Deploy to GitHub Pages: 5f90a31f

develop/doc/_sources/api/v2/fluid/layers.rst.txt

@@ -300,3 +300,7 @@ conv2d_transpose
 ..  autofunction:: paddle.v2.fluid.layers.conv2d_transpose
     :noindex:
 
+sequence_expand
+---------
+..  autofunction:: paddle.v2.fluid.layers.sequence_expand
+    :noindex:

develop/doc/_sources/design/mkl/mkl_packed.md.txt

@@ -30,10 +30,10 @@
 由于在现有的某些情况下（例如RNN），多次调用 cblas_?gemm 会使用相同的原数据，因此，每次调用时对原数据的重复Packing便成为了冗余。
 
 为了最大程度减少多次调用 cblas_?gemm 在Packing上的耗时，Intel® MKL 引入了以下四个API:
-   * cblas_?gemm_alloc
-   * cblas_?gemm_pack 
-   * cblas_?gemm_compute
-   * cblas_?gemm_free
+   * [cblas_?gemm_alloc](https://software.intel.com/en-us/mkl-developer-reference-c-cblas-gemm-alloc)
+   * [cblas_?gemm_pack](https://software.intel.com/en-us/mkl-developer-reference-c-cblas-gemm-pack)
+   * [cblas_?gemm_compute](https://software.intel.com/en-us/mkl-developer-reference-c-cblas-gemm-compute)
+   * [cblas_?gemm_free](https://software.intel.com/en-us/mkl-developer-reference-c-cblas-gemm-free)
 
 通过使用这些API，我们可以先完成对原数据的Packing操作，再把已转换为Packed格式的数据传递给那些复用同一数据的gemm_compute函数，从而避免了Packing冗余。
 
@@ -84,7 +84,20 @@ PaddlePaddle/Paddle
 2. 对比优化后layer与相对应的PaddlePaddle原有layer, 在batch mode下的结果。
 
 ### Python API
-TBD
+计划在`paddle/utils.Flags`中添加`use_mkl_packed`的flag，用于选择是否使用相关功能，并且当编译时`WITH_MKL=ON`的情况下，默认设置为`true`。
+
+同时，在`python/paddle/trainer/config_parser.py`中对应的layer处，添加`use_mkl_packed`这个选择，方便用户在Python端选择是否启用这个功能。
+
+具体实现方式比如：
+
+```python
+use_mkl_packed = bool(int(g_command_config_args.get("use_mkl_packed", 0)))
+if use_mkl_packed:
+    self.layer_type = mkl_packed_*
+```
+
+所有相关的`layer_type`会以*mkl_packed_*开头，这些会在`MKLPacked*Layer`注册layer的时候保证，以示区分。 
+
 
 ### Benchmarking
 会添加相应的脚本用于测试和对比在使用MKL Packed recurrent layers 前后的网络性能。

develop/doc/api/v2/fluid/layers.html

@@ -1065,6 +1065,79 @@ stride_H = stride_W = stride.</li>
 </table>
 </dd></dl>
 
+</div>
+<div class="section" id="sequence-expand">
+<h2>sequence_expand<a class="headerlink" href="#sequence-expand" title="Permalink to this headline">¶</a></h2>
+<dl class="function">
+<dt>
+<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">sequence_expand</code><span class="sig-paren">(</span><em>x</em>, <em>y</em>, <em>main_program=None</em>, <em>startup_program=None</em><span class="sig-paren">)</span></dt>
+<dd><p>Sequence Expand Layer. This layer will expand the input variable <strong>x</strong>
+according to LoD information of <strong>y</strong>. And the following examples will
+explain how sequence_expand works:</p>
+<div class="highlight-text"><div class="highlight"><pre><span></span>* Case 1
+    x is a LoDTensor:
+        x.lod = [[0,       2, 3],
+                 [0, 1,    3, 4]]
+        x.data = [a, b, c, d]
+        x.dims = [4, 1]
+
+    y is a LoDTensor:
+        y.lod = [[0,    2,    4],
+                 [0, 3, 6, 7, 8]]
+
+    with condition len(y.lod[-1]) - 1 == x.dims[0]
+
+    then output is a 2-level LoDTensor:
+        out.lod = [[0,                2,    4],
+                   [0,       3,       6, 7, 8]]
+        out.data = [a, a, a, b, b, b, c, d]
+        out.dims = [8, 1]
+
+* Case 2
+    x is a Tensor:
+        x.data = [a, b, c]
+        x.dims = [3, 1]
+
+    y is a LoDTensor:
+        y.lod = [[0, 2, 3, 6]]
+
+    with condition len(y.lod[-1]) - 1 == x.dims[0]
+
+    then output is a 1-level LoDTensor:
+        out.lod = [[0,    2, 3,      6]]
+        out.data = [a, a, b, c, c, c]
+        out.dims = [6, 1]
+</pre></div>
+</div>
+<table class="docutils field-list" frame="void" rules="none">
+<col class="field-name" />
+<col class="field-body" />
+<tbody valign="top">
+<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
+<li><strong>x</strong> (<em>Variable</em>) &#8211; The input variable which is a Tensor or LoDTensor.</li>
+<li><strong>y</strong> (<em>Variable</em>) &#8211; The input variable which is a LoDTensor.</li>
+<li><strong>main_program</strong> (<em>Program</em>) &#8211; The main program.</li>
+<li><strong>startup_program</strong> (<em>Program</em>) &#8211; The startup program.</li>
+</ul>
+</td>
+</tr>
+<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The expanded variable which is a LoDTensor.</p>
+</td>
+</tr>
+<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">Variable</p>
+</td>
+</tr>
+</tbody>
+</table>
+<p class="rubric">Examples</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">x</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;x&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">10</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">)</span>
+<span class="n">y</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;y&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">10</span><span class="p">,</span> <span class="mi">20</span><span class="p">],</span>
+                 <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">,</span> <span class="n">lod_level</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+<span class="n">out</span> <span class="o">=</span> <span class="n">layers</span><span class="o">.</span><span class="n">sequence_expand</span><span class="p">(</span><span class="n">x</span><span class="o">=</span><span class="n">x</span><span class="p">,</span> <span class="n">y</span><span class="o">=</span><span class="n">y</span><span class="p">)</span>
+</pre></div>
+</div>
+</dd></dl>
+
 </div>
 </div>
 

develop/doc/design/mkl/mkl_packed.html

@@ -238,12 +238,14 @@
 <li>转换冗余 由于在现有的某些情况下（例如RNN），多次调用 cblas_?gemm 会使用相同的原数据，因此，每次调用时对原数据的重复Packing便成为了冗余。</li>
 </ol>
 <p>为了最大程度减少多次调用 cblas_?gemm 在Packing上的耗时，Intel® MKL 引入了以下四个API:</p>
-<ul class="simple">
-<li>cblas_?gemm_alloc</li>
-<li>cblas_?gemm_pack</li>
-<li>cblas_?gemm_compute</li>
-<li>cblas_?gemm_free</li>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://software.intel.com/en-us/mkl-developer-reference-c-cblas-gemm-alloc">cblas</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://software.intel.com/en-us/mkl-developer-reference-c-cblas-gemm-pack">cblas</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://software.intel.com/en-us/mkl-developer-reference-c-cblas-gemm-compute">cblas</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://software.intel.com/en-us/mkl-developer-reference-c-cblas-gemm-free">cblas</a></li>
 </ul>
+</div>
 <p>通过使用这些API，我们可以先完成对原数据的Packing操作，再把已转换为Packed格式的数据传递给那些复用同一数据的gemm_compute函数，从而避免了Packing冗余。</p>
 </div>
 <div class="section" id="solution">
@@ -303,7 +305,15 @@
 </div>
 <div class="section" id="python-api">
 <span id="python-api"></span><h3>Python API<a class="headerlink" href="#python-api" title="Permalink to this headline">¶</a></h3>
-<p>TBD</p>
+<p>计划在<code class="docutils literal"><span class="pre">paddle/utils.Flags</span></code>中添加<code class="docutils literal"><span class="pre">use_mkl_packed</span></code>的flag，用于选择是否使用相关功能，并且当编译时<code class="docutils literal"><span class="pre">WITH_MKL=ON</span></code>的情况下，默认设置为<code class="docutils literal"><span class="pre">true</span></code>。</p>
+<p>同时，在<code class="docutils literal"><span class="pre">python/paddle/trainer/config_parser.py</span></code>中对应的layer处，添加<code class="docutils literal"><span class="pre">use_mkl_packed</span></code>这个选择，方便用户在Python端选择是否启用这个功能。</p>
+<p>具体实现方式比如：</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">use_mkl_packed</span> <span class="o">=</span> <span class="nb">bool</span><span class="p">(</span><span class="nb">int</span><span class="p">(</span><span class="n">g_command_config_args</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;use_mkl_packed&quot;</span><span class="p">,</span> <span class="mi">0</span><span class="p">)))</span>
+<span class="k">if</span> <span class="n">use_mkl_packed</span><span class="p">:</span>
+    <span class="bp">self</span><span class="o">.</span><span class="n">layer_type</span> <span class="o">=</span> <span class="n">mkl_packed_</span><span class="o">*</span>
+</pre></div>
+</div>
+<p>所有相关的<code class="docutils literal"><span class="pre">layer_type</span></code>会以*mkl_packed_*开头，这些会在<code class="docutils literal"><span class="pre">MKLPacked*Layer</span></code>注册layer的时候保证，以示区分。</p>
 </div>
 <div class="section" id="benchmarking">
 <span id="benchmarking"></span><h3>Benchmarking<a class="headerlink" href="#benchmarking" title="Permalink to this headline">¶</a></h3>

develop/doc_cn/_sources/api/v2/fluid/layers.rst.txt

@@ -300,3 +300,7 @@ conv2d_transpose
 ..  autofunction:: paddle.v2.fluid.layers.conv2d_transpose
     :noindex:
 
+sequence_expand
+---------
+..  autofunction:: paddle.v2.fluid.layers.sequence_expand
+    :noindex:

develop/doc_cn/_sources/design/mkl/mkl_packed.md.txt

@@ -30,10 +30,10 @@
 由于在现有的某些情况下（例如RNN），多次调用 cblas_?gemm 会使用相同的原数据，因此，每次调用时对原数据的重复Packing便成为了冗余。
 
 为了最大程度减少多次调用 cblas_?gemm 在Packing上的耗时，Intel® MKL 引入了以下四个API:
-   * cblas_?gemm_alloc
-   * cblas_?gemm_pack 
-   * cblas_?gemm_compute
-   * cblas_?gemm_free
+   * [cblas_?gemm_alloc](https://software.intel.com/en-us/mkl-developer-reference-c-cblas-gemm-alloc)
+   * [cblas_?gemm_pack](https://software.intel.com/en-us/mkl-developer-reference-c-cblas-gemm-pack)
+   * [cblas_?gemm_compute](https://software.intel.com/en-us/mkl-developer-reference-c-cblas-gemm-compute)
+   * [cblas_?gemm_free](https://software.intel.com/en-us/mkl-developer-reference-c-cblas-gemm-free)
 
 通过使用这些API，我们可以先完成对原数据的Packing操作，再把已转换为Packed格式的数据传递给那些复用同一数据的gemm_compute函数，从而避免了Packing冗余。
 
@@ -84,7 +84,20 @@ PaddlePaddle/Paddle
 2. 对比优化后layer与相对应的PaddlePaddle原有layer, 在batch mode下的结果。
 
 ### Python API
-TBD
+计划在`paddle/utils.Flags`中添加`use_mkl_packed`的flag，用于选择是否使用相关功能，并且当编译时`WITH_MKL=ON`的情况下，默认设置为`true`。
+
+同时，在`python/paddle/trainer/config_parser.py`中对应的layer处，添加`use_mkl_packed`这个选择，方便用户在Python端选择是否启用这个功能。
+
+具体实现方式比如：
+
+```python
+use_mkl_packed = bool(int(g_command_config_args.get("use_mkl_packed", 0)))
+if use_mkl_packed:
+    self.layer_type = mkl_packed_*
+```
+
+所有相关的`layer_type`会以*mkl_packed_*开头，这些会在`MKLPacked*Layer`注册layer的时候保证，以示区分。 
+
 
 ### Benchmarking
 会添加相应的脚本用于测试和对比在使用MKL Packed recurrent layers 前后的网络性能。

develop/doc_cn/api/v2/fluid/layers.html

@@ -1084,6 +1084,79 @@ stride_H = stride_W = stride.</li>
 </table>
 </dd></dl>
 
+</div>
+<div class="section" id="sequence-expand">
+<h2>sequence_expand<a class="headerlink" href="#sequence-expand" title="永久链接至标题">¶</a></h2>
+<dl class="function">
+<dt>
+<code class="descclassname">paddle.v2.fluid.layers.</code><code class="descname">sequence_expand</code><span class="sig-paren">(</span><em>x</em>, <em>y</em>, <em>main_program=None</em>, <em>startup_program=None</em><span class="sig-paren">)</span></dt>
+<dd><p>Sequence Expand Layer. This layer will expand the input variable <strong>x</strong>
+according to LoD information of <strong>y</strong>. And the following examples will
+explain how sequence_expand works:</p>
+<div class="highlight-text"><div class="highlight"><pre><span></span>* Case 1
+    x is a LoDTensor:
+        x.lod = [[0,       2, 3],
+                 [0, 1,    3, 4]]
+        x.data = [a, b, c, d]
+        x.dims = [4, 1]
+
+    y is a LoDTensor:
+        y.lod = [[0,    2,    4],
+                 [0, 3, 6, 7, 8]]
+
+    with condition len(y.lod[-1]) - 1 == x.dims[0]
+
+    then output is a 2-level LoDTensor:
+        out.lod = [[0,                2,    4],
+                   [0,       3,       6, 7, 8]]
+        out.data = [a, a, a, b, b, b, c, d]
+        out.dims = [8, 1]
+
+* Case 2
+    x is a Tensor:
+        x.data = [a, b, c]
+        x.dims = [3, 1]
+
+    y is a LoDTensor:
+        y.lod = [[0, 2, 3, 6]]
+
+    with condition len(y.lod[-1]) - 1 == x.dims[0]
+
+    then output is a 1-level LoDTensor:
+        out.lod = [[0,    2, 3,      6]]
+        out.data = [a, a, b, c, c, c]
+        out.dims = [6, 1]
+</pre></div>
+</div>
+<table class="docutils field-list" frame="void" rules="none">
+<col class="field-name" />
+<col class="field-body" />
+<tbody valign="top">
+<tr class="field-odd field"><th class="field-name">参数:</th><td class="field-body"><ul class="first simple">
+<li><strong>x</strong> (<em>Variable</em>) &#8211; The input variable which is a Tensor or LoDTensor.</li>
+<li><strong>y</strong> (<em>Variable</em>) &#8211; The input variable which is a LoDTensor.</li>
+<li><strong>main_program</strong> (<em>Program</em>) &#8211; The main program.</li>
+<li><strong>startup_program</strong> (<em>Program</em>) &#8211; The startup program.</li>
+</ul>
+</td>
+</tr>
+<tr class="field-even field"><th class="field-name">返回:</th><td class="field-body"><p class="first">The expanded variable which is a LoDTensor.</p>
+</td>
+</tr>
+<tr class="field-odd field"><th class="field-name">返回类型:</th><td class="field-body"><p class="first last">Variable</p>
+</td>
+</tr>
+</tbody>
+</table>
+<p class="rubric">Examples</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">x</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;x&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">10</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">)</span>
+<span class="n">y</span> <span class="o">=</span> <span class="n">fluid</span><span class="o">.</span><span class="n">layers</span><span class="o">.</span><span class="n">data</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;y&#39;</span><span class="p">,</span> <span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="mi">10</span><span class="p">,</span> <span class="mi">20</span><span class="p">],</span>
+                 <span class="n">dtype</span><span class="o">=</span><span class="s1">&#39;float32&#39;</span><span class="p">,</span> <span class="n">lod_level</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+<span class="n">out</span> <span class="o">=</span> <span class="n">layers</span><span class="o">.</span><span class="n">sequence_expand</span><span class="p">(</span><span class="n">x</span><span class="o">=</span><span class="n">x</span><span class="p">,</span> <span class="n">y</span><span class="o">=</span><span class="n">y</span><span class="p">)</span>
+</pre></div>
+</div>
+</dd></dl>
+
 </div>
 </div>
 

develop/doc_cn/design/mkl/mkl_packed.html

@@ -257,12 +257,14 @@
 <li>转换冗余 由于在现有的某些情况下（例如RNN），多次调用 cblas_?gemm 会使用相同的原数据，因此，每次调用时对原数据的重复Packing便成为了冗余。</li>
 </ol>
 <p>为了最大程度减少多次调用 cblas_?gemm 在Packing上的耗时，Intel® MKL 引入了以下四个API:</p>
-<ul class="simple">
-<li>cblas_?gemm_alloc</li>
-<li>cblas_?gemm_pack</li>
-<li>cblas_?gemm_compute</li>
-<li>cblas_?gemm_free</li>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://software.intel.com/en-us/mkl-developer-reference-c-cblas-gemm-alloc">cblas</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://software.intel.com/en-us/mkl-developer-reference-c-cblas-gemm-pack">cblas</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://software.intel.com/en-us/mkl-developer-reference-c-cblas-gemm-compute">cblas</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://software.intel.com/en-us/mkl-developer-reference-c-cblas-gemm-free">cblas</a></li>
 </ul>
+</div>
 <p>通过使用这些API，我们可以先完成对原数据的Packing操作，再把已转换为Packed格式的数据传递给那些复用同一数据的gemm_compute函数，从而避免了Packing冗余。</p>
 </div>
 <div class="section" id="solution">
@@ -322,7 +324,15 @@
 </div>
 <div class="section" id="python-api">
 <span id="python-api"></span><h3>Python API<a class="headerlink" href="#python-api" title="永久链接至标题">¶</a></h3>
-<p>TBD</p>
+<p>计划在<code class="docutils literal"><span class="pre">paddle/utils.Flags</span></code>中添加<code class="docutils literal"><span class="pre">use_mkl_packed</span></code>的flag，用于选择是否使用相关功能，并且当编译时<code class="docutils literal"><span class="pre">WITH_MKL=ON</span></code>的情况下，默认设置为<code class="docutils literal"><span class="pre">true</span></code>。</p>
+<p>同时，在<code class="docutils literal"><span class="pre">python/paddle/trainer/config_parser.py</span></code>中对应的layer处，添加<code class="docutils literal"><span class="pre">use_mkl_packed</span></code>这个选择，方便用户在Python端选择是否启用这个功能。</p>
+<p>具体实现方式比如：</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">use_mkl_packed</span> <span class="o">=</span> <span class="nb">bool</span><span class="p">(</span><span class="nb">int</span><span class="p">(</span><span class="n">g_command_config_args</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;use_mkl_packed&quot;</span><span class="p">,</span> <span class="mi">0</span><span class="p">)))</span>
+<span class="k">if</span> <span class="n">use_mkl_packed</span><span class="p">:</span>
+    <span class="bp">self</span><span class="o">.</span><span class="n">layer_type</span> <span class="o">=</span> <span class="n">mkl_packed_</span><span class="o">*</span>
+</pre></div>
+</div>
+<p>所有相关的<code class="docutils literal"><span class="pre">layer_type</span></code>会以*mkl_packed_*开头，这些会在<code class="docutils literal"><span class="pre">MKLPacked*Layer</span></code>注册layer的时候保证，以示区分。</p>
 </div>
 <div class="section" id="benchmarking">
 <span id="benchmarking"></span><h3>Benchmarking<a class="headerlink" href="#benchmarking" title="永久链接至标题">¶</a></h3>