>check 设置成true的话，会根据input_types检查数据的合法性。</li>
+<li>check_fail_continue 如果设置成true的话，即使在check中数据不合法，也会扔到这条数据，继续训练。 如果
+check是false的话，没有作用。</li>
 </ul>
 </div>
 <div class="section" id="input-types">
@@ -371,6 +434,47 @@ DataProvider</li>
 </ul>
 </div>
 </div>
+<div class="section" id="id4">
+<h2>注意事项<a class="headerlink" href="#id4" title="Permalink to this headline">¶</a></h2>
+<div class="section" id="id5">
+<h3>可能的内存泄露问题<a class="headerlink" href="#id5" title="Permalink to this headline">¶</a></h3>
+<p>PaddlePaddle将train.list中的每一行，都传递给process函数，从而生成多个generator。
+即如果train.list中，有100个训练文件，即会生成100个generator。这个本身不是一个很
+严重的问题。</p>
+<p>但是，如果在训练时，每一条训练数据都是一个文件，并且，训练数据非常多的情况下，就
+会生成多个generator。每个generator在没有调用的时候，是几乎不占内存的。但是，当调
+用过一次的时候，generator便会存下当前的上下文(Context)。而这个Context可能会非常
+大。并且，generator至少调用两次才会知道是否停止。所以，即使在process里面只会有一
+个yield，也需要两次随机选择到同样的generator的时候，才会释放该段内存。</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="k">def</span> <span class="nf">func</span><span class="p">():</span>
+    <span class="k">yield</span> <span class="mi">0</span>
+<span class="n">f</span> <span class="o">=</span> <span class="n">func</span><span class="p">()</span>  <span class="c1"># 创建generator</span>
+<span class="n">tmp</span> <span class="o">=</span> <span class="nb">next</span><span class="p">(</span><span class="n">f</span><span class="p">)</span>  <span class="c1"># 调用一次，返回0</span>
+<span class="n">tmp</span> <span class="o">=</span> <span class="nb">next</span><span class="p">(</span><span class="n">f</span><span class="p">)</span>  <span class="c1"># 调用第二次的时候，才会Stop Iteration</span>
+</pre></div>
+</div>
+<p>而如果按顺序调用这些generator就不会出现这个问题。</p>
+<p>所以最佳实践推荐不要将每一个样本都放入train.list。而是将样本的地址放入另一个文本
+文件，train.list写入那个文本文件的地址。 或者在python generator的上下文中尽量留
+下非常少的变量引用。例如</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="k">def</span> <span class="nf">real_process</span><span class="p">(</span><span class="n">fn</span><span class="p">):</span>
+    <span class="c1"># ... read from fn</span>
+    <span class="k">return</span> <span class="n">result</span>   <span class="c1"># 当函数返回的时候，python可以解除掉内部变量的引用。</span>
+<span class="k">def</span> <span class="nf">process</span><span class="p">(</span><span class="n">fn</span><span class="p">):</span>
+    <span class="k">yield</span> <span class="n">real_process</span><span class="p">(</span><span class="n">fn</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>这个问题是PyDataProvider读数据时候的逻辑问题，基本上不能整体修正。</p>
+</div>
+<div class="section" id="id6">
+<h3>内存不够用的情况<a class="headerlink" href="#id6" title="Permalink to this headline">¶</a></h3>
+<p>PyDataProvider2会尽量使用内存。所以如果对于内存比较小的机器，推荐设置
+<code class="code docutils literal"><span class="pre">pool_size</span></code> 变量，而这个变量推荐大于训练的batch size，并且在内存足够
+的情况下越大越好。</p>
+</div>
+</div>
 </div>
@@ -391,6 +495,11 @@ DataProvider</li>
 <li><a class="reference internal" href="#cache">cache</a></li>
 </ul>
 </li>
+<li><a class="reference internal" href="#id4">注意事项</a><ul>
+<li><a class="reference internal" href="#id5">可能的内存泄露问题</a></li>
+<li><a class="reference internal" href="#id6">内存不够用的情况</a></li>
+</ul>
+</li>
 </ul>
 </li>
 </ul>