Skip to content

P
Paddle

PaddlePaddle / Paddle
大约 1 年 前同步成功

通知 2299
Star 20931
Fork 5422
P
Paddle
提交 714155e1 编写于 作者: T Travis CI
浏览文件
操作

Deploy to GitHub Pages: 438ed128

上级 399f84c0
展开全部 显示空白变更内容
内联 并排
Showing with 38 addition and 58 deletion
develop/doc/api/v2/config/layer.html
浏览文件 @ 714155e1
...@@ -833,8 +833,7 @@ as pool_type to identify the mode.</td> ...@@ -833,8 +833,7 @@ as pool_type to identify the mode.</td>
<dd><p>A layer performs spatial pyramid pooling.</p> <dd><p>A layer performs spatial pyramid pooling.</p>
<dl class="docutils"> <dl class="docutils">
<dt>Reference:</dt> <dt>Reference:</dt>
<dd><a href="#id12"><span class="problematic" id="id13">`Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition <dd><a class="reference external" href="https://arxiv.org/abs/1406.4729">Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition</a></dd>
https://arxiv.org/abs/1406.4729`_</span></a></dd>
</dl> </dl>
<p>The example usage is:</p> <p>The example usage is:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">spp</span> <span class="o">=</span> <span class="n">spp</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">data</span><span class="p">,</span> <div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">spp</span> <span class="o">=</span> <span class="n">spp</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">data</span><span class="p">,</span>
...@@ -890,10 +889,8 @@ details.</li> ...@@ -890,10 +889,8 @@ details.</li>
to be devided by groups.</p> to be devided by groups.</p>
<dl class="docutils"> <dl class="docutils">
<dt>Reference:</dt> <dt>Reference:</dt>
<dd><a href="#id14"><span class="problematic" id="id15">`Maxout Networks <dd><a class="reference external" href="http://www.jmlr.org/proceedings/papers/v28/goodfellow13.pdf">Maxout Networks</a>
http://www.jmlr.org/proceedings/papers/v28/goodfellow13.pdf`_</span></a> <a class="reference external" href="https://arxiv.org/pdf/1312.6082v4.pdf">Multi-digit Number Recognition from Street View Imagery using Deep Convolutional Neural Networks</a></dd>
<a href="#id16"><span class="problematic" id="id17">`Multi-digit Number Recognition from Street View Imagery using Deep Convolutional Neural Networks
https://arxiv.org/pdf/1312.6082v4.pdf`_</span></a></dd>
</dl> </dl>
<div class="math"> <div class="math">
\[\begin{split}out = \max_k (in[n, k, o_c , s]) \\ \[\begin{split}out = \max_k (in[n, k, o_c , s]) \\
...@@ -997,8 +994,7 @@ feature map.</p> ...@@ -997,8 +994,7 @@ feature map.</p>
<dd><p>Response normalization across feature maps.</p> <dd><p>Response normalization across feature maps.</p>
<dl class="docutils"> <dl class="docutils">
<dt>Reference:</dt> <dt>Reference:</dt>
<dd><a href="#id18"><span class="problematic" id="id19">`ImageNet Classification with Deep Convolutional Neural Networks <dd><a class="reference external" href="http://www.cs.toronto.edu/~fritz/absps/imagenet.pdf">ImageNet Classification with Deep Convolutional Neural Networks</a></dd>
http://www.cs.toronto.edu/~fritz/absps/imagenet.pdf`_</span></a></dd>
</dl> </dl>
<p>The example usage is:</p> <p>The example usage is:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">norm</span> <span class="o">=</span> <span class="n">img_cmrnorm</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">net</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="mi">5</span><span class="p">)</span> <div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">norm</span> <span class="o">=</span> <span class="n">img_cmrnorm</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">net</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="mi">5</span><span class="p">)</span>
...@@ -1050,9 +1046,8 @@ details.</li> ...@@ -1050,9 +1046,8 @@ details.</li>
y_i &amp;\gets \gamma \hat{x_i} + \beta \qquad &amp;//\ scale\ and\ shift\end{split}\]</div> y_i &amp;\gets \gamma \hat{x_i} + \beta \qquad &amp;//\ scale\ and\ shift\end{split}\]</div>
<dl class="docutils"> <dl class="docutils">
<dt>Reference:</dt> <dt>Reference:</dt>
<dd><a href="#id20"><span class="problematic" id="id21">`Batch Normalization: Accelerating Deep Network Training by Reducing <dd><a class="reference external" href="http://arxiv.org/abs/1502.03167">Batch Normalization: Accelerating Deep Network Training by Reducing
Internal Covariate Shift Internal Covariate Shift</a></dd>
http://arxiv.org/abs/1502.03167`_</span></a></dd>
</dl> </dl>
<p>The example usage is:</p> <p>The example usage is:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">norm</span> <span class="o">=</span> <span class="n">batch_norm</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">net</span><span class="p">,</span> <span class="n">act</span><span class="o">=</span><span class="n">paddle</span><span class="o">.</span><span class="n">v2</span><span class="o">.</span><span class="n">activation</span><span class="o">.</span><span class="n">Relu</span><span class="p">())</span> <div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">norm</span> <span class="o">=</span> <span class="n">batch_norm</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">net</span><span class="p">,</span> <span class="n">act</span><span class="o">=</span><span class="n">paddle</span><span class="o">.</span><span class="n">v2</span><span class="o">.</span><span class="n">activation</span><span class="o">.</span><span class="n">Relu</span><span class="p">())</span>
...@@ -3972,8 +3967,7 @@ details.</li> ...@@ -3972,8 +3967,7 @@ details.</li>
<dd><p>A cost Layer for learning to rank using gradient descent.</p> <dd><p>A cost Layer for learning to rank using gradient descent.</p>
<dl class="docutils"> <dl class="docutils">
<dt>Reference:</dt> <dt>Reference:</dt>
<dd><a href="#id22"><span class="problematic" id="id23">`Learning to Rank using Gradient Descent <dd><a class="reference external" href="http://research.microsoft.com/en-us/um/people/cburges/papers/ICML_ranking.pdf">Learning to Rank using Gradient Descent</a></dd>
http://research.microsoft.com/en-us/um/people/cburges/papers/ICML_ranking.pdf`_</span></a></dd>
</dl> </dl>
<div class="math"> <div class="math">
\[ \begin{align}\begin{aligned}C_{i,j} &amp; = -\tilde{P_{ij}} * o_{i,j} + log(1 + e^{o_{i,j}})\\o_{i,j} &amp; = o_i - o_j\\\tilde{P_{i,j}} &amp; = \{0, 0.5, 1\} \ or \ \{0, 1\}\end{aligned}\end{align} \]</div> \[ \begin{align}\begin{aligned}C_{i,j} &amp; = -\tilde{P_{ij}} * o_{i,j} + log(1 + e^{o_{i,j}})\\o_{i,j} &amp; = o_i - o_j\\\tilde{P_{i,j}} &amp; = \{0, 0.5, 1\} \ or \ \{0, 1\}\end{aligned}\end{align} \]</div>
...@@ -4152,9 +4146,8 @@ classication task. e.g. sequence labeling problems where the ...@@ -4152,9 +4146,8 @@ classication task. e.g. sequence labeling problems where the
alignment between the inputs and the target labels is unknown.</p> alignment between the inputs and the target labels is unknown.</p>
<dl class="docutils"> <dl class="docutils">
<dt>Reference:</dt> <dt>Reference:</dt>
<dd><a href="#id24"><span class="problematic" id="id25">`Connectionist Temporal Classification: Labelling Unsegmented Sequence Data <dd><a class="reference external" href="http://machinelearning.wustl.edu/mlpapers/paper_files/icml2006_GravesFGS06.pdf">Connectionist Temporal Classification: Labelling Unsegmented Sequence Data
with Recurrent Neural Networks with Recurrent Neural Networks</a></dd>
http://machinelearning.wustl.edu/mlpapers/paper_files/icml2006_GravesFGS06.pdf`_</span></a></dd>
</dl> </dl>
<div class="admonition note"> <div class="admonition note">
<p class="first admonition-title">Note</p> <p class="first admonition-title">Note</p>
...@@ -4210,9 +4203,8 @@ building process, PaddlePaddle will clone the source codes, build and ...@@ -4210,9 +4203,8 @@ building process, PaddlePaddle will clone the source codes, build and
install it to <code class="code docutils literal"><span class="pre">third_party/install/warpctc</span></code> directory.</p> install it to <code class="code docutils literal"><span class="pre">third_party/install/warpctc</span></code> directory.</p>
<dl class="docutils"> <dl class="docutils">
<dt>Reference:</dt> <dt>Reference:</dt>
<dd><a href="#id26"><span class="problematic" id="id27">`Connectionist Temporal Classification: Labelling Unsegmented Sequence Data <dd><a class="reference external" href="http://machinelearning.wustl.edu/mlpapers/paper_files/icml2006_GravesFGS06.pdf">Connectionist Temporal Classification: Labelling Unsegmented Sequence Data
with Recurrent Neural Networks with Recurrent Neural Networks</a></dd>
http://machinelearning.wustl.edu/mlpapers/paper_files/icml2006_GravesFGS06.pdf`_</span></a></dd>
</dl> </dl>
<div class="admonition note"> <div class="admonition note">
<p class="first admonition-title">Note</p> <p class="first admonition-title">Note</p>
...@@ -4269,8 +4261,8 @@ details.</li> ...@@ -4269,8 +4261,8 @@ details.</li>
<dd><p>Noise-contrastive estimation.</p> <dd><p>Noise-contrastive estimation.</p>
<dl class="docutils"> <dl class="docutils">
<dt>Reference:</dt> <dt>Reference:</dt>
<dd><a href="#id28"><span class="problematic" id="id29">`A fast and simple algorithm for training neural probabilistic language <dd><a class="reference external" href="https://www.cs.toronto.edu/~amnih/papers/ncelm.pdf">A fast and simple algorithm for training neural probabilistic language
models. https://www.cs.toronto.edu/~amnih/papers/ncelm.pdf`_</span></a></dd> models.</a></dd>
</dl> </dl>
<p>The example usage is:</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">nce</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="p">[</span><span class="n">layer1</span><span class="p">,</span> <span class="n">layer2</span><span class="p">],</span> <span class="n">label</span><span class="o">=</span><span class="n">layer2</span><span class="p">,</span> <div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">cost</span> <span class="o">=</span> <span class="n">nce</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="p">[</span><span class="n">layer1</span><span class="p">,</span> <span class="n">layer2</span><span class="p">],</span> <span class="n">label</span><span class="o">=</span><span class="n">layer2</span><span class="p">,</span>
...@@ -4381,8 +4373,7 @@ sizes of input and label are equal. The formula is as follows,</p> ...@@ -4381,8 +4373,7 @@ sizes of input and label are equal. The formula is as follows,</p>
\[\begin{split}smooth_{L1}(x) = \begin{cases} 0.5x^2&amp; \text{if} \ |x| &lt; 1 \\ |x|-0.5&amp; \text{otherwise} \end{cases}\end{split}\]</div> \[\begin{split}smooth_{L1}(x) = \begin{cases} 0.5x^2&amp; \text{if} \ |x| &lt; 1 \\ |x|-0.5&amp; \text{otherwise} \end{cases}\end{split}\]</div>
<dl class="docutils"> <dl class="docutils">
<dt>Reference:</dt> <dt>Reference:</dt>
<dd><a href="#id30"><span class="problematic" id="id31">`Fast R-CNN <dd><a class="reference external" href="https://arxiv.org/pdf/1504.08083v2.pdf">Fast R-CNN</a></dd>
https://arxiv.org/pdf/1504.08083v2.pdf`_</span></a></dd>
</dl> </dl>
<p>The example usage is:</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">smooth_l1_cost</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="nb">input</span><span class="p">,</span> <div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">cost</span> <span class="o">=</span> <span class="n">smooth_l1_cost</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="nb">input</span><span class="p">,</span>
...@@ -4532,8 +4523,8 @@ details.</li> ...@@ -4532,8 +4523,8 @@ details.</li>
<dd><p>The Parametric Relu activation that actives outputs with a learnable weight.</p> <dd><p>The Parametric Relu activation that actives outputs with a learnable weight.</p>
<dl class="docutils"> <dl class="docutils">
<dt>Reference:</dt> <dt>Reference:</dt>
<dd><a href="#id32"><span class="problematic" id="id33">`Delving Deep into Rectifiers: Surpassing Human-Level Performance on <dd><a class="reference external" href="http://arxiv.org/pdf/1502.01852v1.pdf">Delving Deep into Rectifiers: Surpassing Human-Level Performance on
ImageNet Classification http://arxiv.org/pdf/1502.01852v1.pdf`_</span></a></dd> ImageNet Classification</a></dd>
</dl> </dl>
<div class="math"> <div class="math">
\[\begin{split}z_i &amp;\quad if \quad z_i &gt; 0 \\ \[\begin{split}z_i &amp;\quad if \quad z_i &gt; 0 \\
...@@ -4588,11 +4579,10 @@ details.</li> ...@@ -4588,11 +4579,10 @@ details.</li>
<dd><p>The gated unit layer implements a simple gating mechanism over the input. <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 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 it is also used to produce a gate weight <span class="math">\(\sigma\)</span>. Element-wise
product between <a href="#id10"><span class="problematic" id="id11">:match:`X&#8217;`</span></a> and <span class="math">\(\sigma\)</span> is finally returned.</p> 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>
<dl class="docutils"> <dl class="docutils">
<dt>Reference:</dt> <dt>Reference:</dt>
<dd><a href="#id34"><span class="problematic" id="id35">`Language Modeling with Gated Convolutional Networks <dd><a class="reference external" href="https://arxiv.org/abs/1612.08083">Language Modeling with Gated Convolutional Networks</a></dd>
https://arxiv.org/abs/1612.08083`_</span></a></dd>
</dl> </dl>
<div class="math"> <div class="math">
\[y=\text{act}(X \cdot W + b)\otimes \sigma(X \cdot V + c)\]</div> \[y=\text{act}(X \cdot W + b)\otimes \sigma(X \cdot V + c)\]</div>
......
develop/doc/searchindex.js
浏览文件 @ 714155e1
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 @ 714155e1
...@@ -834,8 +834,7 @@ as pool_type to identify the mode.</td> ...@@ -834,8 +834,7 @@ as pool_type to identify the mode.</td>
<dd><p>A layer performs spatial pyramid pooling.</p> <dd><p>A layer performs spatial pyramid pooling.</p>
<dl class="docutils"> <dl class="docutils">
<dt>Reference:</dt> <dt>Reference:</dt>
<dd><a href="#id12"><span class="problematic" id="id13">`Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition <dd><a class="reference external" href="https://arxiv.org/abs/1406.4729">Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition</a></dd>
https://arxiv.org/abs/1406.4729`_</span></a></dd>
</dl> </dl>
<p>The example usage is:</p> <p>The example usage is:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">spp</span> <span class="o">=</span> <span class="n">spp</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">data</span><span class="p">,</span> <div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">spp</span> <span class="o">=</span> <span class="n">spp</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">data</span><span class="p">,</span>
...@@ -891,10 +890,8 @@ details.</li> ...@@ -891,10 +890,8 @@ details.</li>
to be devided by groups.</p> to be devided by groups.</p>
<dl class="docutils"> <dl class="docutils">
<dt>Reference:</dt> <dt>Reference:</dt>
<dd><a href="#id14"><span class="problematic" id="id15">`Maxout Networks <dd><a class="reference external" href="http://www.jmlr.org/proceedings/papers/v28/goodfellow13.pdf">Maxout Networks</a>
http://www.jmlr.org/proceedings/papers/v28/goodfellow13.pdf`_</span></a> <a class="reference external" href="https://arxiv.org/pdf/1312.6082v4.pdf">Multi-digit Number Recognition from Street View Imagery using Deep Convolutional Neural Networks</a></dd>
<a href="#id16"><span class="problematic" id="id17">`Multi-digit Number Recognition from Street View Imagery using Deep Convolutional Neural Networks
https://arxiv.org/pdf/1312.6082v4.pdf`_</span></a></dd>
</dl> </dl>
<div class="math"> <div class="math">
\[\begin{split}out = \max_k (in[n, k, o_c , s]) \\ \[\begin{split}out = \max_k (in[n, k, o_c , s]) \\
...@@ -998,8 +995,7 @@ feature map.</p> ...@@ -998,8 +995,7 @@ feature map.</p>
<dd><p>Response normalization across feature maps.</p> <dd><p>Response normalization across feature maps.</p>
<dl class="docutils"> <dl class="docutils">
<dt>Reference:</dt> <dt>Reference:</dt>
<dd><a href="#id18"><span class="problematic" id="id19">`ImageNet Classification with Deep Convolutional Neural Networks <dd><a class="reference external" href="http://www.cs.toronto.edu/~fritz/absps/imagenet.pdf">ImageNet Classification with Deep Convolutional Neural Networks</a></dd>
http://www.cs.toronto.edu/~fritz/absps/imagenet.pdf`_</span></a></dd>
</dl> </dl>
<p>The example usage is:</p> <p>The example usage is:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">norm</span> <span class="o">=</span> <span class="n">img_cmrnorm</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">net</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="mi">5</span><span class="p">)</span> <div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">norm</span> <span class="o">=</span> <span class="n">img_cmrnorm</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">net</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="mi">5</span><span class="p">)</span>
...@@ -1051,9 +1047,8 @@ details.</li> ...@@ -1051,9 +1047,8 @@ details.</li>
y_i &amp;\gets \gamma \hat{x_i} + \beta \qquad &amp;//\ scale\ and\ shift\end{split}\]</div> y_i &amp;\gets \gamma \hat{x_i} + \beta \qquad &amp;//\ scale\ and\ shift\end{split}\]</div>
<dl class="docutils"> <dl class="docutils">
<dt>Reference:</dt> <dt>Reference:</dt>
<dd><a href="#id20"><span class="problematic" id="id21">`Batch Normalization: Accelerating Deep Network Training by Reducing <dd><a class="reference external" href="http://arxiv.org/abs/1502.03167">Batch Normalization: Accelerating Deep Network Training by Reducing
Internal Covariate Shift Internal Covariate Shift</a></dd>
http://arxiv.org/abs/1502.03167`_</span></a></dd>
</dl> </dl>
<p>The example usage is:</p> <p>The example usage is:</p>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">norm</span> <span class="o">=</span> <span class="n">batch_norm</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">net</span><span class="p">,</span> <span class="n">act</span><span class="o">=</span><span class="n">paddle</span><span class="o">.</span><span class="n">v2</span><span class="o">.</span><span class="n">activation</span><span class="o">.</span><span class="n">Relu</span><span class="p">())</span> <div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">norm</span> <span class="o">=</span> <span class="n">batch_norm</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">net</span><span class="p">,</span> <span class="n">act</span><span class="o">=</span><span class="n">paddle</span><span class="o">.</span><span class="n">v2</span><span class="o">.</span><span class="n">activation</span><span class="o">.</span><span class="n">Relu</span><span class="p">())</span>
...@@ -3973,8 +3968,7 @@ details.</li> ...@@ -3973,8 +3968,7 @@ details.</li>
<dd><p>A cost Layer for learning to rank using gradient descent.</p> <dd><p>A cost Layer for learning to rank using gradient descent.</p>
<dl class="docutils"> <dl class="docutils">
<dt>Reference:</dt> <dt>Reference:</dt>
<dd><a href="#id22"><span class="problematic" id="id23">`Learning to Rank using Gradient Descent <dd><a class="reference external" href="http://research.microsoft.com/en-us/um/people/cburges/papers/ICML_ranking.pdf">Learning to Rank using Gradient Descent</a></dd>
http://research.microsoft.com/en-us/um/people/cburges/papers/ICML_ranking.pdf`_</span></a></dd>
</dl> </dl>
<div class="math"> <div class="math">
\[ \begin{align}\begin{aligned}C_{i,j} &amp; = -\tilde{P_{ij}} * o_{i,j} + log(1 + e^{o_{i,j}})\\o_{i,j} &amp; = o_i - o_j\\\tilde{P_{i,j}} &amp; = \{0, 0.5, 1\} \ or \ \{0, 1\}\end{aligned}\end{align} \]</div> \[ \begin{align}\begin{aligned}C_{i,j} &amp; = -\tilde{P_{ij}} * o_{i,j} + log(1 + e^{o_{i,j}})\\o_{i,j} &amp; = o_i - o_j\\\tilde{P_{i,j}} &amp; = \{0, 0.5, 1\} \ or \ \{0, 1\}\end{aligned}\end{align} \]</div>
...@@ -4153,9 +4147,8 @@ classication task. e.g. sequence labeling problems where the ...@@ -4153,9 +4147,8 @@ classication task. e.g. sequence labeling problems where the
alignment between the inputs and the target labels is unknown.</p> alignment between the inputs and the target labels is unknown.</p>
<dl class="docutils"> <dl class="docutils">
<dt>Reference:</dt> <dt>Reference:</dt>
<dd><a href="#id24"><span class="problematic" id="id25">`Connectionist Temporal Classification: Labelling Unsegmented Sequence Data <dd><a class="reference external" href="http://machinelearning.wustl.edu/mlpapers/paper_files/icml2006_GravesFGS06.pdf">Connectionist Temporal Classification: Labelling Unsegmented Sequence Data
with Recurrent Neural Networks with Recurrent Neural Networks</a></dd>
http://machinelearning.wustl.edu/mlpapers/paper_files/icml2006_GravesFGS06.pdf`_</span></a></dd>
</dl> </dl>
<div class="admonition note"> <div class="admonition note">
<p class="first admonition-title">注解</p> <p class="first admonition-title">注解</p>
...@@ -4211,9 +4204,8 @@ building process, PaddlePaddle will clone the source codes, build and ...@@ -4211,9 +4204,8 @@ building process, PaddlePaddle will clone the source codes, build and
install it to <code class="code docutils literal"><span class="pre">third_party/install/warpctc</span></code> directory.</p> install it to <code class="code docutils literal"><span class="pre">third_party/install/warpctc</span></code> directory.</p>
<dl class="docutils"> <dl class="docutils">
<dt>Reference:</dt> <dt>Reference:</dt>
<dd><a href="#id26"><span class="problematic" id="id27">`Connectionist Temporal Classification: Labelling Unsegmented Sequence Data <dd><a class="reference external" href="http://machinelearning.wustl.edu/mlpapers/paper_files/icml2006_GravesFGS06.pdf">Connectionist Temporal Classification: Labelling Unsegmented Sequence Data
with Recurrent Neural Networks with Recurrent Neural Networks</a></dd>
http://machinelearning.wustl.edu/mlpapers/paper_files/icml2006_GravesFGS06.pdf`_</span></a></dd>
</dl> </dl>
<div class="admonition note"> <div class="admonition note">
<p class="first admonition-title">注解</p> <p class="first admonition-title">注解</p>
...@@ -4270,8 +4262,8 @@ details.</li> ...@@ -4270,8 +4262,8 @@ details.</li>
<dd><p>Noise-contrastive estimation.</p> <dd><p>Noise-contrastive estimation.</p>
<dl class="docutils"> <dl class="docutils">
<dt>Reference:</dt> <dt>Reference:</dt>
<dd><a href="#id28"><span class="problematic" id="id29">`A fast and simple algorithm for training neural probabilistic language <dd><a class="reference external" href="https://www.cs.toronto.edu/~amnih/papers/ncelm.pdf">A fast and simple algorithm for training neural probabilistic language
models. https://www.cs.toronto.edu/~amnih/papers/ncelm.pdf`_</span></a></dd> models.</a></dd>
</dl> </dl>
<p>The example usage is:</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">nce</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="p">[</span><span class="n">layer1</span><span class="p">,</span> <span class="n">layer2</span><span class="p">],</span> <span class="n">label</span><span class="o">=</span><span class="n">layer2</span><span class="p">,</span> <div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">cost</span> <span class="o">=</span> <span class="n">nce</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="p">[</span><span class="n">layer1</span><span class="p">,</span> <span class="n">layer2</span><span class="p">],</span> <span class="n">label</span><span class="o">=</span><span class="n">layer2</span><span class="p">,</span>
...@@ -4382,8 +4374,7 @@ sizes of input and label are equal. The formula is as follows,</p> ...@@ -4382,8 +4374,7 @@ sizes of input and label are equal. The formula is as follows,</p>
\[\begin{split}smooth_{L1}(x) = \begin{cases} 0.5x^2&amp; \text{if} \ |x| &lt; 1 \\ |x|-0.5&amp; \text{otherwise} \end{cases}\end{split}\]</div> \[\begin{split}smooth_{L1}(x) = \begin{cases} 0.5x^2&amp; \text{if} \ |x| &lt; 1 \\ |x|-0.5&amp; \text{otherwise} \end{cases}\end{split}\]</div>
<dl class="docutils"> <dl class="docutils">
<dt>Reference:</dt> <dt>Reference:</dt>
<dd><a href="#id30"><span class="problematic" id="id31">`Fast R-CNN <dd><a class="reference external" href="https://arxiv.org/pdf/1504.08083v2.pdf">Fast R-CNN</a></dd>
https://arxiv.org/pdf/1504.08083v2.pdf`_</span></a></dd>
</dl> </dl>
<p>The example usage is:</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">smooth_l1_cost</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="nb">input</span><span class="p">,</span> <div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">cost</span> <span class="o">=</span> <span class="n">smooth_l1_cost</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="nb">input</span><span class="p">,</span>
...@@ -4533,8 +4524,8 @@ details.</li> ...@@ -4533,8 +4524,8 @@ details.</li>
<dd><p>The Parametric Relu activation that actives outputs with a learnable weight.</p> <dd><p>The Parametric Relu activation that actives outputs with a learnable weight.</p>
<dl class="docutils"> <dl class="docutils">
<dt>Reference:</dt> <dt>Reference:</dt>
<dd><a href="#id32"><span class="problematic" id="id33">`Delving Deep into Rectifiers: Surpassing Human-Level Performance on <dd><a class="reference external" href="http://arxiv.org/pdf/1502.01852v1.pdf">Delving Deep into Rectifiers: Surpassing Human-Level Performance on
ImageNet Classification http://arxiv.org/pdf/1502.01852v1.pdf`_</span></a></dd> ImageNet Classification</a></dd>
</dl> </dl>
<div class="math"> <div class="math">
\[\begin{split}z_i &amp;\quad if \quad z_i &gt; 0 \\ \[\begin{split}z_i &amp;\quad if \quad z_i &gt; 0 \\
...@@ -4589,11 +4580,10 @@ details.</li> ...@@ -4589,11 +4580,10 @@ details.</li>
<dd><p>The gated unit layer implements a simple gating mechanism over the input. <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 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 it is also used to produce a gate weight <span class="math">\(\sigma\)</span>. Element-wise
product between <a href="#id10"><span class="problematic" id="id11">:match:`X&#8217;`</span></a> and <span class="math">\(\sigma\)</span> is finally returned.</p> 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>
<dl class="docutils"> <dl class="docutils">
<dt>Reference:</dt> <dt>Reference:</dt>
<dd><a href="#id34"><span class="problematic" id="id35">`Language Modeling with Gated Convolutional Networks <dd><a class="reference external" href="https://arxiv.org/abs/1612.08083">Language Modeling with Gated Convolutional Networks</a></dd>
https://arxiv.org/abs/1612.08083`_</span></a></dd>
</dl> </dl>
<div class="math"> <div class="math">
\[y=\text{act}(X \cdot W + b)\otimes \sigma(X \cdot V + c)\]</div> \[y=\text{act}(X \cdot W + b)\otimes \sigma(X \cdot V + c)\]</div>
......
develop/doc_cn/searchindex.js
浏览文件 @ 714155e1
此差异已折叠。
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