Skip to content

P
Paddle

机器未来 / Paddle
与 Fork 源项目一致

Fork自 PaddlePaddle / Paddle

通知 1
Star 1
Fork 0
P
Paddle
提交 8bcf7dab 编写于 作者: T Travis CI
浏览文件
操作

Deploy to GitHub Pages: 5926e9a2

上级 9d853b85
隐藏空白更改
内联 并排
Showing with 240 addition and 340 deletion
develop/doc/api/v2/config/layer.html
浏览文件 @ 8bcf7dab
......@@ -1179,14 +1179,10 @@ factors which dimensions equal to the channel&#8217;s number.</p>
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.layer.</code><code class="descname">row_l2_norm</code></dt>
<dd><blockquote>
<div><p>A layer for L2-normalization in each row.</p>
<dd><p>A layer for L2-normalization in each row.</p>
<div class="math">
\[out[i] =\]</div>
</div></blockquote>
<p>rac{in[i]}{sqrt{sum_{k=1}^N in[k]^{2}}}</p>
<blockquote>
<div><p>where the size of <span class="math">\(in\)</span> is (batchSize x dataDim) ,
\[out[i] = \frac{in[i]} {\sqrt{\sum_{k=1}^N in[k]^{2}}}\]</div>
<p>where the size of <span class="math">\(in\)</span> is (batchSize x dataDim) ,
and the size of <span class="math">\(out\)</span> is a (batchSize x dataDim) .</p>
<p>The example usage is:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">row_l2_norm</span> <span class="o">=</span> <span class="n">row_l2_norm</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">layer</span><span class="p">)</span>
......@@ -1196,28 +1192,22 @@ and the size of <span class="math">\(out\)</span> is a (batchSize x dataDim) .</
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">param input:</th><td class="field-body">The input of this layer.</td>
</tr>
<tr class="field-even field"><th class="field-name">type input:</th><td class="field-body">paddle.v2.config_base.Layer</td>
</tr>
<tr class="field-odd field"><th class="field-name">param name:</th><td class="field-body">The name of this layer. It is optional.</td>
</tr>
<tr class="field-even field"><th class="field-name">type name:</th><td class="field-body">basestring</td>
</tr>
<tr class="field-odd field"><th class="field-name" colspan="2">param layer_attr:</th></tr>
<tr class="field-odd field"><td>&#160;</td><td class="field-body">The extra layer attribute. See paddle.v2.attr.ExtraAttribute
for details.</td>
</tr>
<tr class="field-even field"><th class="field-name" colspan="2">type layer_attr:</th></tr>
<tr class="field-even field"><td>&#160;</td><td class="field-body">paddle.v2.attr.ExtraAttribute</td>
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>paddle.v2.config_base.Layer</em>) &#8211; The input of this layer.</li>
<li><strong>name</strong> (<em>basestring</em>) &#8211; The name of this layer. It is optional.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute</em>) &#8211; The extra layer attribute. See paddle.v2.attr.ExtraAttribute
for details.</li>
</ul>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">return:</th><td class="field-body">paddle.v2.config_base.Layer object.</td>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">paddle.v2.config_base.Layer object.</p>
</td>
</tr>
<tr class="field-even field"><th class="field-name">rtype:</th><td class="field-body">paddle.v2.config_base.Layer</td>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">paddle.v2.config_base.Layer</p>
</td>
</tr>
</tbody>
</table>
</div></blockquote>
</dd></dl>
</div>
......@@ -1286,26 +1276,29 @@ be included in the configuration file to complete the input-to-hidden
mappings before lstmemory is called.</p>
<p>NOTE: This is a low level user interface. You can use network.simple_lstm
to config a simple plain lstm layer.</p>
<p>Please refer to <strong>Generating Sequences With Recurrent Neural Networks</strong> for
more details about LSTM.</p>
<p><a class="reference external" href="http://arxiv.org/abs/1308.0850">Link</a> goes as below.</p>
<dl class="docutils">
<dt>Reference:</dt>
<dd><a class="reference external" href="https://arxiv.org/pdf/1308.0850.pdf">Generating Sequences With Recurrent Neural Networks</a></dd>
</dl>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>name</strong> (<em>basestring</em>) &#8211; The lstmemory layer name.</li>
<li><strong>size</strong> (<em>int</em>) &#8211; DEPRECATED. size of the lstm cell</li>
<li><strong>name</strong> (<em>basestring</em>) &#8211; The name of this layer. It is optional.</li>
<li><strong>size</strong> (<em>int</em>) &#8211; DEPRECATED. The dimension of the lstm cell.</li>
<li><strong>input</strong> (<em>paddle.v2.config_base.Layer</em>) &#8211; The input of this layer.</li>
<li><strong>reverse</strong> (<em>bool</em>) &#8211; is sequence process reversed or not.</li>
<li><strong>reverse</strong> (<em>bool</em>) &#8211; Whether the input sequence is processed in a reverse order.</li>
<li><strong>act</strong> (<em>paddle.v2.activation.Base</em>) &#8211; Activation type. paddle.v2.activation.Tanh is the default activation.</li>
<li><strong>gate_act</strong> (<em>paddle.v2.activation.Base</em>) &#8211; gate activation type, paddle.v2.activation.Sigmoid by default.</li>
<li><strong>state_act</strong> (<em>paddle.v2.activation.Base</em>) &#8211; state activation type, paddle.v2.activation.Tanh by default.</li>
<li><strong>gate_act</strong> (<em>paddle.v2.activation.Base</em>) &#8211; Activation type of this layer&#8217;s gates. paddle.v2.activation.Sigmoid is the
default activation.</li>
<li><strong>state_act</strong> (<em>paddle.v2.activation.Base</em>) &#8211; Activation type of the state. paddle.v2.activation.Tanh is the default activation.</li>
<li><strong>bias_attr</strong> (<em>paddle.v2.attr.ParameterAttribute | None | bool | Any</em>) &#8211; The bias attribute. If the parameter is set to False or an object
whose type is not paddle.v2.attr.ParameterAttribute, no bias is defined. If the
parameter is set to True, the bias is initialized to zero.</li>
<li><strong>param_attr</strong> (<em>paddle.v2.attr.ParameterAttribute | None | False</em>) &#8211; Parameter Attribute.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute | None</em>) &#8211; Extra Layer attribute</li>
<li><strong>param_attr</strong> (<em>paddle.v2.attr.ParameterAttribute</em>) &#8211; The parameter attribute. See paddle.v2.attr.ParameterAttribute for details.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute | None</em>) &#8211; The extra layer attribute. See paddle.v2.attr.ExtraAttribute for
details.</li>
</ul>
</td>
</tr>
......@@ -1346,12 +1339,14 @@ candidate activation <span class="math">\(\tilde{h_t}\)</span>:</p>
<div class="math">
\[h_t = (1 - z_t) h_{t-1} + z_t {\tilde{h_t}}\]</div>
<p>NOTE: In PaddlePaddle&#8217;s implementation, the multiplication operations
<span class="math">\(W_{r}x_{t}\)</span>, <span class="math">\(W_{z}x_{t}\)</span> and <span class="math">\(W x_t\)</span> are not computed in
gate_recurrent layer. Consequently, an additional mixed with
<span class="math">\(W_{r}x_{t}\)</span>, <span class="math">\(W_{z}x_{t}\)</span> and <span class="math">\(W x_t\)</span> are not performed
in gate_recurrent layer. Consequently, an additional mixed with
full_matrix_projection or a fc must be included before grumemory
is called.</p>
<p>More details can be found by referring to <a class="reference external" href="https://arxiv.org/abs/1412.3555">Empirical Evaluation of Gated
Recurrent Neural Networks on Sequence Modeling.</a></p>
<dl class="docutils">
<dt>Reference:</dt>
<dd><a class="reference external" href="https://arxiv.org/abs/1412.3555">Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling</a></dd>
</dl>
<p>The simple usage is:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">gru</span> <span class="o">=</span> <span class="n">grumemory</span><span class="p">(</span><span class="nb">input</span><span class="p">)</span>
</pre></div>
......@@ -1361,20 +1356,21 @@ Recurrent Neural Networks on Sequence Modeling.</a></p>
<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>name</strong> (<em>None | basestring</em>) &#8211; The gru layer name.</li>
<li><strong>name</strong> (<em>basestring</em>) &#8211; The name of this layer. It is optional.</li>
<li><strong>input</strong> (<em>paddle.v2.config_base.Layer.</em>) &#8211; The input of this layer.</li>
<li><strong>size</strong> (<em>int</em>) &#8211; DEPRECATED. size of the gru cell</li>
<li><strong>reverse</strong> (<em>bool</em>) &#8211; Whether sequence process is reversed or not.</li>
<li><strong>size</strong> (<em>int</em>) &#8211; DEPRECATED. The dimension of the gru cell.</li>
<li><strong>reverse</strong> (<em>bool</em>) &#8211; Whether the input sequence is processed in a reverse order.</li>
<li><strong>act</strong> (<em>paddle.v2.activation.Base</em>) &#8211; Activation type, paddle.v2.activation.Tanh is the default. This activation
affects the <span class="math">\({\tilde{h_t}}\)</span>.</li>
<li><strong>gate_act</strong> (<em>paddle.v2.activation.Base</em>) &#8211; gate activation type, paddle.v2.activation.Sigmoid by default.
This activation affects the <span class="math">\(z_t\)</span> and <span class="math">\(r_t\)</span>. It is the
<span class="math">\(\sigma\)</span> in the above formula.</li>
<li><strong>gate_act</strong> (<em>paddle.v2.activation.Base</em>) &#8211; Activation type of this layer&#8217;s two gates. paddle.v2.activation.Sigmoid is
the default activation. This activation affects the <span class="math">\(z_t\)</span>
and <span class="math">\(r_t\)</span>. It is the <span class="math">\(\sigma\)</span> in the above formula.</li>
<li><strong>bias_attr</strong> (<em>paddle.v2.attr.ParameterAttribute | None | bool | Any</em>) &#8211; The bias attribute. If the parameter is set to False or an object
whose type is not paddle.v2.attr.ParameterAttribute, no bias is defined. If the
parameter is set to True, the bias is initialized to zero.</li>
<li><strong>param_attr</strong> (<em>paddle.v2.attr.ParameterAttribute | None | False</em>) &#8211; Parameter Attribute.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute | None</em>) &#8211; Extra Layer attribute</li>
<li><strong>param_attr</strong> (<em>paddle.v2.attr.ParameterAttribute</em>) &#8211; The parameter attribute. See paddle.v2.attr.ParameterAttribute for details.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute | None</em>) &#8211; The extra layer attribute. See paddle.v2.attr.ExtraAttribute for
details.</li>
</ul>
</td>
</tr>
......@@ -2219,10 +2215,10 @@ parameter is set to True, the bias is initialized to zero.</li>
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.layer.</code><code class="descname">last_seq</code></dt>
<dd><p>Get Last Timestamp Activation of a sequence.</p>
<p>If stride &gt; 0, this layer slides a window whose size is determined by stride,
and return the last value of the window as the output. Thus, a long sequence
will be shorten. Note that for sequence with sub-sequence, the default value
of stride is -1.</p>
<p>If stride &gt; 0, this layer will slide a window whose size is determined by stride,
and return the last value of the sequence in the window as the output. Thus, a
long sequence will be shortened. Note that for sequence with sub-sequence, the
default value of stride is -1.</p>
<p>The simple usage is:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">seq</span> <span class="o">=</span> <span class="n">last_seq</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">layer</span><span class="p">)</span>
</pre></div>
......@@ -2232,11 +2228,12 @@ of stride is -1.</p>
<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>agg_level</strong> &#8211; Aggregated level</li>
<li><strong>agg_level</strong> (<em>AggregateLevel</em>) &#8211; Aggregated level</li>
<li><strong>name</strong> (<em>basestring</em>) &#8211; The name of this layer. It is optional.</li>
<li><strong>input</strong> (<em>paddle.v2.config_base.Layer</em>) &#8211; The input of this layer.</li>
<li><strong>stride</strong> (<em>Int</em>) &#8211; The step size between successive pooling regions.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute</em>) &#8211; extra layer attributes.</li>
<li><strong>stride</strong> (<em>int</em>) &#8211; The step size between successive pooling regions.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute</em>) &#8211; The extra layer attribute. See paddle.v2.attr.ExtraAttribute for
details.</li>
</ul>
</td>
</tr>
......@@ -2257,10 +2254,10 @@ of stride is -1.</p>
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.layer.</code><code class="descname">first_seq</code></dt>
<dd><p>Get First Timestamp Activation of a sequence.</p>
<p>If stride &gt; 0, this layer slides a window whose size is determined by stride,
and return the first value of the window as the output. Thus, a long sequence
will be shorten. Note that for sequence with sub-sequence, the default value
of stride is -1.</p>
<p>If stride &gt; 0, this layer will slide a window whose size is determined by stride,
and return the first value of the sequence in the window as the output. Thus, a
long sequence will be shortened. Note that for sequence with sub-sequence, the
default value of stride is -1.</p>
<p>The simple usage is:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">seq</span> <span class="o">=</span> <span class="n">first_seq</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">layer</span><span class="p">)</span>
</pre></div>
......@@ -2270,11 +2267,12 @@ of stride is -1.</p>
<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>agg_level</strong> &#8211; aggregation level</li>
<li><strong>agg_level</strong> (<em>AggregateLevel</em>) &#8211; aggregation level</li>
<li><strong>name</strong> (<em>basestring</em>) &#8211; The name of this layer. It is optional.</li>
<li><strong>input</strong> (<em>paddle.v2.config_base.Layer</em>) &#8211; The input of this layer.</li>
<li><strong>stride</strong> (<em>Int</em>) &#8211; The step size between successive pooling regions.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute</em>) &#8211; extra layer attributes.</li>
<li><strong>stride</strong> (<em>int</em>) &#8211; The step size between successive pooling regions.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute</em>) &#8211; The extra layer attribute. See paddle.v2.attr.ExtraAttribute for
details.</li>
</ul>
</td>
</tr>
......@@ -2547,8 +2545,8 @@ details.</li>
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.layer.</code><code class="descname">expand</code></dt>
<dd><p>A layer for &#8220;Expand Dense data or (sequence data where the length of each
sequence is one) to sequence data.&#8221;</p>
<dd><p>A layer for expanding dense data or (sequence data where the length of each
sequence is one) to sequence data.</p>
<p>The example usage is:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">expand</span> <span class="o">=</span> <span class="n">expand</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">layer1</span><span class="p">,</span>
<span class="n">expand_as</span><span class="o">=</span><span class="n">layer2</span><span class="p">,</span>
......@@ -2561,13 +2559,16 @@ sequence is one) to sequence data.&#8221;</p>
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>paddle.v2.config_base.Layer</em>) &#8211; The input of this layer.</li>
<li><strong>expand_as</strong> (<em>paddle.v2.config_base.Layer</em>) &#8211; Expand as this layer&#8217;s sequence info.</li>
<li><strong>expand_as</strong> (<em>paddle.v2.config_base.Layer</em>) &#8211; Expand the input according to this layer&#8217;s sequence infomation. And
after the operation, the input expanded will have the same number of
elememts as this layer.</li>
<li><strong>name</strong> (<em>basestring</em>) &#8211; The name of this layer. It is optional.</li>
<li><strong>bias_attr</strong> (<em>paddle.v2.attr.ParameterAttribute | None | bool | Any</em>) &#8211; The bias attribute. If the parameter is set to False or an object
whose type is not paddle.v2.attr.ParameterAttribute, no bias is defined. If the
parameter is set to True, the bias is initialized to zero.</li>
<li><strong>expand_level</strong> (<em>ExpandLevel</em>) &#8211; whether input layer is timestep(default) or sequence.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute</em>) &#8211; extra layer attributes.</li>
<li><strong>expand_level</strong> (<em>ExpandLevel</em>) &#8211; Whether the input layer is a sequence or the element of a sequence.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute</em>) &#8211; The extra layer attribute. See paddle.v2.attr.ExtraAttribute for
details.</li>
</ul>
</td>
</tr>
......@@ -3055,44 +3056,32 @@ details.</li>
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.layer.</code><code class="descname">clip</code></dt>
<dd><blockquote>
<div><p>A layer for clipping the input value by the threshold.</p>
<dd><p>A layer for clipping the input value by the threshold.</p>
<div class="math">
\[out[i] = \min\left(\max\left(in[i],p_{1}\]</div>
</div></blockquote>
<p>ight),p_{2}
ight)</p>
<blockquote>
<div><div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">clip</span> <span class="o">=</span> <span class="n">clip</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="nb">input</span><span class="p">,</span> <span class="nb">min</span><span class="o">=-</span><span class="mi">10</span><span class="p">,</span> <span class="nb">max</span><span class="o">=</span><span class="mi">10</span><span class="p">)</span>
\[out[i] = \min (\max (in[i],p_{1} ),p_{2} )\]</div>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">clip</span> <span class="o">=</span> <span class="n">clip</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="nb">input</span><span class="p">,</span> <span class="nb">min</span><span class="o">=-</span><span class="mi">10</span><span class="p">,</span> <span class="nb">max</span><span class="o">=</span><span class="mi">10</span><span class="p">)</span>
</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">param name:</th><td class="field-body">The name of this layer. It is optional.</td>
</tr>
<tr class="field-even field"><th class="field-name">type name:</th><td class="field-body">basestring</td>
</tr>
<tr class="field-odd field"><th class="field-name">param input:</th><td class="field-body">The input of this layer.</td>
</tr>
<tr class="field-even field"><th class="field-name">type input:</th><td class="field-body">paddle.v2.config_base.Layer.</td>
</tr>
<tr class="field-odd field"><th class="field-name">param min:</th><td class="field-body">The lower threshold for clipping.</td>
</tr>
<tr class="field-even field"><th class="field-name">type min:</th><td class="field-body">float</td>
</tr>
<tr class="field-odd field"><th class="field-name">param max:</th><td class="field-body">The upper threshold for clipping.</td>
</tr>
<tr class="field-even field"><th class="field-name">type max:</th><td class="field-body">float</td>
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>name</strong> (<em>basestring</em>) &#8211; The name of this layer. It is optional.</li>
<li><strong>input</strong> (<em>paddle.v2.config_base.Layer.</em>) &#8211; The input of this layer.</li>
<li><strong>min</strong> (<em>float</em>) &#8211; The lower threshold for clipping.</li>
<li><strong>max</strong> (<em>float</em>) &#8211; The upper threshold for clipping.</li>
</ul>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">return:</th><td class="field-body">paddle.v2.config_base.Layer object.</td>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">paddle.v2.config_base.Layer object.</p>
</td>
</tr>
<tr class="field-even field"><th class="field-name">rtype:</th><td class="field-body">paddle.v2.config_base.Layer</td>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">paddle.v2.config_base.Layer</p>
</td>
</tr>
</tbody>
</table>
</div></blockquote>
</dd></dl>
</div>
......@@ -3762,18 +3751,13 @@ details.</li>
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.layer.</code><code class="descname">huber_regression_cost</code></dt>
<dd><blockquote>
<div>In statistics, the Huber loss is a loss function used in robust regression,
<dd><p>In statistics, the Huber loss is a loss function used in robust regression,
that is less sensitive to outliers in data than the squared error loss.
Given a prediction f(x), a label y and <span class="math">\(\delta\)</span>, the loss function
is defined as:</div></blockquote>
<p>ight )^2, left | y-f(x)
ight <a href="#id2"><span class="problematic" id="id3">|</span></a>leq delta</p>
<blockquote>
<div>loss = delta left | y-f(x)</div></blockquote>
<p>ight <a href="#id4"><span class="problematic" id="id5">|</span></a>-0.5delta ^2, otherwise</p>
<blockquote>
<div><p>The example usage is:</p>
is defined as:</p>
<div class="math">
\[ \begin{align}\begin{aligned}loss = 0.5*(y-f(x))^{2}, | y-f(x) | &lt; \delta\\loss = \delta | y-f(x) | - 0.5 \delta ^2, otherwise\end{aligned}\end{align} \]</div>
<p>The example usage is:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">cost</span> <span class="o">=</span> <span class="n">huber_regression_cost</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="nb">input</span><span class="p">,</span> <span class="n">label</span><span class="o">=</span><span class="n">label</span><span class="p">)</span>
</pre></div>
</div>
......@@ -3781,41 +3765,26 @@ ight <a href="#id2"><span class="problematic" id="id3">|</span></a>leq delta</p>
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">param input:</th><td class="field-body">The first input layer.</td>
</tr>
<tr class="field-even field"><th class="field-name">type input:</th><td class="field-body">paddle.v2.config_base.Layer</td>
</tr>
<tr class="field-odd field"><th class="field-name">param label:</th><td class="field-body">The input label.</td>
</tr>
<tr class="field-even field"><th class="field-name">type input:</th><td class="field-body">paddle.v2.config_base.Layer</td>
</tr>
<tr class="field-odd field"><th class="field-name">param name:</th><td class="field-body">The name of this layer. It is optional.</td>
</tr>
<tr class="field-even field"><th class="field-name">type name:</th><td class="field-body">basestring</td>
</tr>
<tr class="field-odd field"><th class="field-name">param delta:</th><td class="field-body">The difference between the observed and predicted values.</td>
</tr>
<tr class="field-even field"><th class="field-name">type delta:</th><td class="field-body">float</td>
</tr>
<tr class="field-odd field"><th class="field-name">param coeff:</th><td class="field-body">The weight of the gradient in the back propagation.
1.0 is the default value.</td>
</tr>
<tr class="field-even field"><th class="field-name">type coeff:</th><td class="field-body">float</td>
</tr>
<tr class="field-odd field"><th class="field-name" colspan="2">param layer_attr:</th></tr>
<tr class="field-odd field"><td>&#160;</td><td class="field-body">The extra layer attribute. See paddle.v2.attr.ExtraAttribute for
details.</td>
</tr>
<tr class="field-even field"><th class="field-name" colspan="2">type layer_attr:</th></tr>
<tr class="field-even field"><td>&#160;</td><td class="field-body">paddle.v2.attr.ExtraAttribute</td>
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>paddle.v2.config_base.Layer</em>) &#8211; The first input layer.</li>
<li><strong>label</strong> &#8211; The input label.</li>
<li><strong>name</strong> (<em>basestring</em>) &#8211; The name of this layer. It is optional.</li>
<li><strong>delta</strong> (<em>float</em>) &#8211; The difference between the observed and predicted values.</li>
<li><strong>coeff</strong> (<em>float</em>) &#8211; The weight of the gradient in the back propagation.
1.0 is the default value.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute</em>) &#8211; The extra layer attribute. See paddle.v2.attr.ExtraAttribute for
details.</li>
</ul>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">return:</th><td class="field-body">paddle.v2.config_base.Layer object.</td>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">paddle.v2.config_base.Layer object.</p>
</td>
</tr>
<tr class="field-even field"><th class="field-name">rtype:</th><td class="field-body">paddle.v2.config_base.Layer.</td>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">paddle.v2.config_base.Layer.</p>
</td>
</tr>
</tbody>
</table>
</div></blockquote>
</dd></dl>
</div>
......@@ -3824,56 +3793,37 @@ details.</td>
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.layer.</code><code class="descname">huber_classification_cost</code></dt>
<dd><blockquote>
<div>For classification purposes, a variant of the Huber loss called modified Huber
<dd><p>For classification purposes, a variant of the Huber loss called modified Huber
is sometimes used. Given a prediction f(x) (a real-valued classifier score) and
a true binary class label :math:<a href="#id6"><span class="problematic" id="id7">`</span></a>yin left {-1, 1</div></blockquote>
<dl class="docutils">
<dt>ight }`, the modified Huber</dt>
<dd>loss is defined as:</dd>
<dt>ight )^2, yf(x)geq 1</dt>
<dd><blockquote class="first">
<div>loss = -4yf(x), ext{otherwise}</div></blockquote>
a true binary class label <span class="math">\(y\in \{-1, 1 \}\)</span>, the modified Huber
loss is defined as:</p>
<p>The example usage is:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">cost</span> <span class="o">=</span> <span class="n">huber_classification_cost</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="nb">input</span><span class="p">,</span> <span class="n">label</span><span class="o">=</span><span class="n">label</span><span class="p">)</span>
</pre></div>
</div>
<table class="last docutils field-list" frame="void" rules="none">
<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">param input:</th><td class="field-body">The first input layer.</td>
</tr>
<tr class="field-even field"><th class="field-name">type input:</th><td class="field-body">paddle.v2.config_base.Layer</td>
</tr>
<tr class="field-odd field"><th class="field-name">param label:</th><td class="field-body">The input label.</td>
</tr>
<tr class="field-even field"><th class="field-name">type input:</th><td class="field-body">paddle.v2.config_base.Layer</td>
</tr>
<tr class="field-odd field"><th class="field-name">param name:</th><td class="field-body">The name of this layer. It is optional.</td>
</tr>
<tr class="field-even field"><th class="field-name">type name:</th><td class="field-body">basestring</td>
</tr>
<tr class="field-odd field"><th class="field-name">param coeff:</th><td class="field-body">The weight of the gradient in the back propagation.
1.0 is the default value.</td>
</tr>
<tr class="field-even field"><th class="field-name">type coeff:</th><td class="field-body">float</td>
</tr>
<tr class="field-odd field"><th class="field-name" colspan="2">param layer_attr:</th></tr>
<tr class="field-odd field"><td>&#160;</td><td class="field-body">The extra layer attribute. See paddle.v2.attr.ExtraAttribute for
details.</td>
</tr>
<tr class="field-even field"><th class="field-name" colspan="2">type layer_attr:</th></tr>
<tr class="field-even field"><td>&#160;</td><td class="field-body">paddle.v2.attr.ExtraAttribute</td>
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>paddle.v2.config_base.Layer</em>) &#8211; The first input layer.</li>
<li><strong>label</strong> &#8211; The input label.</li>
<li><strong>name</strong> (<em>basestring</em>) &#8211; The name of this layer. It is optional.</li>
<li><strong>coeff</strong> (<em>float</em>) &#8211; The weight of the gradient in the back propagation.
1.0 is the default value.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute</em>) &#8211; The extra layer attribute. See paddle.v2.attr.ExtraAttribute for
details.</li>
</ul>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">return:</th><td class="field-body">paddle.v2.config_base.Layer object.</td>
<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">paddle.v2.config_base.Layer object.</p>
</td>
</tr>
<tr class="field-even field"><th class="field-name">rtype:</th><td class="field-body">paddle.v2.config_base.Layer</td>
<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">paddle.v2.config_base.Layer</p>
</td>
</tr>
</tbody>
</table>
</dd>
</dl>
</dd></dl>
</div>
......@@ -4579,7 +4529,7 @@ details.</li>
<dd><p>The gated unit layer implements a simple gating mechanism over the input.
The input <span class="math">\(X\)</span> is first projected into a new space <span class="math">\(X'\)</span>, and
it is also used to produce a gate weight <span class="math">\(\sigma\)</span>. Element-wise
product between <a href="#id11"><span class="problematic" id="id12">:match:`X&#8217;`</span></a> and <span class="math">\(\sigma\)</span> is finally returned.</p>
product between <a href="#id5"><span class="problematic" id="id6">:match:`X&#8217;`</span></a> and <span class="math">\(\sigma\)</span> is finally returned.</p>
<dl class="docutils">
<dt>Reference:</dt>
<dd><a class="reference external" href="https://arxiv.org/abs/1612.08083">Language Modeling with Gated Convolutional Networks</a></dd>
......
develop/doc/searchindex.js
浏览文件 @ 8bcf7dab
因为 它太大了无法显示 source diff 。你可以改为 查看blob。
develop/doc_cn/api/v2/config/layer.html
浏览文件 @ 8bcf7dab
......@@ -1180,14 +1180,10 @@ factors which dimensions equal to the channel&#8217;s number.</p>
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.layer.</code><code class="descname">row_l2_norm</code></dt>
<dd><blockquote>
<div><p>A layer for L2-normalization in each row.</p>
<dd><p>A layer for L2-normalization in each row.</p>
<div class="math">
\[out[i] =\]</div>
</div></blockquote>
<p>rac{in[i]}{sqrt{sum_{k=1}^N in[k]^{2}}}</p>
<blockquote>
<div><p>where the size of <span class="math">\(in\)</span> is (batchSize x dataDim) ,
\[out[i] = \frac{in[i]} {\sqrt{\sum_{k=1}^N in[k]^{2}}}\]</div>
<p>where the size of <span class="math">\(in\)</span> is (batchSize x dataDim) ,
and the size of <span class="math">\(out\)</span> is a (batchSize x dataDim) .</p>
<p>The example usage is:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">row_l2_norm</span> <span class="o">=</span> <span class="n">row_l2_norm</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">layer</span><span class="p">)</span>
......@@ -1197,28 +1193,22 @@ and the size of <span class="math">\(out\)</span> is a (batchSize x dataDim) .</
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">param input:</th><td class="field-body">The input of this layer.</td>
</tr>
<tr class="field-even field"><th class="field-name">type input:</th><td class="field-body">paddle.v2.config_base.Layer</td>
</tr>
<tr class="field-odd field"><th class="field-name">param name:</th><td class="field-body">The name of this layer. It is optional.</td>
</tr>
<tr class="field-even field"><th class="field-name">type name:</th><td class="field-body">basestring</td>
</tr>
<tr class="field-odd field"><th class="field-name" colspan="2">param layer_attr:</th></tr>
<tr class="field-odd field"><td>&#160;</td><td class="field-body">The extra layer attribute. See paddle.v2.attr.ExtraAttribute
for details.</td>
</tr>
<tr class="field-even field"><th class="field-name" colspan="2">type layer_attr:</th></tr>
<tr class="field-even field"><td>&#160;</td><td class="field-body">paddle.v2.attr.ExtraAttribute</td>
<tr class="field-odd field"><th class="field-name">参数:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>paddle.v2.config_base.Layer</em>) &#8211; The input of this layer.</li>
<li><strong>name</strong> (<em>basestring</em>) &#8211; The name of this layer. It is optional.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute</em>) &#8211; The extra layer attribute. See paddle.v2.attr.ExtraAttribute
for details.</li>
</ul>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">return:</th><td class="field-body">paddle.v2.config_base.Layer object.</td>
<tr class="field-even field"><th class="field-name">返回:</th><td class="field-body"><p class="first">paddle.v2.config_base.Layer object.</p>
</td>
</tr>
<tr class="field-even field"><th class="field-name">rtype:</th><td class="field-body">paddle.v2.config_base.Layer</td>
<tr class="field-odd field"><th class="field-name">返回类型:</th><td class="field-body"><p class="first last">paddle.v2.config_base.Layer</p>
</td>
</tr>
</tbody>
</table>
</div></blockquote>
</dd></dl>
</div>
......@@ -1287,26 +1277,29 @@ be included in the configuration file to complete the input-to-hidden
mappings before lstmemory is called.</p>
<p>NOTE: This is a low level user interface. You can use network.simple_lstm
to config a simple plain lstm layer.</p>
<p>Please refer to <strong>Generating Sequences With Recurrent Neural Networks</strong> for
more details about LSTM.</p>
<p><a class="reference external" href="http://arxiv.org/abs/1308.0850">Link</a> goes as below.</p>
<dl class="docutils">
<dt>Reference:</dt>
<dd><a class="reference external" href="https://arxiv.org/pdf/1308.0850.pdf">Generating Sequences With Recurrent Neural Networks</a></dd>
</dl>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">参数:</th><td class="field-body"><ul class="first simple">
<li><strong>name</strong> (<em>basestring</em>) &#8211; The lstmemory layer name.</li>
<li><strong>size</strong> (<em>int</em>) &#8211; DEPRECATED. size of the lstm cell</li>
<li><strong>name</strong> (<em>basestring</em>) &#8211; The name of this layer. It is optional.</li>
<li><strong>size</strong> (<em>int</em>) &#8211; DEPRECATED. The dimension of the lstm cell.</li>
<li><strong>input</strong> (<em>paddle.v2.config_base.Layer</em>) &#8211; The input of this layer.</li>
<li><strong>reverse</strong> (<em>bool</em>) &#8211; is sequence process reversed or not.</li>
<li><strong>reverse</strong> (<em>bool</em>) &#8211; Whether the input sequence is processed in a reverse order.</li>
<li><strong>act</strong> (<em>paddle.v2.activation.Base</em>) &#8211; Activation type. paddle.v2.activation.Tanh is the default activation.</li>
<li><strong>gate_act</strong> (<em>paddle.v2.activation.Base</em>) &#8211; gate activation type, paddle.v2.activation.Sigmoid by default.</li>
<li><strong>state_act</strong> (<em>paddle.v2.activation.Base</em>) &#8211; state activation type, paddle.v2.activation.Tanh by default.</li>
<li><strong>gate_act</strong> (<em>paddle.v2.activation.Base</em>) &#8211; Activation type of this layer&#8217;s gates. paddle.v2.activation.Sigmoid is the
default activation.</li>
<li><strong>state_act</strong> (<em>paddle.v2.activation.Base</em>) &#8211; Activation type of the state. paddle.v2.activation.Tanh is the default activation.</li>
<li><strong>bias_attr</strong> (<em>paddle.v2.attr.ParameterAttribute | None | bool | Any</em>) &#8211; The bias attribute. If the parameter is set to False or an object
whose type is not paddle.v2.attr.ParameterAttribute, no bias is defined. If the
parameter is set to True, the bias is initialized to zero.</li>
<li><strong>param_attr</strong> (<em>paddle.v2.attr.ParameterAttribute | None | False</em>) &#8211; Parameter Attribute.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute | None</em>) &#8211; Extra Layer attribute</li>
<li><strong>param_attr</strong> (<em>paddle.v2.attr.ParameterAttribute</em>) &#8211; The parameter attribute. See paddle.v2.attr.ParameterAttribute for details.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute | None</em>) &#8211; The extra layer attribute. See paddle.v2.attr.ExtraAttribute for
details.</li>
</ul>
</td>
</tr>
......@@ -1347,12 +1340,14 @@ candidate activation <span class="math">\(\tilde{h_t}\)</span>:</p>
<div class="math">
\[h_t = (1 - z_t) h_{t-1} + z_t {\tilde{h_t}}\]</div>
<p>NOTE: In PaddlePaddle&#8217;s implementation, the multiplication operations
<span class="math">\(W_{r}x_{t}\)</span>, <span class="math">\(W_{z}x_{t}\)</span> and <span class="math">\(W x_t\)</span> are not computed in
gate_recurrent layer. Consequently, an additional mixed with
<span class="math">\(W_{r}x_{t}\)</span>, <span class="math">\(W_{z}x_{t}\)</span> and <span class="math">\(W x_t\)</span> are not performed
in gate_recurrent layer. Consequently, an additional mixed with
full_matrix_projection or a fc must be included before grumemory
is called.</p>
<p>More details can be found by referring to <a class="reference external" href="https://arxiv.org/abs/1412.3555">Empirical Evaluation of Gated
Recurrent Neural Networks on Sequence Modeling.</a></p>
<dl class="docutils">
<dt>Reference:</dt>
<dd><a class="reference external" href="https://arxiv.org/abs/1412.3555">Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling</a></dd>
</dl>
<p>The simple usage is:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">gru</span> <span class="o">=</span> <span class="n">grumemory</span><span class="p">(</span><span class="nb">input</span><span class="p">)</span>
</pre></div>
......@@ -1362,20 +1357,21 @@ Recurrent Neural Networks on Sequence Modeling.</a></p>
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">参数:</th><td class="field-body"><ul class="first simple">
<li><strong>name</strong> (<em>None | basestring</em>) &#8211; The gru layer name.</li>
<li><strong>name</strong> (<em>basestring</em>) &#8211; The name of this layer. It is optional.</li>
<li><strong>input</strong> (<em>paddle.v2.config_base.Layer.</em>) &#8211; The input of this layer.</li>
<li><strong>size</strong> (<em>int</em>) &#8211; DEPRECATED. size of the gru cell</li>
<li><strong>reverse</strong> (<em>bool</em>) &#8211; Whether sequence process is reversed or not.</li>
<li><strong>size</strong> (<em>int</em>) &#8211; DEPRECATED. The dimension of the gru cell.</li>
<li><strong>reverse</strong> (<em>bool</em>) &#8211; Whether the input sequence is processed in a reverse order.</li>
<li><strong>act</strong> (<em>paddle.v2.activation.Base</em>) &#8211; Activation type, paddle.v2.activation.Tanh is the default. This activation
affects the <span class="math">\({\tilde{h_t}}\)</span>.</li>
<li><strong>gate_act</strong> (<em>paddle.v2.activation.Base</em>) &#8211; gate activation type, paddle.v2.activation.Sigmoid by default.
This activation affects the <span class="math">\(z_t\)</span> and <span class="math">\(r_t\)</span>. It is the
<span class="math">\(\sigma\)</span> in the above formula.</li>
<li><strong>gate_act</strong> (<em>paddle.v2.activation.Base</em>) &#8211; Activation type of this layer&#8217;s two gates. paddle.v2.activation.Sigmoid is
the default activation. This activation affects the <span class="math">\(z_t\)</span>
and <span class="math">\(r_t\)</span>. It is the <span class="math">\(\sigma\)</span> in the above formula.</li>
<li><strong>bias_attr</strong> (<em>paddle.v2.attr.ParameterAttribute | None | bool | Any</em>) &#8211; The bias attribute. If the parameter is set to False or an object
whose type is not paddle.v2.attr.ParameterAttribute, no bias is defined. If the
parameter is set to True, the bias is initialized to zero.</li>
<li><strong>param_attr</strong> (<em>paddle.v2.attr.ParameterAttribute | None | False</em>) &#8211; Parameter Attribute.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute | None</em>) &#8211; Extra Layer attribute</li>
<li><strong>param_attr</strong> (<em>paddle.v2.attr.ParameterAttribute</em>) &#8211; The parameter attribute. See paddle.v2.attr.ParameterAttribute for details.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute | None</em>) &#8211; The extra layer attribute. See paddle.v2.attr.ExtraAttribute for
details.</li>
</ul>
</td>
</tr>
......@@ -2220,10 +2216,10 @@ parameter is set to True, the bias is initialized to zero.</li>
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.layer.</code><code class="descname">last_seq</code></dt>
<dd><p>Get Last Timestamp Activation of a sequence.</p>
<p>If stride &gt; 0, this layer slides a window whose size is determined by stride,
and return the last value of the window as the output. Thus, a long sequence
will be shorten. Note that for sequence with sub-sequence, the default value
of stride is -1.</p>
<p>If stride &gt; 0, this layer will slide a window whose size is determined by stride,
and return the last value of the sequence in the window as the output. Thus, a
long sequence will be shortened. Note that for sequence with sub-sequence, the
default value of stride is -1.</p>
<p>The simple usage is:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">seq</span> <span class="o">=</span> <span class="n">last_seq</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">layer</span><span class="p">)</span>
</pre></div>
......@@ -2233,11 +2229,12 @@ of stride is -1.</p>
<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>agg_level</strong> &#8211; Aggregated level</li>
<li><strong>agg_level</strong> (<em>AggregateLevel</em>) &#8211; Aggregated level</li>
<li><strong>name</strong> (<em>basestring</em>) &#8211; The name of this layer. It is optional.</li>
<li><strong>input</strong> (<em>paddle.v2.config_base.Layer</em>) &#8211; The input of this layer.</li>
<li><strong>stride</strong> (<em>Int</em>) &#8211; The step size between successive pooling regions.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute</em>) &#8211; extra layer attributes.</li>
<li><strong>stride</strong> (<em>int</em>) &#8211; The step size between successive pooling regions.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute</em>) &#8211; The extra layer attribute. See paddle.v2.attr.ExtraAttribute for
details.</li>
</ul>
</td>
</tr>
......@@ -2258,10 +2255,10 @@ of stride is -1.</p>
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.layer.</code><code class="descname">first_seq</code></dt>
<dd><p>Get First Timestamp Activation of a sequence.</p>
<p>If stride &gt; 0, this layer slides a window whose size is determined by stride,
and return the first value of the window as the output. Thus, a long sequence
will be shorten. Note that for sequence with sub-sequence, the default value
of stride is -1.</p>
<p>If stride &gt; 0, this layer will slide a window whose size is determined by stride,
and return the first value of the sequence in the window as the output. Thus, a
long sequence will be shortened. Note that for sequence with sub-sequence, the
default value of stride is -1.</p>
<p>The simple usage is:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">seq</span> <span class="o">=</span> <span class="n">first_seq</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">layer</span><span class="p">)</span>
</pre></div>
......@@ -2271,11 +2268,12 @@ of stride is -1.</p>
<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>agg_level</strong> &#8211; aggregation level</li>
<li><strong>agg_level</strong> (<em>AggregateLevel</em>) &#8211; aggregation level</li>
<li><strong>name</strong> (<em>basestring</em>) &#8211; The name of this layer. It is optional.</li>
<li><strong>input</strong> (<em>paddle.v2.config_base.Layer</em>) &#8211; The input of this layer.</li>
<li><strong>stride</strong> (<em>Int</em>) &#8211; The step size between successive pooling regions.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute</em>) &#8211; extra layer attributes.</li>
<li><strong>stride</strong> (<em>int</em>) &#8211; The step size between successive pooling regions.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute</em>) &#8211; The extra layer attribute. See paddle.v2.attr.ExtraAttribute for
details.</li>
</ul>
</td>
</tr>
......@@ -2548,8 +2546,8 @@ details.</li>
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.layer.</code><code class="descname">expand</code></dt>
<dd><p>A layer for &#8220;Expand Dense data or (sequence data where the length of each
sequence is one) to sequence data.&#8221;</p>
<dd><p>A layer for expanding dense data or (sequence data where the length of each
sequence is one) to sequence data.</p>
<p>The example usage is:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">expand</span> <span class="o">=</span> <span class="n">expand</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">layer1</span><span class="p">,</span>
<span class="n">expand_as</span><span class="o">=</span><span class="n">layer2</span><span class="p">,</span>
......@@ -2562,13 +2560,16 @@ sequence is one) to sequence data.&#8221;</p>
<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>input</strong> (<em>paddle.v2.config_base.Layer</em>) &#8211; The input of this layer.</li>
<li><strong>expand_as</strong> (<em>paddle.v2.config_base.Layer</em>) &#8211; Expand as this layer&#8217;s sequence info.</li>
<li><strong>expand_as</strong> (<em>paddle.v2.config_base.Layer</em>) &#8211; Expand the input according to this layer&#8217;s sequence infomation. And
after the operation, the input expanded will have the same number of
elememts as this layer.</li>
<li><strong>name</strong> (<em>basestring</em>) &#8211; The name of this layer. It is optional.</li>
<li><strong>bias_attr</strong> (<em>paddle.v2.attr.ParameterAttribute | None | bool | Any</em>) &#8211; The bias attribute. If the parameter is set to False or an object
whose type is not paddle.v2.attr.ParameterAttribute, no bias is defined. If the
parameter is set to True, the bias is initialized to zero.</li>
<li><strong>expand_level</strong> (<em>ExpandLevel</em>) &#8211; whether input layer is timestep(default) or sequence.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute</em>) &#8211; extra layer attributes.</li>
<li><strong>expand_level</strong> (<em>ExpandLevel</em>) &#8211; Whether the input layer is a sequence or the element of a sequence.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute</em>) &#8211; The extra layer attribute. See paddle.v2.attr.ExtraAttribute for
details.</li>
</ul>
</td>
</tr>
......@@ -3056,44 +3057,32 @@ details.</li>
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.layer.</code><code class="descname">clip</code></dt>
<dd><blockquote>
<div><p>A layer for clipping the input value by the threshold.</p>
<dd><p>A layer for clipping the input value by the threshold.</p>
<div class="math">
\[out[i] = \min\left(\max\left(in[i],p_{1}\]</div>
</div></blockquote>
<p>ight),p_{2}
ight)</p>
<blockquote>
<div><div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">clip</span> <span class="o">=</span> <span class="n">clip</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="nb">input</span><span class="p">,</span> <span class="nb">min</span><span class="o">=-</span><span class="mi">10</span><span class="p">,</span> <span class="nb">max</span><span class="o">=</span><span class="mi">10</span><span class="p">)</span>
\[out[i] = \min (\max (in[i],p_{1} ),p_{2} )\]</div>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">clip</span> <span class="o">=</span> <span class="n">clip</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="nb">input</span><span class="p">,</span> <span class="nb">min</span><span class="o">=-</span><span class="mi">10</span><span class="p">,</span> <span class="nb">max</span><span class="o">=</span><span class="mi">10</span><span class="p">)</span>
</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">param name:</th><td class="field-body">The name of this layer. It is optional.</td>
</tr>
<tr class="field-even field"><th class="field-name">type name:</th><td class="field-body">basestring</td>
</tr>
<tr class="field-odd field"><th class="field-name">param input:</th><td class="field-body">The input of this layer.</td>
</tr>
<tr class="field-even field"><th class="field-name">type input:</th><td class="field-body">paddle.v2.config_base.Layer.</td>
</tr>
<tr class="field-odd field"><th class="field-name">param min:</th><td class="field-body">The lower threshold for clipping.</td>
</tr>
<tr class="field-even field"><th class="field-name">type min:</th><td class="field-body">float</td>
</tr>
<tr class="field-odd field"><th class="field-name">param max:</th><td class="field-body">The upper threshold for clipping.</td>
</tr>
<tr class="field-even field"><th class="field-name">type max:</th><td class="field-body">float</td>
<tr class="field-odd field"><th class="field-name">参数:</th><td class="field-body"><ul class="first simple">
<li><strong>name</strong> (<em>basestring</em>) &#8211; The name of this layer. It is optional.</li>
<li><strong>input</strong> (<em>paddle.v2.config_base.Layer.</em>) &#8211; The input of this layer.</li>
<li><strong>min</strong> (<em>float</em>) &#8211; The lower threshold for clipping.</li>
<li><strong>max</strong> (<em>float</em>) &#8211; The upper threshold for clipping.</li>
</ul>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">return:</th><td class="field-body">paddle.v2.config_base.Layer object.</td>
<tr class="field-even field"><th class="field-name">返回:</th><td class="field-body"><p class="first">paddle.v2.config_base.Layer object.</p>
</td>
</tr>
<tr class="field-even field"><th class="field-name">rtype:</th><td class="field-body">paddle.v2.config_base.Layer</td>
<tr class="field-odd field"><th class="field-name">返回类型:</th><td class="field-body"><p class="first last">paddle.v2.config_base.Layer</p>
</td>
</tr>
</tbody>
</table>
</div></blockquote>
</dd></dl>
</div>
......@@ -3763,18 +3752,13 @@ details.</li>
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.layer.</code><code class="descname">huber_regression_cost</code></dt>
<dd><blockquote>
<div>In statistics, the Huber loss is a loss function used in robust regression,
<dd><p>In statistics, the Huber loss is a loss function used in robust regression,
that is less sensitive to outliers in data than the squared error loss.
Given a prediction f(x), a label y and <span class="math">\(\delta\)</span>, the loss function
is defined as:</div></blockquote>
<p>ight )^2, left | y-f(x)
ight <a href="#id2"><span class="problematic" id="id3">|</span></a>leq delta</p>
<blockquote>
<div>loss = delta left | y-f(x)</div></blockquote>
<p>ight <a href="#id4"><span class="problematic" id="id5">|</span></a>-0.5delta ^2, otherwise</p>
<blockquote>
<div><p>The example usage is:</p>
is defined as:</p>
<div class="math">
\[ \begin{align}\begin{aligned}loss = 0.5*(y-f(x))^{2}, | y-f(x) | &lt; \delta\\loss = \delta | y-f(x) | - 0.5 \delta ^2, otherwise\end{aligned}\end{align} \]</div>
<p>The example usage is:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">cost</span> <span class="o">=</span> <span class="n">huber_regression_cost</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="nb">input</span><span class="p">,</span> <span class="n">label</span><span class="o">=</span><span class="n">label</span><span class="p">)</span>
</pre></div>
</div>
......@@ -3782,41 +3766,26 @@ ight <a href="#id2"><span class="problematic" id="id3">|</span></a>leq delta</p>
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">param input:</th><td class="field-body">The first input layer.</td>
</tr>
<tr class="field-even field"><th class="field-name">type input:</th><td class="field-body">paddle.v2.config_base.Layer</td>
</tr>
<tr class="field-odd field"><th class="field-name">param label:</th><td class="field-body">The input label.</td>
</tr>
<tr class="field-even field"><th class="field-name">type input:</th><td class="field-body">paddle.v2.config_base.Layer</td>
</tr>
<tr class="field-odd field"><th class="field-name">param name:</th><td class="field-body">The name of this layer. It is optional.</td>
</tr>
<tr class="field-even field"><th class="field-name">type name:</th><td class="field-body">basestring</td>
</tr>
<tr class="field-odd field"><th class="field-name">param delta:</th><td class="field-body">The difference between the observed and predicted values.</td>
</tr>
<tr class="field-even field"><th class="field-name">type delta:</th><td class="field-body">float</td>
</tr>
<tr class="field-odd field"><th class="field-name">param coeff:</th><td class="field-body">The weight of the gradient in the back propagation.
1.0 is the default value.</td>
</tr>
<tr class="field-even field"><th class="field-name">type coeff:</th><td class="field-body">float</td>
</tr>
<tr class="field-odd field"><th class="field-name" colspan="2">param layer_attr:</th></tr>
<tr class="field-odd field"><td>&#160;</td><td class="field-body">The extra layer attribute. See paddle.v2.attr.ExtraAttribute for
details.</td>
</tr>
<tr class="field-even field"><th class="field-name" colspan="2">type layer_attr:</th></tr>
<tr class="field-even field"><td>&#160;</td><td class="field-body">paddle.v2.attr.ExtraAttribute</td>
<tr class="field-odd field"><th class="field-name">参数:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>paddle.v2.config_base.Layer</em>) &#8211; The first input layer.</li>
<li><strong>label</strong> &#8211; The input label.</li>
<li><strong>name</strong> (<em>basestring</em>) &#8211; The name of this layer. It is optional.</li>
<li><strong>delta</strong> (<em>float</em>) &#8211; The difference between the observed and predicted values.</li>
<li><strong>coeff</strong> (<em>float</em>) &#8211; The weight of the gradient in the back propagation.
1.0 is the default value.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute</em>) &#8211; The extra layer attribute. See paddle.v2.attr.ExtraAttribute for
details.</li>
</ul>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">return:</th><td class="field-body">paddle.v2.config_base.Layer object.</td>
<tr class="field-even field"><th class="field-name">返回:</th><td class="field-body"><p class="first">paddle.v2.config_base.Layer object.</p>
</td>
</tr>
<tr class="field-even field"><th class="field-name">rtype:</th><td class="field-body">paddle.v2.config_base.Layer.</td>
<tr class="field-odd field"><th class="field-name">返回类型:</th><td class="field-body"><p class="first last">paddle.v2.config_base.Layer.</p>
</td>
</tr>
</tbody>
</table>
</div></blockquote>
</dd></dl>
</div>
......@@ -3825,56 +3794,37 @@ details.</td>
<dl class="class">
<dt>
<em class="property">class </em><code class="descclassname">paddle.v2.layer.</code><code class="descname">huber_classification_cost</code></dt>
<dd><blockquote>
<div>For classification purposes, a variant of the Huber loss called modified Huber
<dd><p>For classification purposes, a variant of the Huber loss called modified Huber
is sometimes used. Given a prediction f(x) (a real-valued classifier score) and
a true binary class label :math:<a href="#id6"><span class="problematic" id="id7">`</span></a>yin left {-1, 1</div></blockquote>
<dl class="docutils">
<dt>ight }`, the modified Huber</dt>
<dd>loss is defined as:</dd>
<dt>ight )^2, yf(x)geq 1</dt>
<dd><blockquote class="first">
<div>loss = -4yf(x), ext{otherwise}</div></blockquote>
a true binary class label <span class="math">\(y\in \{-1, 1 \}\)</span>, the modified Huber
loss is defined as:</p>
<p>The example usage is:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">cost</span> <span class="o">=</span> <span class="n">huber_classification_cost</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="nb">input</span><span class="p">,</span> <span class="n">label</span><span class="o">=</span><span class="n">label</span><span class="p">)</span>
</pre></div>
</div>
<table class="last docutils field-list" frame="void" rules="none">
<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">param input:</th><td class="field-body">The first input layer.</td>
</tr>
<tr class="field-even field"><th class="field-name">type input:</th><td class="field-body">paddle.v2.config_base.Layer</td>
</tr>
<tr class="field-odd field"><th class="field-name">param label:</th><td class="field-body">The input label.</td>
</tr>
<tr class="field-even field"><th class="field-name">type input:</th><td class="field-body">paddle.v2.config_base.Layer</td>
</tr>
<tr class="field-odd field"><th class="field-name">param name:</th><td class="field-body">The name of this layer. It is optional.</td>
</tr>
<tr class="field-even field"><th class="field-name">type name:</th><td class="field-body">basestring</td>
</tr>
<tr class="field-odd field"><th class="field-name">param coeff:</th><td class="field-body">The weight of the gradient in the back propagation.
1.0 is the default value.</td>
</tr>
<tr class="field-even field"><th class="field-name">type coeff:</th><td class="field-body">float</td>
</tr>
<tr class="field-odd field"><th class="field-name" colspan="2">param layer_attr:</th></tr>
<tr class="field-odd field"><td>&#160;</td><td class="field-body">The extra layer attribute. See paddle.v2.attr.ExtraAttribute for
details.</td>
</tr>
<tr class="field-even field"><th class="field-name" colspan="2">type layer_attr:</th></tr>
<tr class="field-even field"><td>&#160;</td><td class="field-body">paddle.v2.attr.ExtraAttribute</td>
<tr class="field-odd field"><th class="field-name">参数:</th><td class="field-body"><ul class="first simple">
<li><strong>input</strong> (<em>paddle.v2.config_base.Layer</em>) &#8211; The first input layer.</li>
<li><strong>label</strong> &#8211; The input label.</li>
<li><strong>name</strong> (<em>basestring</em>) &#8211; The name of this layer. It is optional.</li>
<li><strong>coeff</strong> (<em>float</em>) &#8211; The weight of the gradient in the back propagation.
1.0 is the default value.</li>
<li><strong>layer_attr</strong> (<em>paddle.v2.attr.ExtraAttribute</em>) &#8211; The extra layer attribute. See paddle.v2.attr.ExtraAttribute for
details.</li>
</ul>
</td>
</tr>
<tr class="field-odd field"><th class="field-name">return:</th><td class="field-body">paddle.v2.config_base.Layer object.</td>
<tr class="field-even field"><th class="field-name">返回:</th><td class="field-body"><p class="first">paddle.v2.config_base.Layer object.</p>
</td>
</tr>
<tr class="field-even field"><th class="field-name">rtype:</th><td class="field-body">paddle.v2.config_base.Layer</td>
<tr class="field-odd field"><th class="field-name">返回类型:</th><td class="field-body"><p class="first last">paddle.v2.config_base.Layer</p>
</td>
</tr>
</tbody>
</table>
</dd>
</dl>
</dd></dl>
</div>
......@@ -4580,7 +4530,7 @@ details.</li>
<dd><p>The gated unit layer implements a simple gating mechanism over the input.
The input <span class="math">\(X\)</span> is first projected into a new space <span class="math">\(X'\)</span>, and
it is also used to produce a gate weight <span class="math">\(\sigma\)</span>. Element-wise
product between <a href="#id11"><span class="problematic" id="id12">:match:`X&#8217;`</span></a> and <span class="math">\(\sigma\)</span> is finally returned.</p>
product between <a href="#id5"><span class="problematic" id="id6">:match:`X&#8217;`</span></a> and <span class="math">\(\sigma\)</span> is finally returned.</p>
<dl class="docutils">
<dt>Reference:</dt>
<dd><a class="reference external" href="https://arxiv.org/abs/1612.08083">Language Modeling with Gated Convolutional Networks</a></dd>
......
develop/doc_cn/searchindex.js
浏览文件 @ 8bcf7dab
因为 它太大了无法显示 source diff 。你可以改为 查看blob。
Markdown is supported
0% 或 .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑！
想要评论请 注册 或