Google Git
Sign in
apache / mxnet-site / refs/heads/test-revert / . / _modules / mxnet / rnn / rnn_cell.html
blob: 32d7e568c939473bbe42fe0cdb265095f1389130 [file] [log] [blame]
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="utf-8"/>
<meta content="IE=edge" http-equiv="X-UA-Compatible"/>
<meta content="width=device-width, initial-scale=1" name="viewport"/>
<title>mxnet.rnn.rnn_cell — mxnet documentation</title>
<link crossorigin="anonymous" href="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.6/css/bootstrap.min.css" integrity="sha384-1q8mTJOASx8j1Au+a5WDVnPi2lkFfwwEAa8hDDdjZlpLegxhjVME1fgjWPGmkzs7" rel="stylesheet"/>
<link href="https://maxcdn.bootstrapcdn.com/font-awesome/4.5.0/css/font-awesome.min.css" rel="stylesheet"/>
<link href="../../../_static/basic.css" rel="stylesheet" type="text/css">
<link href="../../../_static/pygments.css" rel="stylesheet" type="text/css">
<link href="../../../_static/mxnet.css" rel="stylesheet" type="text/css"/>
<script type="text/javascript">
var DOCUMENTATION_OPTIONS = {
URL_ROOT: '../../../',
VERSION: '',
COLLAPSE_INDEX: false,
FILE_SUFFIX: '.html',
HAS_SOURCE: true,
SOURCELINK_SUFFIX: ''
};
</script>
<script src="https://code.jquery.com/jquery-1.11.1.min.js" type="text/javascript"></script>
<script src="../../../_static/underscore.js" type="text/javascript"></script>
<script src="../../../_static/searchtools_custom.js" type="text/javascript"></script>
<script src="../../../_static/doctools.js" type="text/javascript"></script>
<script src="../../../_static/selectlang.js" type="text/javascript"></script>
<script src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.1/MathJax.js?config=TeX-AMS-MML_HTMLorMML" type="text/javascript"></script>
<script type="text/javascript"> jQuery(function() { Search.loadIndex("/searchindex.js"); Search.init();}); </script>
<script>
(function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){
(i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new
Date();a=s.createElement(o),
m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m)
})(window,document,'script','https://www.google-analytics.com/analytics.js','ga');
ga('create', 'UA-96378503-1', 'auto');
ga('send', 'pageview');
</script>
<!-- -->
<!-- <script type="text/javascript" src="../../../_static/jquery.js"></script> -->
<!-- -->
<!-- <script type="text/javascript" src="../../../_static/underscore.js"></script> -->
<!-- -->
<!-- <script type="text/javascript" src="../../../_static/doctools.js"></script> -->
<!-- -->
<!-- <script type="text/javascript" src="https://cdn.mathjax.org/mathjax/latest/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script> -->
<!-- -->
<link href="../../index.html" rel="up" title="Module code">
<link href="https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/mxnet-icon.png" rel="icon" type="image/png"/>
</link></link></link></head>
<body background="https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/mxnet-background-compressed.jpeg" role="document">
<div class="content-block"><div class="navbar navbar-fixed-top">
<div class="container" id="navContainer">
<div class="innder" id="header-inner">
<h1 id="logo-wrap">
<a href="../../../" id="logo"><img src="https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/mxnet_logo.png"/></a>
</h1>
<nav class="nav-bar" id="main-nav">
<a class="main-nav-link" href="../../../install/index.html">Install</a>
<a class="main-nav-link" href="../../../tutorials/index.html">Tutorials</a>
<span id="dropdown-menu-position-anchor">
<a aria-expanded="true" aria-haspopup="true" class="main-nav-link dropdown-toggle" data-toggle="dropdown" href="#" role="button">Gluon <span class="caret"></span></a>
<ul class="dropdown-menu navbar-menu" id="package-dropdown-menu">
<li><a class="main-nav-link" href="../../../gluon/index.html">About</a></li>
<li><a class="main-nav-link" href="http://gluon.mxnet.io">Tutorials</a></li>
</ul>
</span>
<span id="dropdown-menu-position-anchor">
<a aria-expanded="true" aria-haspopup="true" class="main-nav-link dropdown-toggle" data-toggle="dropdown" href="#" role="button">API <span class="caret"></span></a>
<ul class="dropdown-menu navbar-menu" id="package-dropdown-menu">
<li><a class="main-nav-link" href="../../../api/python/index.html">Python</a></li>
<li><a class="main-nav-link" href="../../../api/scala/index.html">Scala</a></li>
<li><a class="main-nav-link" href="../../../api/r/index.html">R</a></li>
<li><a class="main-nav-link" href="../../../api/julia/index.html">Julia</a></li>
<li><a class="main-nav-link" href="../../../api/c++/index.html">C++</a></li>
<li><a class="main-nav-link" href="../../../api/perl/index.html">Perl</a></li>
</ul>
</span>
<span id="dropdown-menu-position-anchor-docs">
<a aria-expanded="true" aria-haspopup="true" class="main-nav-link dropdown-toggle" data-toggle="dropdown" href="#" role="button">Docs <span class="caret"></span></a>
<ul class="dropdown-menu navbar-menu" id="package-dropdown-menu-docs">
<li><a class="main-nav-link" href="../../../faq/index.html">FAQ</a></li>
<li><a class="main-nav-link" href="../../../architecture/index.html">Architecture</a></li>
<li><a class="main-nav-link" href="https://github.com/apache/incubator-mxnet/tree/1.0.0/example">Examples</a></li>
<li><a class="main-nav-link" href="../../../model_zoo/index.html">Model Zoo</a></li>
</ul>
</span>
<a class="main-nav-link" href="https://github.com/dmlc/mxnet">Github</a>
<span id="dropdown-menu-position-anchor-community">
<a aria-expanded="true" aria-haspopup="true" class="main-nav-link dropdown-toggle" data-toggle="dropdown" href="#" role="button">Community <span class="caret"></span></a>
<ul class="dropdown-menu navbar-menu" id="package-dropdown-menu-community">
<li><a class="main-nav-link" href="../../../community/index.html">Community</a></li>
<li><a class="main-nav-link" href="../../../community/contribute.html">Contribute</a></li>
<li><a class="main-nav-link" href="../../../community/powered_by.html">Powered By</a></li>
</ul>
</span>
<a class="main-nav-link" href="http://discuss.mxnet.io">Discuss</a>
<span id="dropdown-menu-position-anchor-version" style="position: relative"><a href="#" class="main-nav-link dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">Versions(1.0.0)<span class="caret"></span></a><ul id="package-dropdown-menu" class="dropdown-menu"><li><a class="main-nav-link" href=https://mxnet.incubator.apache.org/>1.0.0</a></li><li><a class="main-nav-link" href=https://mxnet.incubator.apache.org/versions/0.12.1/index.html>0.12.1</a></li><li><a class="main-nav-link" href=https://mxnet.incubator.apache.org/versions/0.12.0/index.html>0.12.0</a></li><li><a class="main-nav-link" href=https://mxnet.incubator.apache.org/versions/0.11.0/index.html>0.11.0</a></li><li><a class="main-nav-link" href=https://mxnet.incubator.apache.org/versions/master/index.html>master</a></li></ul></span></nav>
<script> function getRootPath(){ return "../../../" } </script>
<div class="burgerIcon dropdown">
<a class="dropdown-toggle" data-toggle="dropdown" href="#" role="button">☰</a>
<ul class="dropdown-menu" id="burgerMenu">
<li><a href="../../../install/index.html">Install</a></li>
<li><a class="main-nav-link" href="../../../tutorials/index.html">Tutorials</a></li>
<li class="dropdown-submenu">
<a href="#" tabindex="-1">Community</a>
<ul class="dropdown-menu">
<li><a href="../../../community/index.html" tabindex="-1">Community</a></li>
<li><a href="../../../community/contribute.html" tabindex="-1">Contribute</a></li>
<li><a href="../../../community/powered_by.html" tabindex="-1">Powered By</a></li>
</ul>
</li>
<li class="dropdown-submenu">
<a href="#" tabindex="-1">API</a>
<ul class="dropdown-menu">
<li><a href="../../../api/python/index.html" tabindex="-1">Python</a>
</li>
<li><a href="../../../api/scala/index.html" tabindex="-1">Scala</a>
</li>
<li><a href="../../../api/r/index.html" tabindex="-1">R</a>
</li>
<li><a href="../../../api/julia/index.html" tabindex="-1">Julia</a>
</li>
<li><a href="../../../api/c++/index.html" tabindex="-1">C++</a>
</li>
<li><a href="../../../api/perl/index.html" tabindex="-1">Perl</a>
</li>
</ul>
</li>
<li class="dropdown-submenu">
<a href="#" tabindex="-1">Docs</a>
<ul class="dropdown-menu">
<li><a href="../../../tutorials/index.html" tabindex="-1">Tutorials</a></li>
<li><a href="../../../faq/index.html" tabindex="-1">FAQ</a></li>
<li><a href="../../../architecture/index.html" tabindex="-1">Architecture</a></li>
<li><a href="https://github.com/apache/incubator-mxnet/tree/1.0.0/example" tabindex="-1">Examples</a></li>
<li><a href="../../../model_zoo/index.html" tabindex="-1">Model Zoo</a></li>
</ul>
</li>
<li><a href="../../../architecture/index.html">Architecture</a></li>
<li><a class="main-nav-link" href="https://github.com/dmlc/mxnet">Github</a></li>
<li id="dropdown-menu-position-anchor-version-mobile" class="dropdown-submenu" style="position: relative"><a href="#" tabindex="-1">Versions(1.0.0)</a><ul class="dropdown-menu"><li><a tabindex="-1" href=https://mxnet.incubator.apache.org/>1.0.0</a></li><li><a tabindex="-1" href=https://mxnet.incubator.apache.org/versions/0.12.1/index.html>0.12.1</a></li><li><a tabindex="-1" href=https://mxnet.incubator.apache.org/versions/0.12.0/index.html>0.12.0</a></li><li><a tabindex="-1" href=https://mxnet.incubator.apache.org/versions/0.11.0/index.html>0.11.0</a></li><li><a tabindex="-1" href=https://mxnet.incubator.apache.org/versions/master/index.html>master</a></li></ul></li></ul>
</div>
<div class="plusIcon dropdown">
<a class="dropdown-toggle" data-toggle="dropdown" href="#" role="button"><span aria-hidden="true" class="glyphicon glyphicon-plus"></span></a>
<ul class="dropdown-menu dropdown-menu-right" id="plusMenu"></ul>
</div>
<div id="search-input-wrap">
<form action="../../../search.html" autocomplete="off" class="" method="get" role="search">
<div class="form-group inner-addon left-addon">
<i class="glyphicon glyphicon-search"></i>
<input class="form-control" name="q" placeholder="Search" type="text"/>
</div>
<input name="check_keywords" type="hidden" value="yes">
<input name="area" type="hidden" value="default"/>
</input></form>
<div id="search-preview"></div>
</div>
<div id="searchIcon">
<span aria-hidden="true" class="glyphicon glyphicon-search"></span>
</div>
<!-- <div id="lang-select-wrap"> -->
<!-- <label id="lang-select-label"> -->
<!-- <\!-- <i class="fa fa-globe"></i> -\-> -->
<!-- <span></span> -->
<!-- </label> -->
<!-- <select id="lang-select"> -->
<!-- <option value="en">Eng</option> -->
<!-- <option value="zh">中文</option> -->
<!-- </select> -->
<!-- </div> -->
<!-- <a id="mobile-nav-toggle">
<span class="mobile-nav-toggle-bar"></span>
<span class="mobile-nav-toggle-bar"></span>
<span class="mobile-nav-toggle-bar"></span>
</a> -->
</div>
</div>
</div>
<script type="text/javascript">
$('body').css('background', 'white');
</script>
<div class="container">
<div class="row">
<div aria-label="main navigation" class="sphinxsidebar leftsidebar" role="navigation">
<div class="sphinxsidebarwrapper">
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../../api/python/index.html">Python Documents</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../../api/r/index.html">R Documents</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../../api/julia/index.html">Julia Documents</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../../api/c++/index.html">C++ Documents</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../../api/scala/index.html">Scala Documents</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../../api/perl/index.html">Perl Documents</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../../faq/index.html">HowTo Documents</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../../architecture/index.html">System Documents</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../../tutorials/index.html">Tutorials</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../../community/index.html">Community</a></li>
</ul>
</div>
</div>
<div class="content">
<div class="page-tracker"></div>
<h1>Source code for mxnet.rnn.rnn_cell</h1><div class="highlight"><pre>
<span></span><span class="c1"># Licensed to the Apache Software Foundation (ASF) under one</span>
<span class="c1"># or more contributor license agreements. See the NOTICE file</span>
<span class="c1"># distributed with this work for additional information</span>
<span class="c1"># regarding copyright ownership. The ASF licenses this file</span>
<span class="c1"># to you under the Apache License, Version 2.0 (the</span>
<span class="c1"># "License"); you may not use this file except in compliance</span>
<span class="c1"># with the License. You may obtain a copy of the License at</span>
<span class="c1">#</span>
<span class="c1"># http://www.apache.org/licenses/LICENSE-2.0</span>
<span class="c1">#</span>
<span class="c1"># Unless required by applicable law or agreed to in writing,</span>
<span class="c1"># software distributed under the License is distributed on an</span>
<span class="c1"># "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY</span>
<span class="c1"># KIND, either express or implied. See the License for the</span>
<span class="c1"># specific language governing permissions and limitations</span>
<span class="c1"># under the License.</span>
<span class="c1"># coding: utf-8</span>
<span class="c1"># pylint: disable=no-member, invalid-name, protected-access, no-self-use</span>
<span class="c1"># pylint: disable=too-many-branches, too-many-arguments, no-self-use</span>
<span class="c1"># pylint: disable=too-many-lines</span>
<span class="sd">"""Definition of various recurrent neural network cells."""</span>
<span class="kn">from</span> <span class="nn">__future__</span> <span class="kn">import</span> <span class="n">print_function</span>
<span class="kn">import</span> <span class="nn">warnings</span>
<span class="kn">import</span> <span class="nn">functools</span>
<span class="kn">from</span> <span class="nn">..</span> <span class="kn">import</span> <span class="n">symbol</span><span class="p">,</span> <span class="n">init</span><span class="p">,</span> <span class="n">ndarray</span>
<span class="kn">from</span> <span class="nn">..base</span> <span class="kn">import</span> <span class="n">string_types</span><span class="p">,</span> <span class="n">numeric_types</span>
<span class="k">def</span> <span class="nf">_cells_state_shape</span><span class="p">(</span><span class="n">cells</span><span class="p">):</span>
<span class="k">return</span> <span class="nb">sum</span><span class="p">([</span><span class="n">c</span><span class="o">.</span><span class="n">state_shape</span> <span class="k">for</span> <span class="n">c</span> <span class="ow">in</span> <span class="n">cells</span><span class="p">],</span> <span class="p">[])</span>
<span class="k">def</span> <span class="nf">_cells_state_info</span><span class="p">(</span><span class="n">cells</span><span class="p">):</span>
<span class="k">return</span> <span class="nb">sum</span><span class="p">([</span><span class="n">c</span><span class="o">.</span><span class="n">state_info</span> <span class="k">for</span> <span class="n">c</span> <span class="ow">in</span> <span class="n">cells</span><span class="p">],</span> <span class="p">[])</span>
<span class="k">def</span> <span class="nf">_cells_begin_state</span><span class="p">(</span><span class="n">cells</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
<span class="k">return</span> <span class="nb">sum</span><span class="p">([</span><span class="n">c</span><span class="o">.</span><span class="n">begin_state</span><span class="p">(</span><span class="o">**</span><span class="n">kwargs</span><span class="p">)</span> <span class="k">for</span> <span class="n">c</span> <span class="ow">in</span> <span class="n">cells</span><span class="p">],</span> <span class="p">[])</span>
<span class="k">def</span> <span class="nf">_cells_unpack_weights</span><span class="p">(</span><span class="n">cells</span><span class="p">,</span> <span class="n">args</span><span class="p">):</span>
<span class="k">for</span> <span class="n">cell</span> <span class="ow">in</span> <span class="n">cells</span><span class="p">:</span>
<span class="n">args</span> <span class="o">=</span> <span class="n">cell</span><span class="o">.</span><span class="n">unpack_weights</span><span class="p">(</span><span class="n">args</span><span class="p">)</span>
<span class="k">return</span> <span class="n">args</span>
<span class="k">def</span> <span class="nf">_cells_pack_weights</span><span class="p">(</span><span class="n">cells</span><span class="p">,</span> <span class="n">args</span><span class="p">):</span>
<span class="k">for</span> <span class="n">cell</span> <span class="ow">in</span> <span class="n">cells</span><span class="p">:</span>
<span class="n">args</span> <span class="o">=</span> <span class="n">cell</span><span class="o">.</span><span class="n">pack_weights</span><span class="p">(</span><span class="n">args</span><span class="p">)</span>
<span class="k">return</span> <span class="n">args</span>
<span class="k">def</span> <span class="nf">_normalize_sequence</span><span class="p">(</span><span class="n">length</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">layout</span><span class="p">,</span> <span class="n">merge</span><span class="p">,</span> <span class="n">in_layout</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
<span class="k">assert</span> <span class="n">inputs</span> <span class="ow">is</span> <span class="ow">not</span> <span class="bp">None</span><span class="p">,</span> \
<span class="s2">"unroll(inputs=None) has been deprecated. "</span> \
<span class="s2">"Please create input variables outside unroll."</span>
<span class="n">axis</span> <span class="o">=</span> <span class="n">layout</span><span class="o">.</span><span class="n">find</span><span class="p">(</span><span class="s1">'T'</span><span class="p">)</span>
<span class="n">in_axis</span> <span class="o">=</span> <span class="n">in_layout</span><span class="o">.</span><span class="n">find</span><span class="p">(</span><span class="s1">'T'</span><span class="p">)</span> <span class="k">if</span> <span class="n">in_layout</span> <span class="ow">is</span> <span class="ow">not</span> <span class="bp">None</span> <span class="k">else</span> <span class="n">axis</span>
<span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Symbol</span><span class="p">):</span>
<span class="k">if</span> <span class="n">merge</span> <span class="ow">is</span> <span class="bp">False</span><span class="p">:</span>
<span class="k">assert</span> <span class="nb">len</span><span class="p">(</span><span class="n">inputs</span><span class="o">.</span><span class="n">list_outputs</span><span class="p">())</span> <span class="o">==</span> <span class="mi">1</span><span class="p">,</span> \
<span class="s2">"unroll doesn't allow grouped symbol as input. Please convert "</span> \
<span class="s2">"to list with list(inputs) first or let unroll handle splitting."</span>
<span class="n">inputs</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">symbol</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="n">axis</span><span class="o">=</span><span class="n">in_axis</span><span class="p">,</span> <span class="n">num_outputs</span><span class="o">=</span><span class="n">length</span><span class="p">,</span>
<span class="n">squeeze_axis</span><span class="o">=</span><span class="mi">1</span><span class="p">))</span>
<span class="k">else</span><span class="p">:</span>
<span class="k">assert</span> <span class="n">length</span> <span class="ow">is</span> <span class="bp">None</span> <span class="ow">or</span> <span class="nb">len</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span> <span class="o">==</span> <span class="n">length</span>
<span class="k">if</span> <span class="n">merge</span> <span class="ow">is</span> <span class="bp">True</span><span class="p">:</span>
<span class="n">inputs</span> <span class="o">=</span> <span class="p">[</span><span class="n">symbol</span><span class="o">.</span><span class="n">expand_dims</span><span class="p">(</span><span class="n">i</span><span class="p">,</span> <span class="n">axis</span><span class="o">=</span><span class="n">axis</span><span class="p">)</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">inputs</span><span class="p">]</span>
<span class="n">inputs</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Concat</span><span class="p">(</span><span class="o">*</span><span class="n">inputs</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="n">axis</span><span class="p">)</span>
<span class="n">in_axis</span> <span class="o">=</span> <span class="n">axis</span>
<span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Symbol</span><span class="p">)</span> <span class="ow">and</span> <span class="n">axis</span> <span class="o">!=</span> <span class="n">in_axis</span><span class="p">:</span>
<span class="n">inputs</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">swapaxes</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="n">dim0</span><span class="o">=</span><span class="n">axis</span><span class="p">,</span> <span class="n">dim1</span><span class="o">=</span><span class="n">in_axis</span><span class="p">)</span>
<span class="k">return</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">axis</span>
<div class="viewcode-block" id="RNNParams"><a class="viewcode-back" href="../../../api/python/symbol/rnn.html#mxnet.rnn.RNNParams">[docs]</a><span class="k">class</span> <span class="nc">RNNParams</span><span class="p">(</span><span class="nb">object</span><span class="p">):</span>
<span class="sd">"""Container for holding variables.</span>
<span class="sd"> Used by RNN cells for parameter sharing between cells.</span>
<span class="sd"> Parameters</span>
<span class="sd"> ----------</span>
<span class="sd"> prefix : str</span>
<span class="sd"> Names of all variables created by this container will</span>
<span class="sd"> be prepended with prefix.</span>
<span class="sd"> """</span>
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">prefix</span><span class="o">=</span><span class="s1">''</span><span class="p">):</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span> <span class="o">=</span> <span class="n">prefix</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_params</span> <span class="o">=</span> <span class="p">{}</span>
<div class="viewcode-block" id="RNNParams.get"><a class="viewcode-back" href="../../../api/python/symbol/rnn.html#mxnet.rnn.RNNParams.get">[docs]</a> <span class="k">def</span> <span class="nf">get</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">name</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
<span class="sd">"""Get the variable given a name if one exists or create a new one if missing.</span>
<span class="sd"> Parameters</span>
<span class="sd"> ----------</span>
<span class="sd"> name : str</span>
<span class="sd"> name of the variable</span>
<span class="sd"> **kwargs :</span>
<span class="sd"> more arguments that's passed to symbol.Variable</span>
<span class="sd"> """</span>
<span class="n">name</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span> <span class="o">+</span> <span class="n">name</span>
<span class="k">if</span> <span class="n">name</span> <span class="ow">not</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_params</span><span class="p">:</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_params</span><span class="p">[</span><span class="n">name</span><span class="p">]</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Variable</span><span class="p">(</span><span class="n">name</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_params</span><span class="p">[</span><span class="n">name</span><span class="p">]</span></div></div>
<div class="viewcode-block" id="BaseRNNCell"><a class="viewcode-back" href="../../../api/python/symbol/rnn.html#mxnet.rnn.BaseRNNCell">[docs]</a><span class="k">class</span> <span class="nc">BaseRNNCell</span><span class="p">(</span><span class="nb">object</span><span class="p">):</span>
<span class="sd">"""Abstract base class for RNN cells</span>
<span class="sd"> Parameters</span>
<span class="sd"> ----------</span>
<span class="sd"> prefix : str, optional</span>
<span class="sd"> Prefix for names of layers</span>
<span class="sd"> (this prefix is also used for names of weights if `params` is None</span>
<span class="sd"> i.e. if `params` are being created and not reused)</span>
<span class="sd"> params : RNNParams, default None.</span>
<span class="sd"> Container for weight sharing between cells.</span>
<span class="sd"> A new RNNParams container is created if `params` is None.</span>
<span class="sd"> """</span>
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">prefix</span><span class="o">=</span><span class="s1">''</span><span class="p">,</span> <span class="n">params</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
<span class="k">if</span> <span class="n">params</span> <span class="ow">is</span> <span class="bp">None</span><span class="p">:</span>
<span class="n">params</span> <span class="o">=</span> <span class="n">RNNParams</span><span class="p">(</span><span class="n">prefix</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_own_params</span> <span class="o">=</span> <span class="bp">True</span>
<span class="k">else</span><span class="p">:</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_own_params</span> <span class="o">=</span> <span class="bp">False</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span> <span class="o">=</span> <span class="n">prefix</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_params</span> <span class="o">=</span> <span class="n">params</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_modified</span> <span class="o">=</span> <span class="bp">False</span>
<span class="bp">self</span><span class="o">.</span><span class="n">reset</span><span class="p">()</span>
<div class="viewcode-block" id="BaseRNNCell.reset"><a class="viewcode-back" href="../../../api/python/symbol/rnn.html#mxnet.rnn.BaseRNNCell.reset">[docs]</a> <span class="k">def</span> <span class="nf">reset</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="sd">"""Reset before re-using the cell for another graph."""</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_init_counter</span> <span class="o">=</span> <span class="o">-</span><span class="mi">1</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_counter</span> <span class="o">=</span> <span class="o">-</span><span class="mi">1</span>
<span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="s1">'_cells'</span><span class="p">):</span>
<span class="k">for</span> <span class="n">cell</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_cells</span><span class="p">:</span>
<span class="n">cell</span><span class="o">.</span><span class="n">reset</span><span class="p">()</span></div>
<div class="viewcode-block" id="BaseRNNCell.__call__"><a class="viewcode-back" href="../../../api/python/symbol/rnn.html#mxnet.rnn.BaseRNNCell.__call__">[docs]</a> <span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">states</span><span class="p">):</span>
<span class="sd">"""Unroll the RNN for one time step.</span>
<span class="sd"> Parameters</span>
<span class="sd"> ----------</span>
<span class="sd"> inputs : sym.Variable</span>
<span class="sd"> input symbol, 2D, batch * num_units</span>
<span class="sd"> states : list of sym.Variable</span>
<span class="sd"> RNN state from previous step or the output of begin_state().</span>
<span class="sd"> Returns</span>
<span class="sd"> -------</span>
<span class="sd"> output : Symbol</span>
<span class="sd"> Symbol corresponding to the output from the RNN when unrolling</span>
<span class="sd"> for a single time step.</span>
<span class="sd"> states : nested list of Symbol</span>
<span class="sd"> The new state of this RNN after this unrolling.</span>
<span class="sd"> The type of this symbol is same as the output of begin_state().</span>
<span class="sd"> This can be used as input state to the next time step</span>
<span class="sd"> of this RNN.</span>
<span class="sd"> See Also</span>
<span class="sd"> --------</span>
<span class="sd"> begin_state: This function can provide the states for the first time step.</span>
<span class="sd"> unroll: This function unrolls an RNN for a given number of (>=1) time steps.</span>
<span class="sd"> """</span>
<span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">()</span></div>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">params</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="sd">"""Parameters of this cell"""</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_own_params</span> <span class="o">=</span> <span class="bp">False</span>
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_params</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">state_info</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="sd">"""shape and layout information of states"""</span>
<span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">()</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">state_shape</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="sd">"""shape(s) of states"""</span>
<span class="k">return</span> <span class="p">[</span><span class="n">ele</span><span class="p">[</span><span class="s1">'shape'</span><span class="p">]</span> <span class="k">for</span> <span class="n">ele</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">state_info</span><span class="p">]</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">_gate_names</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="sd">"""name(s) of gates"""</span>
<span class="k">return</span> <span class="p">()</span>
<div class="viewcode-block" id="BaseRNNCell.begin_state"><a class="viewcode-back" href="../../../api/python/symbol/rnn.html#mxnet.rnn.BaseRNNCell.begin_state">[docs]</a> <span class="k">def</span> <span class="nf">begin_state</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">func</span><span class="o">=</span><span class="n">symbol</span><span class="o">.</span><span class="n">zeros</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
<span class="sd">"""Initial state for this cell.</span>
<span class="sd"> Parameters</span>
<span class="sd"> ----------</span>
<span class="sd"> func : callable, default symbol.zeros</span>
<span class="sd"> Function for creating initial state. Can be symbol.zeros,</span>
<span class="sd"> symbol.uniform, symbol.Variable etc.</span>
<span class="sd"> Use symbol.Variable if you want to directly</span>
<span class="sd"> feed input as states.</span>
<span class="sd"> **kwargs :</span>
<span class="sd"> more keyword arguments passed to func. For example</span>
<span class="sd"> mean, std, dtype, etc.</span>
<span class="sd"> Returns</span>
<span class="sd"> -------</span>
<span class="sd"> states : nested list of Symbol</span>
<span class="sd"> Starting states for the first RNN step.</span>
<span class="sd"> """</span>
<span class="k">assert</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_modified</span><span class="p">,</span> \
<span class="s2">"After applying modifier cells (e.g. DropoutCell) the base "</span> \
<span class="s2">"cell cannot be called directly. Call the modifier cell instead."</span>
<span class="n">states</span> <span class="o">=</span> <span class="p">[]</span>
<span class="k">for</span> <span class="n">info</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">state_info</span><span class="p">:</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_init_counter</span> <span class="o">+=</span> <span class="mi">1</span>
<span class="k">if</span> <span class="n">info</span> <span class="ow">is</span> <span class="bp">None</span><span class="p">:</span>
<span class="n">state</span> <span class="o">=</span> <span class="n">func</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">begin_state_</span><span class="si">%d</span><span class="s1">'</span><span class="o">%</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_init_counter</span><span class="p">),</span>
<span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
<span class="k">else</span><span class="p">:</span>
<span class="n">kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">info</span><span class="p">)</span>
<span class="n">state</span> <span class="o">=</span> <span class="n">func</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">begin_state_</span><span class="si">%d</span><span class="s1">'</span><span class="o">%</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_init_counter</span><span class="p">),</span>
<span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
<span class="n">states</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">state</span><span class="p">)</span>
<span class="k">return</span> <span class="n">states</span></div>
<div class="viewcode-block" id="BaseRNNCell.unpack_weights"><a class="viewcode-back" href="../../../api/python/symbol/rnn.html#mxnet.rnn.BaseRNNCell.unpack_weights">[docs]</a> <span class="k">def</span> <span class="nf">unpack_weights</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">args</span><span class="p">):</span>
<span class="sd">"""Unpack fused weight matrices into separate</span>
<span class="sd"> weight matrices.</span>
<span class="sd"> For example, say you use a module object `mod` to run a network that has an lstm cell.</span>
<span class="sd"> In `mod.get_params()[0]`, the lstm parameters are all represented as a single big vector.</span>
<span class="sd"> `cell.unpack_weights(mod.get_params()[0])` will unpack this vector into a dictionary of</span>
<span class="sd"> more readable lstm parameters - c, f, i, o gates for i2h (input to hidden) and</span>
<span class="sd"> h2h (hidden to hidden) weights.</span>
<span class="sd"> Parameters</span>
<span class="sd"> ----------</span>
<span class="sd"> args : dict of str -> NDArray</span>
<span class="sd"> Dictionary containing packed weights.</span>
<span class="sd"> usually from `Module.get_params()[0]`.</span>
<span class="sd"> Returns</span>
<span class="sd"> -------</span>
<span class="sd"> args : dict of str -> NDArray</span>
<span class="sd"> Dictionary with unpacked weights associated with</span>
<span class="sd"> this cell.</span>
<span class="sd"> See Also</span>
<span class="sd"> --------</span>
<span class="sd"> pack_weights: Performs the reverse operation of this function.</span>
<span class="sd"> """</span>
<span class="n">args</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">copy</span><span class="p">()</span>
<span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_gate_names</span><span class="p">:</span>
<span class="k">return</span> <span class="n">args</span>
<span class="n">h</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span>
<span class="k">for</span> <span class="n">group_name</span> <span class="ow">in</span> <span class="p">[</span><span class="s1">'i2h'</span><span class="p">,</span> <span class="s1">'h2h'</span><span class="p">]:</span>
<span class="n">weight</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">'</span><span class="si">%s%s</span><span class="s1">_weight'</span><span class="o">%</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="n">group_name</span><span class="p">))</span>
<span class="n">bias</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">'</span><span class="si">%s%s</span><span class="s1">_bias'</span> <span class="o">%</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="n">group_name</span><span class="p">))</span>
<span class="k">for</span> <span class="n">j</span><span class="p">,</span> <span class="n">gate</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_gate_names</span><span class="p">):</span>
<span class="n">wname</span> <span class="o">=</span> <span class="s1">'</span><span class="si">%s%s%s</span><span class="s1">_weight'</span> <span class="o">%</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="n">group_name</span><span class="p">,</span> <span class="n">gate</span><span class="p">)</span>
<span class="n">args</span><span class="p">[</span><span class="n">wname</span><span class="p">]</span> <span class="o">=</span> <span class="n">weight</span><span class="p">[</span><span class="n">j</span><span class="o">*</span><span class="n">h</span><span class="p">:(</span><span class="n">j</span><span class="o">+</span><span class="mi">1</span><span class="p">)</span><span class="o">*</span><span class="n">h</span><span class="p">]</span><span class="o">.</span><span class="n">copy</span><span class="p">()</span>
<span class="n">bname</span> <span class="o">=</span> <span class="s1">'</span><span class="si">%s%s%s</span><span class="s1">_bias'</span> <span class="o">%</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="n">group_name</span><span class="p">,</span> <span class="n">gate</span><span class="p">)</span>
<span class="n">args</span><span class="p">[</span><span class="n">bname</span><span class="p">]</span> <span class="o">=</span> <span class="n">bias</span><span class="p">[</span><span class="n">j</span><span class="o">*</span><span class="n">h</span><span class="p">:(</span><span class="n">j</span><span class="o">+</span><span class="mi">1</span><span class="p">)</span><span class="o">*</span><span class="n">h</span><span class="p">]</span><span class="o">.</span><span class="n">copy</span><span class="p">()</span>
<span class="k">return</span> <span class="n">args</span></div>
<div class="viewcode-block" id="BaseRNNCell.pack_weights"><a class="viewcode-back" href="../../../api/python/symbol/rnn.html#mxnet.rnn.BaseRNNCell.pack_weights">[docs]</a> <span class="k">def</span> <span class="nf">pack_weights</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">args</span><span class="p">):</span>
<span class="sd">"""Pack separate weight matrices into a single packed</span>
<span class="sd"> weight.</span>
<span class="sd"> Parameters</span>
<span class="sd"> ----------</span>
<span class="sd"> args : dict of str -> NDArray</span>
<span class="sd"> Dictionary containing unpacked weights.</span>
<span class="sd"> Returns</span>
<span class="sd"> -------</span>
<span class="sd"> args : dict of str -> NDArray</span>
<span class="sd"> Dictionary with packed weights associated with</span>
<span class="sd"> this cell.</span>
<span class="sd"> """</span>
<span class="n">args</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">copy</span><span class="p">()</span>
<span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_gate_names</span><span class="p">:</span>
<span class="k">return</span> <span class="n">args</span>
<span class="k">for</span> <span class="n">group_name</span> <span class="ow">in</span> <span class="p">[</span><span class="s1">'i2h'</span><span class="p">,</span> <span class="s1">'h2h'</span><span class="p">]:</span>
<span class="n">weight</span> <span class="o">=</span> <span class="p">[]</span>
<span class="n">bias</span> <span class="o">=</span> <span class="p">[]</span>
<span class="k">for</span> <span class="n">gate</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_gate_names</span><span class="p">:</span>
<span class="n">wname</span> <span class="o">=</span> <span class="s1">'</span><span class="si">%s%s%s</span><span class="s1">_weight'</span><span class="o">%</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="n">group_name</span><span class="p">,</span> <span class="n">gate</span><span class="p">)</span>
<span class="n">weight</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="n">wname</span><span class="p">))</span>
<span class="n">bname</span> <span class="o">=</span> <span class="s1">'</span><span class="si">%s%s%s</span><span class="s1">_bias'</span><span class="o">%</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="n">group_name</span><span class="p">,</span> <span class="n">gate</span><span class="p">)</span>
<span class="n">bias</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="n">bname</span><span class="p">))</span>
<span class="n">args</span><span class="p">[</span><span class="s1">'</span><span class="si">%s%s</span><span class="s1">_weight'</span><span class="o">%</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="n">group_name</span><span class="p">)]</span> <span class="o">=</span> <span class="n">ndarray</span><span class="o">.</span><span class="n">concatenate</span><span class="p">(</span><span class="n">weight</span><span class="p">)</span>
<span class="n">args</span><span class="p">[</span><span class="s1">'</span><span class="si">%s%s</span><span class="s1">_bias'</span><span class="o">%</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="n">group_name</span><span class="p">)]</span> <span class="o">=</span> <span class="n">ndarray</span><span class="o">.</span><span class="n">concatenate</span><span class="p">(</span><span class="n">bias</span><span class="p">)</span>
<span class="k">return</span> <span class="n">args</span></div>
<div class="viewcode-block" id="BaseRNNCell.unroll"><a class="viewcode-back" href="../../../api/python/symbol/rnn.html#mxnet.rnn.BaseRNNCell.unroll">[docs]</a> <span class="k">def</span> <span class="nf">unroll</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">length</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">begin_state</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">layout</span><span class="o">=</span><span class="s1">'NTC'</span><span class="p">,</span> <span class="n">merge_outputs</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
<span class="sd">"""Unroll an RNN cell across time steps.</span>
<span class="sd"> Parameters</span>
<span class="sd"> ----------</span>
<span class="sd"> length : int</span>
<span class="sd"> Number of steps to unroll.</span>
<span class="sd"> inputs : Symbol, list of Symbol, or None</span>
<span class="sd"> If `inputs` is a single Symbol (usually the output</span>
<span class="sd"> of Embedding symbol), it should have shape</span>
<span class="sd"> (batch_size, length, ...) if layout == 'NTC',</span>
<span class="sd"> or (length, batch_size, ...) if layout == 'TNC'.</span>
<span class="sd"> If `inputs` is a list of symbols (usually output of</span>
<span class="sd"> previous unroll), they should all have shape</span>
<span class="sd"> (batch_size, ...).</span>
<span class="sd"> begin_state : nested list of Symbol, default None</span>
<span class="sd"> Input states created by `begin_state()`</span>
<span class="sd"> or output state of another cell.</span>
<span class="sd"> Created from `begin_state()` if None.</span>
<span class="sd"> layout : str, optional</span>
<span class="sd"> `layout` of input symbol. Only used if inputs</span>
<span class="sd"> is a single Symbol.</span>
<span class="sd"> merge_outputs : bool, optional</span>
<span class="sd"> If False, return outputs as a list of Symbols.</span>
<span class="sd"> If True, concatenate output across time steps</span>
<span class="sd"> and return a single symbol with shape</span>
<span class="sd"> (batch_size, length, ...) if layout == 'NTC',</span>
<span class="sd"> or (length, batch_size, ...) if layout == 'TNC'.</span>
<span class="sd"> If None, output whatever is faster.</span>
<span class="sd"> Returns</span>
<span class="sd"> -------</span>
<span class="sd"> outputs : list of Symbol or Symbol</span>
<span class="sd"> Symbol (if `merge_outputs` is True) or list of Symbols</span>
<span class="sd"> (if `merge_outputs` is False) corresponding to the output from</span>
<span class="sd"> the RNN from this unrolling.</span>
<span class="sd"> states : nested list of Symbol</span>
<span class="sd"> The new state of this RNN after this unrolling.</span>
<span class="sd"> The type of this symbol is same as the output of begin_state().</span>
<span class="sd"> """</span>
<span class="bp">self</span><span class="o">.</span><span class="n">reset</span><span class="p">()</span>
<span class="n">inputs</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">_normalize_sequence</span><span class="p">(</span><span class="n">length</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">layout</span><span class="p">,</span> <span class="bp">False</span><span class="p">)</span>
<span class="k">if</span> <span class="n">begin_state</span> <span class="ow">is</span> <span class="bp">None</span><span class="p">:</span>
<span class="n">begin_state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">begin_state</span><span class="p">()</span>
<span class="n">states</span> <span class="o">=</span> <span class="n">begin_state</span>
<span class="n">outputs</span> <span class="o">=</span> <span class="p">[]</span>
<span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">length</span><span class="p">):</span>
<span class="n">output</span><span class="p">,</span> <span class="n">states</span> <span class="o">=</span> <span class="bp">self</span><span class="p">(</span><span class="n">inputs</span><span class="p">[</span><span class="n">i</span><span class="p">],</span> <span class="n">states</span><span class="p">)</span>
<span class="n">outputs</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">output</span><span class="p">)</span>
<span class="n">outputs</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">_normalize_sequence</span><span class="p">(</span><span class="n">length</span><span class="p">,</span> <span class="n">outputs</span><span class="p">,</span> <span class="n">layout</span><span class="p">,</span> <span class="n">merge_outputs</span><span class="p">)</span>
<span class="k">return</span> <span class="n">outputs</span><span class="p">,</span> <span class="n">states</span></div>
<span class="c1">#pylint: disable=no-self-use</span>
<span class="k">def</span> <span class="nf">_get_activation</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">activation</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
<span class="sd">"""Get activation function. Convert if is string"""</span>
<span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">activation</span><span class="p">,</span> <span class="n">string_types</span><span class="p">):</span>
<span class="k">return</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Activation</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="n">act_type</span><span class="o">=</span><span class="n">activation</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
<span class="k">else</span><span class="p">:</span>
<span class="k">return</span> <span class="n">activation</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
<div class="viewcode-block" id="RNNCell"><a class="viewcode-back" href="../../../api/python/symbol/rnn.html#mxnet.rnn.RNNCell">[docs]</a><span class="k">class</span> <span class="nc">RNNCell</span><span class="p">(</span><span class="n">BaseRNNCell</span><span class="p">):</span>
<span class="sd">"""Simple recurrent neural network cell.</span>
<span class="sd"> Parameters</span>
<span class="sd"> ----------</span>
<span class="sd"> num_hidden : int</span>
<span class="sd"> Number of units in output symbol.</span>
<span class="sd"> activation : str or Symbol, default 'tanh'</span>
<span class="sd"> Type of activation function. Options are 'relu' and 'tanh'.</span>
<span class="sd"> prefix : str, default 'rnn_'</span>
<span class="sd"> Prefix for name of layers (and name of weight if params is None).</span>
<span class="sd"> params : RNNParams, default None</span>
<span class="sd"> Container for weight sharing between cells. Created if None.</span>
<span class="sd"> """</span>
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">num_hidden</span><span class="p">,</span> <span class="n">activation</span><span class="o">=</span><span class="s1">'tanh'</span><span class="p">,</span> <span class="n">prefix</span><span class="o">=</span><span class="s1">'rnn_'</span><span class="p">,</span> <span class="n">params</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
<span class="nb">super</span><span class="p">(</span><span class="n">RNNCell</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">prefix</span><span class="o">=</span><span class="n">prefix</span><span class="p">,</span> <span class="n">params</span><span class="o">=</span><span class="n">params</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span> <span class="o">=</span> <span class="n">num_hidden</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_activation</span> <span class="o">=</span> <span class="n">activation</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_iW</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'i2h_weight'</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_iB</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'i2h_bias'</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_hW</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'h2h_weight'</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_hB</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'h2h_bias'</span><span class="p">)</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">state_info</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">return</span> <span class="p">[{</span><span class="s1">'shape'</span><span class="p">:</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span><span class="p">),</span> <span class="s1">'__layout__'</span><span class="p">:</span> <span class="s1">'NC'</span><span class="p">}]</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">_gate_names</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">return</span> <span class="p">(</span><span class="s1">''</span><span class="p">,)</span>
<span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">states</span><span class="p">):</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_counter</span> <span class="o">+=</span> <span class="mi">1</span>
<span class="n">name</span> <span class="o">=</span> <span class="s1">'</span><span class="si">%s</span><span class="s1">t</span><span class="si">%d</span><span class="s1">_'</span><span class="o">%</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_counter</span><span class="p">)</span>
<span class="n">i2h</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">FullyConnected</span><span class="p">(</span><span class="n">data</span><span class="o">=</span><span class="n">inputs</span><span class="p">,</span> <span class="n">weight</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_iW</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_iB</span><span class="p">,</span>
<span class="n">num_hidden</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">i2h'</span><span class="o">%</span><span class="n">name</span><span class="p">)</span>
<span class="n">h2h</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">FullyConnected</span><span class="p">(</span><span class="n">data</span><span class="o">=</span><span class="n">states</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">weight</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_hW</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_hB</span><span class="p">,</span>
<span class="n">num_hidden</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">h2h'</span><span class="o">%</span><span class="n">name</span><span class="p">)</span>
<span class="n">output</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_activation</span><span class="p">(</span><span class="n">i2h</span> <span class="o">+</span> <span class="n">h2h</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_activation</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">out'</span><span class="o">%</span><span class="n">name</span><span class="p">)</span>
<span class="k">return</span> <span class="n">output</span><span class="p">,</span> <span class="p">[</span><span class="n">output</span><span class="p">]</span></div>
<div class="viewcode-block" id="LSTMCell"><a class="viewcode-back" href="../../../api/python/symbol/rnn.html#mxnet.rnn.LSTMCell">[docs]</a><span class="k">class</span> <span class="nc">LSTMCell</span><span class="p">(</span><span class="n">BaseRNNCell</span><span class="p">):</span>
<span class="sd">"""Long-Short Term Memory (LSTM) network cell.</span>
<span class="sd"> Parameters</span>
<span class="sd"> ----------</span>
<span class="sd"> num_hidden : int</span>
<span class="sd"> Number of units in output symbol.</span>
<span class="sd"> prefix : str, default 'lstm_'</span>
<span class="sd"> Prefix for name of layers (and name of weight if params is None).</span>
<span class="sd"> params : RNNParams, default None</span>
<span class="sd"> Container for weight sharing between cells. Created if None.</span>
<span class="sd"> forget_bias : bias added to forget gate, default 1.0.</span>
<span class="sd"> Jozefowicz et al. 2015 recommends setting this to 1.0</span>
<span class="sd"> """</span>
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">num_hidden</span><span class="p">,</span> <span class="n">prefix</span><span class="o">=</span><span class="s1">'lstm_'</span><span class="p">,</span> <span class="n">params</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">forget_bias</span><span class="o">=</span><span class="mf">1.0</span><span class="p">):</span>
<span class="nb">super</span><span class="p">(</span><span class="n">LSTMCell</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">prefix</span><span class="o">=</span><span class="n">prefix</span><span class="p">,</span> <span class="n">params</span><span class="o">=</span><span class="n">params</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span> <span class="o">=</span> <span class="n">num_hidden</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_iW</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'i2h_weight'</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_hW</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'h2h_weight'</span><span class="p">)</span>
<span class="c1"># we add the forget_bias to i2h_bias, this adds the bias to the forget gate activation</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_iB</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'i2h_bias'</span><span class="p">,</span> <span class="n">init</span><span class="o">=</span><span class="n">init</span><span class="o">.</span><span class="n">LSTMBias</span><span class="p">(</span><span class="n">forget_bias</span><span class="o">=</span><span class="n">forget_bias</span><span class="p">))</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_hB</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'h2h_bias'</span><span class="p">)</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">state_info</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">return</span> <span class="p">[{</span><span class="s1">'shape'</span><span class="p">:</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span><span class="p">),</span> <span class="s1">'__layout__'</span><span class="p">:</span> <span class="s1">'NC'</span><span class="p">},</span>
<span class="p">{</span><span class="s1">'shape'</span><span class="p">:</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span><span class="p">),</span> <span class="s1">'__layout__'</span><span class="p">:</span> <span class="s1">'NC'</span><span class="p">}]</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">_gate_names</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">return</span> <span class="p">[</span><span class="s1">'_i'</span><span class="p">,</span> <span class="s1">'_f'</span><span class="p">,</span> <span class="s1">'_c'</span><span class="p">,</span> <span class="s1">'_o'</span><span class="p">]</span>
<span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">states</span><span class="p">):</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_counter</span> <span class="o">+=</span> <span class="mi">1</span>
<span class="n">name</span> <span class="o">=</span> <span class="s1">'</span><span class="si">%s</span><span class="s1">t</span><span class="si">%d</span><span class="s1">_'</span><span class="o">%</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_counter</span><span class="p">)</span>
<span class="n">i2h</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">FullyConnected</span><span class="p">(</span><span class="n">data</span><span class="o">=</span><span class="n">inputs</span><span class="p">,</span> <span class="n">weight</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_iW</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_iB</span><span class="p">,</span>
<span class="n">num_hidden</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span><span class="o">*</span><span class="mi">4</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">i2h'</span><span class="o">%</span><span class="n">name</span><span class="p">)</span>
<span class="n">h2h</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">FullyConnected</span><span class="p">(</span><span class="n">data</span><span class="o">=</span><span class="n">states</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">weight</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_hW</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_hB</span><span class="p">,</span>
<span class="n">num_hidden</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span><span class="o">*</span><span class="mi">4</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">h2h'</span><span class="o">%</span><span class="n">name</span><span class="p">)</span>
<span class="n">gates</span> <span class="o">=</span> <span class="n">i2h</span> <span class="o">+</span> <span class="n">h2h</span>
<span class="n">slice_gates</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">SliceChannel</span><span class="p">(</span><span class="n">gates</span><span class="p">,</span> <span class="n">num_outputs</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s2">"</span><span class="si">%s</span><span class="s2">slice"</span><span class="o">%</span><span class="n">name</span><span class="p">)</span>
<span class="n">in_gate</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Activation</span><span class="p">(</span><span class="n">slice_gates</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">act_type</span><span class="o">=</span><span class="s2">"sigmoid"</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">i'</span><span class="o">%</span><span class="n">name</span><span class="p">)</span>
<span class="n">forget_gate</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Activation</span><span class="p">(</span><span class="n">slice_gates</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">act_type</span><span class="o">=</span><span class="s2">"sigmoid"</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">f'</span><span class="o">%</span><span class="n">name</span><span class="p">)</span>
<span class="n">in_transform</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Activation</span><span class="p">(</span><span class="n">slice_gates</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span> <span class="n">act_type</span><span class="o">=</span><span class="s2">"tanh"</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">c'</span><span class="o">%</span><span class="n">name</span><span class="p">)</span>
<span class="n">out_gate</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Activation</span><span class="p">(</span><span class="n">slice_gates</span><span class="p">[</span><span class="mi">3</span><span class="p">],</span> <span class="n">act_type</span><span class="o">=</span><span class="s2">"sigmoid"</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">o'</span><span class="o">%</span><span class="n">name</span><span class="p">)</span>
<span class="n">next_c</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">_internal</span><span class="o">.</span><span class="n">_plus</span><span class="p">(</span><span class="n">forget_gate</span> <span class="o">*</span> <span class="n">states</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">in_gate</span> <span class="o">*</span> <span class="n">in_transform</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">state'</span><span class="o">%</span><span class="n">name</span><span class="p">)</span>
<span class="n">next_h</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">_internal</span><span class="o">.</span><span class="n">_mul</span><span class="p">(</span><span class="n">out_gate</span><span class="p">,</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Activation</span><span class="p">(</span><span class="n">next_c</span><span class="p">,</span> <span class="n">act_type</span><span class="o">=</span><span class="s2">"tanh"</span><span class="p">),</span>
<span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">out'</span><span class="o">%</span><span class="n">name</span><span class="p">)</span>
<span class="k">return</span> <span class="n">next_h</span><span class="p">,</span> <span class="p">[</span><span class="n">next_h</span><span class="p">,</span> <span class="n">next_c</span><span class="p">]</span></div>
<div class="viewcode-block" id="GRUCell"><a class="viewcode-back" href="../../../api/python/symbol/rnn.html#mxnet.rnn.GRUCell">[docs]</a><span class="k">class</span> <span class="nc">GRUCell</span><span class="p">(</span><span class="n">BaseRNNCell</span><span class="p">):</span>
<span class="sd">"""Gated Rectified Unit (GRU) network cell.</span>
<span class="sd"> Note: this is an implementation of the cuDNN version of GRUs</span>
<span class="sd"> (slight modification compared to Cho et al. 2014).</span>
<span class="sd"> Parameters</span>
<span class="sd"> ----------</span>
<span class="sd"> num_hidden : int</span>
<span class="sd"> Number of units in output symbol.</span>
<span class="sd"> prefix : str, default 'gru_'</span>
<span class="sd"> Prefix for name of layers (and name of weight if params is None).</span>
<span class="sd"> params : RNNParams, default None</span>
<span class="sd"> Container for weight sharing between cells. Created if None.</span>
<span class="sd"> """</span>
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">num_hidden</span><span class="p">,</span> <span class="n">prefix</span><span class="o">=</span><span class="s1">'gru_'</span><span class="p">,</span> <span class="n">params</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
<span class="nb">super</span><span class="p">(</span><span class="n">GRUCell</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">prefix</span><span class="o">=</span><span class="n">prefix</span><span class="p">,</span> <span class="n">params</span><span class="o">=</span><span class="n">params</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span> <span class="o">=</span> <span class="n">num_hidden</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_iW</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">"i2h_weight"</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_iB</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">"i2h_bias"</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_hW</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">"h2h_weight"</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_hB</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">"h2h_bias"</span><span class="p">)</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">state_info</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">return</span> <span class="p">[{</span><span class="s1">'shape'</span><span class="p">:</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span><span class="p">),</span>
<span class="s1">'__layout__'</span><span class="p">:</span> <span class="s1">'NC'</span><span class="p">}]</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">_gate_names</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">return</span> <span class="p">[</span><span class="s1">'_r'</span><span class="p">,</span> <span class="s1">'_z'</span><span class="p">,</span> <span class="s1">'_o'</span><span class="p">]</span>
<span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">states</span><span class="p">):</span>
<span class="c1"># pylint: disable=too-many-locals</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_counter</span> <span class="o">+=</span> <span class="mi">1</span>
<span class="n">seq_idx</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_counter</span>
<span class="n">name</span> <span class="o">=</span> <span class="s1">'</span><span class="si">%s</span><span class="s1">t</span><span class="si">%d</span><span class="s1">_'</span> <span class="o">%</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="n">seq_idx</span><span class="p">)</span>
<span class="n">prev_state_h</span> <span class="o">=</span> <span class="n">states</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
<span class="n">i2h</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">FullyConnected</span><span class="p">(</span><span class="n">data</span><span class="o">=</span><span class="n">inputs</span><span class="p">,</span>
<span class="n">weight</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_iW</span><span class="p">,</span>
<span class="n">bias</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_iB</span><span class="p">,</span>
<span class="n">num_hidden</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span> <span class="o">*</span> <span class="mi">3</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s2">"</span><span class="si">%s</span><span class="s2">_i2h"</span> <span class="o">%</span> <span class="n">name</span><span class="p">)</span>
<span class="n">h2h</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">FullyConnected</span><span class="p">(</span><span class="n">data</span><span class="o">=</span><span class="n">prev_state_h</span><span class="p">,</span>
<span class="n">weight</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_hW</span><span class="p">,</span>
<span class="n">bias</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_hB</span><span class="p">,</span>
<span class="n">num_hidden</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span> <span class="o">*</span> <span class="mi">3</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s2">"</span><span class="si">%s</span><span class="s2">_h2h"</span> <span class="o">%</span> <span class="n">name</span><span class="p">)</span>
<span class="n">i2h_r</span><span class="p">,</span> <span class="n">i2h_z</span><span class="p">,</span> <span class="n">i2h</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">SliceChannel</span><span class="p">(</span><span class="n">i2h</span><span class="p">,</span> <span class="n">num_outputs</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">name</span><span class="o">=</span><span class="s2">"</span><span class="si">%s</span><span class="s2">_i2h_slice"</span> <span class="o">%</span> <span class="n">name</span><span class="p">)</span>
<span class="n">h2h_r</span><span class="p">,</span> <span class="n">h2h_z</span><span class="p">,</span> <span class="n">h2h</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">SliceChannel</span><span class="p">(</span><span class="n">h2h</span><span class="p">,</span> <span class="n">num_outputs</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">name</span><span class="o">=</span><span class="s2">"</span><span class="si">%s</span><span class="s2">_h2h_slice"</span> <span class="o">%</span> <span class="n">name</span><span class="p">)</span>
<span class="n">reset_gate</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Activation</span><span class="p">(</span><span class="n">i2h_r</span> <span class="o">+</span> <span class="n">h2h_r</span><span class="p">,</span> <span class="n">act_type</span><span class="o">=</span><span class="s2">"sigmoid"</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s2">"</span><span class="si">%s</span><span class="s2">_r_act"</span> <span class="o">%</span> <span class="n">name</span><span class="p">)</span>
<span class="n">update_gate</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Activation</span><span class="p">(</span><span class="n">i2h_z</span> <span class="o">+</span> <span class="n">h2h_z</span><span class="p">,</span> <span class="n">act_type</span><span class="o">=</span><span class="s2">"sigmoid"</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s2">"</span><span class="si">%s</span><span class="s2">_z_act"</span> <span class="o">%</span> <span class="n">name</span><span class="p">)</span>
<span class="n">next_h_tmp</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Activation</span><span class="p">(</span><span class="n">i2h</span> <span class="o">+</span> <span class="n">reset_gate</span> <span class="o">*</span> <span class="n">h2h</span><span class="p">,</span> <span class="n">act_type</span><span class="o">=</span><span class="s2">"tanh"</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s2">"</span><span class="si">%s</span><span class="s2">_h_act"</span> <span class="o">%</span> <span class="n">name</span><span class="p">)</span>
<span class="n">next_h</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">_internal</span><span class="o">.</span><span class="n">_plus</span><span class="p">((</span><span class="mf">1.</span> <span class="o">-</span> <span class="n">update_gate</span><span class="p">)</span> <span class="o">*</span> <span class="n">next_h_tmp</span><span class="p">,</span> <span class="n">update_gate</span> <span class="o">*</span> <span class="n">prev_state_h</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">out'</span> <span class="o">%</span> <span class="n">name</span><span class="p">)</span>
<span class="k">return</span> <span class="n">next_h</span><span class="p">,</span> <span class="p">[</span><span class="n">next_h</span><span class="p">]</span></div>
<div class="viewcode-block" id="FusedRNNCell"><a class="viewcode-back" href="../../../api/python/symbol/rnn.html#mxnet.rnn.FusedRNNCell">[docs]</a><span class="k">class</span> <span class="nc">FusedRNNCell</span><span class="p">(</span><span class="n">BaseRNNCell</span><span class="p">):</span>
<span class="sd">"""Fusing RNN layers across time step into one kernel.</span>
<span class="sd"> Improves speed but is less flexible. Currently only</span>
<span class="sd"> supported if using cuDNN on GPU.</span>
<span class="sd"> Parameters</span>
<span class="sd"> ----------</span>
<span class="sd"> num_hidden : int</span>
<span class="sd"> Number of units in output symbol.</span>
<span class="sd"> num_layers : int, default 1</span>
<span class="sd"> Number of layers in the cell.</span>
<span class="sd"> mode : str, default 'lstm'</span>
<span class="sd"> Type of RNN. options are 'rnn_relu', 'rnn_tanh', 'lstm', 'gru'.</span>
<span class="sd"> bidirectional : bool, default False</span>
<span class="sd"> Whether to use bidirectional unroll. The output dimension size is doubled if bidrectional.</span>
<span class="sd"> dropout : float, default 0.</span>
<span class="sd"> Fraction of the input that gets dropped out during training time.</span>
<span class="sd"> get_next_state : bool, default False</span>
<span class="sd"> Whether to return the states that can be used as starting states next time.</span>
<span class="sd"> forget_bias : bias added to forget gate, default 1.0.</span>
<span class="sd"> Jozefowicz et al. 2015 recommends setting this to 1.0</span>
<span class="sd"> prefix : str, default '$mode_' such as 'lstm_'</span>
<span class="sd"> Prefix for names of layers</span>
<span class="sd"> (this prefix is also used for names of weights if `params` is None</span>
<span class="sd"> i.e. if `params` are being created and not reused)</span>
<span class="sd"> params : RNNParams, default None</span>
<span class="sd"> Container for weight sharing between cells. Created if None.</span>
<span class="sd"> """</span>
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">num_hidden</span><span class="p">,</span> <span class="n">num_layers</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s1">'lstm'</span><span class="p">,</span> <span class="n">bidirectional</span><span class="o">=</span><span class="bp">False</span><span class="p">,</span>
<span class="n">dropout</span><span class="o">=</span><span class="mf">0.</span><span class="p">,</span> <span class="n">get_next_state</span><span class="o">=</span><span class="bp">False</span><span class="p">,</span> <span class="n">forget_bias</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span>
<span class="n">prefix</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">params</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
<span class="k">if</span> <span class="n">prefix</span> <span class="ow">is</span> <span class="bp">None</span><span class="p">:</span>
<span class="n">prefix</span> <span class="o">=</span> <span class="s1">'</span><span class="si">%s</span><span class="s1">_'</span><span class="o">%</span><span class="n">mode</span>
<span class="nb">super</span><span class="p">(</span><span class="n">FusedRNNCell</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">prefix</span><span class="o">=</span><span class="n">prefix</span><span class="p">,</span> <span class="n">params</span><span class="o">=</span><span class="n">params</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span> <span class="o">=</span> <span class="n">num_hidden</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_num_layers</span> <span class="o">=</span> <span class="n">num_layers</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_mode</span> <span class="o">=</span> <span class="n">mode</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_bidirectional</span> <span class="o">=</span> <span class="n">bidirectional</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_dropout</span> <span class="o">=</span> <span class="n">dropout</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_get_next_state</span> <span class="o">=</span> <span class="n">get_next_state</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_directions</span> <span class="o">=</span> <span class="p">[</span><span class="s1">'l'</span><span class="p">,</span> <span class="s1">'r'</span><span class="p">]</span> <span class="k">if</span> <span class="n">bidirectional</span> <span class="k">else</span> <span class="p">[</span><span class="s1">'l'</span><span class="p">]</span>
<span class="n">initializer</span> <span class="o">=</span> <span class="n">init</span><span class="o">.</span><span class="n">FusedRNN</span><span class="p">(</span><span class="bp">None</span><span class="p">,</span> <span class="n">num_hidden</span><span class="p">,</span> <span class="n">num_layers</span><span class="p">,</span> <span class="n">mode</span><span class="p">,</span>
<span class="n">bidirectional</span><span class="p">,</span> <span class="n">forget_bias</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_parameter</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'parameters'</span><span class="p">,</span> <span class="n">init</span><span class="o">=</span><span class="n">initializer</span><span class="p">)</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">state_info</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="n">b</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_bidirectional</span> <span class="o">+</span> <span class="mi">1</span>
<span class="n">n</span> <span class="o">=</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_mode</span> <span class="o">==</span> <span class="s1">'lstm'</span><span class="p">)</span> <span class="o">+</span> <span class="mi">1</span>
<span class="k">return</span> <span class="p">[{</span><span class="s1">'shape'</span><span class="p">:</span> <span class="p">(</span><span class="n">b</span><span class="o">*</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_layers</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span><span class="p">),</span> <span class="s1">'__layout__'</span><span class="p">:</span> <span class="s1">'LNC'</span><span class="p">}</span>
<span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">n</span><span class="p">)]</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">_gate_names</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">return</span> <span class="p">{</span><span class="s1">'rnn_relu'</span><span class="p">:</span> <span class="p">[</span><span class="s1">''</span><span class="p">],</span>
<span class="s1">'rnn_tanh'</span><span class="p">:</span> <span class="p">[</span><span class="s1">''</span><span class="p">],</span>
<span class="s1">'lstm'</span><span class="p">:</span> <span class="p">[</span><span class="s1">'_i'</span><span class="p">,</span> <span class="s1">'_f'</span><span class="p">,</span> <span class="s1">'_c'</span><span class="p">,</span> <span class="s1">'_o'</span><span class="p">],</span>
<span class="s1">'gru'</span><span class="p">:</span> <span class="p">[</span><span class="s1">'_r'</span><span class="p">,</span> <span class="s1">'_z'</span><span class="p">,</span> <span class="s1">'_o'</span><span class="p">]}[</span><span class="bp">self</span><span class="o">.</span><span class="n">_mode</span><span class="p">]</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">_num_gates</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_gate_names</span><span class="p">)</span>
<span class="k">def</span> <span class="nf">_slice_weights</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">arr</span><span class="p">,</span> <span class="n">li</span><span class="p">,</span> <span class="n">lh</span><span class="p">):</span>
<span class="sd">"""slice fused rnn weights"""</span>
<span class="n">args</span> <span class="o">=</span> <span class="p">{}</span>
<span class="n">gate_names</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_gate_names</span>
<span class="n">directions</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_directions</span>
<span class="n">b</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">directions</span><span class="p">)</span>
<span class="n">p</span> <span class="o">=</span> <span class="mi">0</span>
<span class="k">for</span> <span class="n">layer</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_layers</span><span class="p">):</span>
<span class="k">for</span> <span class="n">direction</span> <span class="ow">in</span> <span class="n">directions</span><span class="p">:</span>
<span class="k">for</span> <span class="n">gate</span> <span class="ow">in</span> <span class="n">gate_names</span><span class="p">:</span>
<span class="n">name</span> <span class="o">=</span> <span class="s1">'</span><span class="si">%s%s%d</span><span class="s1">_i2h</span><span class="si">%s</span><span class="s1">_weight'</span><span class="o">%</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="n">direction</span><span class="p">,</span> <span class="n">layer</span><span class="p">,</span> <span class="n">gate</span><span class="p">)</span>
<span class="k">if</span> <span class="n">layer</span> <span class="o">></span> <span class="mi">0</span><span class="p">:</span>
<span class="n">size</span> <span class="o">=</span> <span class="n">b</span><span class="o">*</span><span class="n">lh</span><span class="o">*</span><span class="n">lh</span>
<span class="n">args</span><span class="p">[</span><span class="n">name</span><span class="p">]</span> <span class="o">=</span> <span class="n">arr</span><span class="p">[</span><span class="n">p</span><span class="p">:</span><span class="n">p</span><span class="o">+</span><span class="n">size</span><span class="p">]</span><span class="o">.</span><span class="n">reshape</span><span class="p">((</span><span class="n">lh</span><span class="p">,</span> <span class="n">b</span><span class="o">*</span><span class="n">lh</span><span class="p">))</span>
<span class="k">else</span><span class="p">:</span>
<span class="n">size</span> <span class="o">=</span> <span class="n">li</span><span class="o">*</span><span class="n">lh</span>
<span class="n">args</span><span class="p">[</span><span class="n">name</span><span class="p">]</span> <span class="o">=</span> <span class="n">arr</span><span class="p">[</span><span class="n">p</span><span class="p">:</span><span class="n">p</span><span class="o">+</span><span class="n">size</span><span class="p">]</span><span class="o">.</span><span class="n">reshape</span><span class="p">((</span><span class="n">lh</span><span class="p">,</span> <span class="n">li</span><span class="p">))</span>
<span class="n">p</span> <span class="o">+=</span> <span class="n">size</span>
<span class="k">for</span> <span class="n">gate</span> <span class="ow">in</span> <span class="n">gate_names</span><span class="p">:</span>
<span class="n">name</span> <span class="o">=</span> <span class="s1">'</span><span class="si">%s%s%d</span><span class="s1">_h2h</span><span class="si">%s</span><span class="s1">_weight'</span><span class="o">%</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="n">direction</span><span class="p">,</span> <span class="n">layer</span><span class="p">,</span> <span class="n">gate</span><span class="p">)</span>
<span class="n">size</span> <span class="o">=</span> <span class="n">lh</span><span class="o">**</span><span class="mi">2</span>
<span class="n">args</span><span class="p">[</span><span class="n">name</span><span class="p">]</span> <span class="o">=</span> <span class="n">arr</span><span class="p">[</span><span class="n">p</span><span class="p">:</span><span class="n">p</span><span class="o">+</span><span class="n">size</span><span class="p">]</span><span class="o">.</span><span class="n">reshape</span><span class="p">((</span><span class="n">lh</span><span class="p">,</span> <span class="n">lh</span><span class="p">))</span>
<span class="n">p</span> <span class="o">+=</span> <span class="n">size</span>
<span class="k">for</span> <span class="n">layer</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_layers</span><span class="p">):</span>
<span class="k">for</span> <span class="n">direction</span> <span class="ow">in</span> <span class="n">directions</span><span class="p">:</span>
<span class="k">for</span> <span class="n">gate</span> <span class="ow">in</span> <span class="n">gate_names</span><span class="p">:</span>
<span class="n">name</span> <span class="o">=</span> <span class="s1">'</span><span class="si">%s%s%d</span><span class="s1">_i2h</span><span class="si">%s</span><span class="s1">_bias'</span><span class="o">%</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="n">direction</span><span class="p">,</span> <span class="n">layer</span><span class="p">,</span> <span class="n">gate</span><span class="p">)</span>
<span class="n">args</span><span class="p">[</span><span class="n">name</span><span class="p">]</span> <span class="o">=</span> <span class="n">arr</span><span class="p">[</span><span class="n">p</span><span class="p">:</span><span class="n">p</span><span class="o">+</span><span class="n">lh</span><span class="p">]</span>
<span class="n">p</span> <span class="o">+=</span> <span class="n">lh</span>
<span class="k">for</span> <span class="n">gate</span> <span class="ow">in</span> <span class="n">gate_names</span><span class="p">:</span>
<span class="n">name</span> <span class="o">=</span> <span class="s1">'</span><span class="si">%s%s%d</span><span class="s1">_h2h</span><span class="si">%s</span><span class="s1">_bias'</span><span class="o">%</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="n">direction</span><span class="p">,</span> <span class="n">layer</span><span class="p">,</span> <span class="n">gate</span><span class="p">)</span>
<span class="n">args</span><span class="p">[</span><span class="n">name</span><span class="p">]</span> <span class="o">=</span> <span class="n">arr</span><span class="p">[</span><span class="n">p</span><span class="p">:</span><span class="n">p</span><span class="o">+</span><span class="n">lh</span><span class="p">]</span>
<span class="n">p</span> <span class="o">+=</span> <span class="n">lh</span>
<span class="k">assert</span> <span class="n">p</span> <span class="o">==</span> <span class="n">arr</span><span class="o">.</span><span class="n">size</span><span class="p">,</span> <span class="s2">"Invalid parameters size for FusedRNNCell"</span>
<span class="k">return</span> <span class="n">args</span>
<span class="k">def</span> <span class="nf">unpack_weights</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">args</span><span class="p">):</span>
<span class="n">args</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">copy</span><span class="p">()</span>
<span class="n">arr</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_parameter</span><span class="o">.</span><span class="n">name</span><span class="p">)</span>
<span class="n">b</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_directions</span><span class="p">)</span>
<span class="n">m</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_num_gates</span>
<span class="n">h</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span>
<span class="n">num_input</span> <span class="o">=</span> <span class="n">arr</span><span class="o">.</span><span class="n">size</span><span class="o">//</span><span class="n">b</span><span class="o">//</span><span class="n">h</span><span class="o">//</span><span class="n">m</span> <span class="o">-</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_layers</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span><span class="o">*</span><span class="p">(</span><span class="n">h</span><span class="o">+</span><span class="n">b</span><span class="o">*</span><span class="n">h</span><span class="o">+</span><span class="mi">2</span><span class="p">)</span> <span class="o">-</span> <span class="n">h</span> <span class="o">-</span> <span class="mi">2</span>
<span class="n">nargs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_slice_weights</span><span class="p">(</span><span class="n">arr</span><span class="p">,</span> <span class="n">num_input</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span><span class="p">)</span>
<span class="n">args</span><span class="o">.</span><span class="n">update</span><span class="p">({</span><span class="n">name</span><span class="p">:</span> <span class="n">nd</span><span class="o">.</span><span class="n">copy</span><span class="p">()</span> <span class="k">for</span> <span class="n">name</span><span class="p">,</span> <span class="n">nd</span> <span class="ow">in</span> <span class="n">nargs</span><span class="o">.</span><span class="n">items</span><span class="p">()})</span>
<span class="k">return</span> <span class="n">args</span>
<span class="k">def</span> <span class="nf">pack_weights</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">args</span><span class="p">):</span>
<span class="n">args</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">copy</span><span class="p">()</span>
<span class="n">b</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_bidirectional</span> <span class="o">+</span> <span class="mi">1</span>
<span class="n">m</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_num_gates</span>
<span class="n">c</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_gate_names</span>
<span class="n">h</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span>
<span class="n">w0</span> <span class="o">=</span> <span class="n">args</span><span class="p">[</span><span class="s1">'</span><span class="si">%s</span><span class="s1">l0_i2h</span><span class="si">%s</span><span class="s1">_weight'</span><span class="o">%</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="n">c</span><span class="p">[</span><span class="mi">0</span><span class="p">])]</span>
<span class="n">num_input</span> <span class="o">=</span> <span class="n">w0</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
<span class="n">total</span> <span class="o">=</span> <span class="p">(</span><span class="n">num_input</span><span class="o">+</span><span class="n">h</span><span class="o">+</span><span class="mi">2</span><span class="p">)</span><span class="o">*</span><span class="n">h</span><span class="o">*</span><span class="n">m</span><span class="o">*</span><span class="n">b</span> <span class="o">+</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_layers</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span><span class="o">*</span><span class="n">m</span><span class="o">*</span><span class="n">h</span><span class="o">*</span><span class="p">(</span><span class="n">h</span><span class="o">+</span><span class="n">b</span><span class="o">*</span><span class="n">h</span><span class="o">+</span><span class="mi">2</span><span class="p">)</span><span class="o">*</span><span class="n">b</span>
<span class="n">arr</span> <span class="o">=</span> <span class="n">ndarray</span><span class="o">.</span><span class="n">zeros</span><span class="p">((</span><span class="n">total</span><span class="p">,),</span> <span class="n">ctx</span><span class="o">=</span><span class="n">w0</span><span class="o">.</span><span class="n">context</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">w0</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
<span class="k">for</span> <span class="n">name</span><span class="p">,</span> <span class="n">nd</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_slice_weights</span><span class="p">(</span><span class="n">arr</span><span class="p">,</span> <span class="n">num_input</span><span class="p">,</span> <span class="n">h</span><span class="p">)</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
<span class="n">nd</span><span class="p">[:]</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="n">name</span><span class="p">)</span>
<span class="n">args</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">_parameter</span><span class="o">.</span><span class="n">name</span><span class="p">]</span> <span class="o">=</span> <span class="n">arr</span>
<span class="k">return</span> <span class="n">args</span>
<span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">states</span><span class="p">):</span>
<span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="s2">"FusedRNNCell cannot be stepped. Please use unroll"</span><span class="p">)</span>
<span class="k">def</span> <span class="nf">unroll</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">length</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">begin_state</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">layout</span><span class="o">=</span><span class="s1">'NTC'</span><span class="p">,</span> <span class="n">merge_outputs</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
<span class="bp">self</span><span class="o">.</span><span class="n">reset</span><span class="p">()</span>
<span class="n">inputs</span><span class="p">,</span> <span class="n">axis</span> <span class="o">=</span> <span class="n">_normalize_sequence</span><span class="p">(</span><span class="n">length</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">layout</span><span class="p">,</span> <span class="bp">True</span><span class="p">)</span>
<span class="k">if</span> <span class="n">axis</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
<span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">"NTC layout detected. Consider using "</span>
<span class="s2">"TNC for FusedRNNCell for faster speed"</span><span class="p">)</span>
<span class="n">inputs</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">swapaxes</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="n">dim1</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">dim2</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
<span class="k">else</span><span class="p">:</span>
<span class="k">assert</span> <span class="n">axis</span> <span class="o">==</span> <span class="mi">0</span><span class="p">,</span> <span class="s2">"Unsupported layout </span><span class="si">%s</span><span class="s2">"</span><span class="o">%</span><span class="n">layout</span>
<span class="k">if</span> <span class="n">begin_state</span> <span class="ow">is</span> <span class="bp">None</span><span class="p">:</span>
<span class="n">begin_state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">begin_state</span><span class="p">()</span>
<span class="n">states</span> <span class="o">=</span> <span class="n">begin_state</span>
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_mode</span> <span class="o">==</span> <span class="s1">'lstm'</span><span class="p">:</span>
<span class="n">states</span> <span class="o">=</span> <span class="p">{</span><span class="s1">'state'</span><span class="p">:</span> <span class="n">states</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="s1">'state_cell'</span><span class="p">:</span> <span class="n">states</span><span class="p">[</span><span class="mi">1</span><span class="p">]}</span> <span class="c1"># pylint: disable=redefined-variable-type</span>
<span class="k">else</span><span class="p">:</span>
<span class="n">states</span> <span class="o">=</span> <span class="p">{</span><span class="s1">'state'</span><span class="p">:</span> <span class="n">states</span><span class="p">[</span><span class="mi">0</span><span class="p">]}</span>
<span class="n">rnn</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">RNN</span><span class="p">(</span><span class="n">data</span><span class="o">=</span><span class="n">inputs</span><span class="p">,</span> <span class="n">parameters</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_parameter</span><span class="p">,</span>
<span class="n">state_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span><span class="p">,</span> <span class="n">num_layers</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_layers</span><span class="p">,</span>
<span class="n">bidirectional</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_bidirectional</span><span class="p">,</span> <span class="n">p</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_dropout</span><span class="p">,</span>
<span class="n">state_outputs</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_get_next_state</span><span class="p">,</span>
<span class="n">mode</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_mode</span><span class="p">,</span> <span class="n">name</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="o">+</span><span class="s1">'rnn'</span><span class="p">,</span>
<span class="o">**</span><span class="n">states</span><span class="p">)</span>
<span class="n">attr</span> <span class="o">=</span> <span class="p">{</span><span class="s1">'__layout__'</span> <span class="p">:</span> <span class="s1">'LNC'</span><span class="p">}</span>
<span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_next_state</span><span class="p">:</span>
<span class="n">outputs</span><span class="p">,</span> <span class="n">states</span> <span class="o">=</span> <span class="n">rnn</span><span class="p">,</span> <span class="p">[]</span>
<span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">_mode</span> <span class="o">==</span> <span class="s1">'lstm'</span><span class="p">:</span>
<span class="n">rnn</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">_set_attr</span><span class="p">(</span><span class="o">**</span><span class="n">attr</span><span class="p">)</span>
<span class="n">rnn</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span><span class="o">.</span><span class="n">_set_attr</span><span class="p">(</span><span class="o">**</span><span class="n">attr</span><span class="p">)</span>
<span class="n">outputs</span><span class="p">,</span> <span class="n">states</span> <span class="o">=</span> <span class="n">rnn</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="p">[</span><span class="n">rnn</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">rnn</span><span class="p">[</span><span class="mi">2</span><span class="p">]]</span>
<span class="k">else</span><span class="p">:</span>
<span class="n">rnn</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">_set_attr</span><span class="p">(</span><span class="o">**</span><span class="n">attr</span><span class="p">)</span>
<span class="n">outputs</span><span class="p">,</span> <span class="n">states</span> <span class="o">=</span> <span class="n">rnn</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="p">[</span><span class="n">rnn</span><span class="p">[</span><span class="mi">1</span><span class="p">]]</span>
<span class="k">if</span> <span class="n">axis</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
<span class="n">outputs</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">swapaxes</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="n">dim1</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">dim2</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
<span class="n">outputs</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">_normalize_sequence</span><span class="p">(</span><span class="n">length</span><span class="p">,</span> <span class="n">outputs</span><span class="p">,</span> <span class="n">layout</span><span class="p">,</span> <span class="n">merge_outputs</span><span class="p">)</span>
<span class="k">return</span> <span class="n">outputs</span><span class="p">,</span> <span class="n">states</span>
<div class="viewcode-block" id="FusedRNNCell.unfuse"><a class="viewcode-back" href="../../../api/python/symbol/rnn.html#mxnet.rnn.FusedRNNCell.unfuse">[docs]</a> <span class="k">def</span> <span class="nf">unfuse</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="sd">"""Unfuse the fused RNN in to a stack of rnn cells.</span>
<span class="sd"> Returns</span>
<span class="sd"> -------</span>
<span class="sd"> cell : SequentialRNNCell</span>
<span class="sd"> unfused cell that can be used for stepping, and can run on CPU.</span>
<span class="sd"> """</span>
<span class="n">stack</span> <span class="o">=</span> <span class="n">SequentialRNNCell</span><span class="p">()</span>
<span class="n">get_cell</span> <span class="o">=</span> <span class="p">{</span><span class="s1">'rnn_relu'</span><span class="p">:</span> <span class="k">lambda</span> <span class="n">cell_prefix</span><span class="p">:</span> <span class="n">RNNCell</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span><span class="p">,</span>
<span class="n">activation</span><span class="o">=</span><span class="s1">'relu'</span><span class="p">,</span>
<span class="n">prefix</span><span class="o">=</span><span class="n">cell_prefix</span><span class="p">),</span>
<span class="s1">'rnn_tanh'</span><span class="p">:</span> <span class="k">lambda</span> <span class="n">cell_prefix</span><span class="p">:</span> <span class="n">RNNCell</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span><span class="p">,</span>
<span class="n">activation</span><span class="o">=</span><span class="s1">'tanh'</span><span class="p">,</span>
<span class="n">prefix</span><span class="o">=</span><span class="n">cell_prefix</span><span class="p">),</span>
<span class="s1">'lstm'</span><span class="p">:</span> <span class="k">lambda</span> <span class="n">cell_prefix</span><span class="p">:</span> <span class="n">LSTMCell</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span><span class="p">,</span>
<span class="n">prefix</span><span class="o">=</span><span class="n">cell_prefix</span><span class="p">),</span>
<span class="s1">'gru'</span><span class="p">:</span> <span class="k">lambda</span> <span class="n">cell_prefix</span><span class="p">:</span> <span class="n">GRUCell</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span><span class="p">,</span>
<span class="n">prefix</span><span class="o">=</span><span class="n">cell_prefix</span><span class="p">)}[</span><span class="bp">self</span><span class="o">.</span><span class="n">_mode</span><span class="p">]</span>
<span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_layers</span><span class="p">):</span>
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_bidirectional</span><span class="p">:</span>
<span class="n">stack</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">BidirectionalCell</span><span class="p">(</span>
<span class="n">get_cell</span><span class="p">(</span><span class="s1">'</span><span class="si">%s</span><span class="s1">l</span><span class="si">%d</span><span class="s1">_'</span><span class="o">%</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="n">i</span><span class="p">)),</span>
<span class="n">get_cell</span><span class="p">(</span><span class="s1">'</span><span class="si">%s</span><span class="s1">r</span><span class="si">%d</span><span class="s1">_'</span><span class="o">%</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="n">i</span><span class="p">)),</span>
<span class="n">output_prefix</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">bi_l</span><span class="si">%d</span><span class="s1">_'</span><span class="o">%</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="n">i</span><span class="p">)))</span>
<span class="k">else</span><span class="p">:</span>
<span class="n">stack</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">get_cell</span><span class="p">(</span><span class="s1">'</span><span class="si">%s</span><span class="s1">l</span><span class="si">%d</span><span class="s1">_'</span><span class="o">%</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="n">i</span><span class="p">)))</span>
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_dropout</span> <span class="o">></span> <span class="mi">0</span> <span class="ow">and</span> <span class="n">i</span> <span class="o">!=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_num_layers</span> <span class="o">-</span> <span class="mi">1</span><span class="p">:</span>
<span class="n">stack</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">DropoutCell</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_dropout</span><span class="p">,</span> <span class="n">prefix</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">_dropout</span><span class="si">%d</span><span class="s1">_'</span><span class="o">%</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="n">i</span><span class="p">)))</span>
<span class="k">return</span> <span class="n">stack</span></div></div>
<div class="viewcode-block" id="SequentialRNNCell"><a class="viewcode-back" href="../../../api/python/symbol/rnn.html#mxnet.rnn.SequentialRNNCell">[docs]</a><span class="k">class</span> <span class="nc">SequentialRNNCell</span><span class="p">(</span><span class="n">BaseRNNCell</span><span class="p">):</span>
<span class="sd">"""Sequantially stacking multiple RNN cells.</span>
<span class="sd"> Parameters</span>
<span class="sd"> ----------</span>
<span class="sd"> params : RNNParams, default None</span>
<span class="sd"> Container for weight sharing between cells. Created if None.</span>
<span class="sd"> """</span>
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">params</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
<span class="nb">super</span><span class="p">(</span><span class="n">SequentialRNNCell</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">prefix</span><span class="o">=</span><span class="s1">''</span><span class="p">,</span> <span class="n">params</span><span class="o">=</span><span class="n">params</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_override_cell_params</span> <span class="o">=</span> <span class="n">params</span> <span class="ow">is</span> <span class="ow">not</span> <span class="bp">None</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_cells</span> <span class="o">=</span> <span class="p">[]</span>
<div class="viewcode-block" id="SequentialRNNCell.add"><a class="viewcode-back" href="../../../api/python/symbol/rnn.html#mxnet.rnn.SequentialRNNCell.add">[docs]</a> <span class="k">def</span> <span class="nf">add</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">cell</span><span class="p">):</span>
<span class="sd">"""Append a cell into the stack.</span>
<span class="sd"> Parameters</span>
<span class="sd"> ----------</span>
<span class="sd"> cell : BaseRNNCell</span>
<span class="sd"> The cell to be appended. During unroll, previous cell's output (or raw inputs if</span>
<span class="sd"> no previous cell) is used as the input to this cell.</span>
<span class="sd"> """</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_cells</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">cell</span><span class="p">)</span>
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_override_cell_params</span><span class="p">:</span>
<span class="k">assert</span> <span class="n">cell</span><span class="o">.</span><span class="n">_own_params</span><span class="p">,</span> \
<span class="s2">"Either specify params for SequentialRNNCell "</span> \
<span class="s2">"or child cells, not both."</span>
<span class="n">cell</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">_params</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">_params</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">_params</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">cell</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">_params</span><span class="p">)</span></div>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">state_info</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">return</span> <span class="n">_cells_state_info</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_cells</span><span class="p">)</span>
<span class="k">def</span> <span class="nf">begin_state</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span> <span class="c1"># pylint: disable=arguments-differ</span>
<span class="k">assert</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_modified</span><span class="p">,</span> \
<span class="s2">"After applying modifier cells (e.g. ZoneoutCell) the base "</span> \
<span class="s2">"cell cannot be called directly. Call the modifier cell instead."</span>
<span class="k">return</span> <span class="n">_cells_begin_state</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_cells</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
<span class="k">def</span> <span class="nf">unpack_weights</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">args</span><span class="p">):</span>
<span class="k">return</span> <span class="n">_cells_unpack_weights</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_cells</span><span class="p">,</span> <span class="n">args</span><span class="p">)</span>
<span class="k">def</span> <span class="nf">pack_weights</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">args</span><span class="p">):</span>
<span class="k">return</span> <span class="n">_cells_pack_weights</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_cells</span><span class="p">,</span> <span class="n">args</span><span class="p">)</span>
<span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">states</span><span class="p">):</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_counter</span> <span class="o">+=</span> <span class="mi">1</span>
<span class="n">next_states</span> <span class="o">=</span> <span class="p">[]</span>
<span class="n">p</span> <span class="o">=</span> <span class="mi">0</span>
<span class="k">for</span> <span class="n">cell</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_cells</span><span class="p">:</span>
<span class="k">assert</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">cell</span><span class="p">,</span> <span class="n">BidirectionalCell</span><span class="p">)</span>
<span class="n">n</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">cell</span><span class="o">.</span><span class="n">state_info</span><span class="p">)</span>
<span class="n">state</span> <span class="o">=</span> <span class="n">states</span><span class="p">[</span><span class="n">p</span><span class="p">:</span><span class="n">p</span><span class="o">+</span><span class="n">n</span><span class="p">]</span>
<span class="n">p</span> <span class="o">+=</span> <span class="n">n</span>
<span class="n">inputs</span><span class="p">,</span> <span class="n">state</span> <span class="o">=</span> <span class="n">cell</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="n">state</span><span class="p">)</span>
<span class="n">next_states</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">state</span><span class="p">)</span>
<span class="k">return</span> <span class="n">inputs</span><span class="p">,</span> <span class="nb">sum</span><span class="p">(</span><span class="n">next_states</span><span class="p">,</span> <span class="p">[])</span>
<span class="k">def</span> <span class="nf">unroll</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">length</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">begin_state</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">layout</span><span class="o">=</span><span class="s1">'NTC'</span><span class="p">,</span> <span class="n">merge_outputs</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
<span class="bp">self</span><span class="o">.</span><span class="n">reset</span><span class="p">()</span>
<span class="n">num_cells</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_cells</span><span class="p">)</span>
<span class="k">if</span> <span class="n">begin_state</span> <span class="ow">is</span> <span class="bp">None</span><span class="p">:</span>
<span class="n">begin_state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">begin_state</span><span class="p">()</span>
<span class="n">p</span> <span class="o">=</span> <span class="mi">0</span>
<span class="n">next_states</span> <span class="o">=</span> <span class="p">[]</span>
<span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">cell</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_cells</span><span class="p">):</span>
<span class="n">n</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">cell</span><span class="o">.</span><span class="n">state_info</span><span class="p">)</span>
<span class="n">states</span> <span class="o">=</span> <span class="n">begin_state</span><span class="p">[</span><span class="n">p</span><span class="p">:</span><span class="n">p</span><span class="o">+</span><span class="n">n</span><span class="p">]</span>
<span class="n">p</span> <span class="o">+=</span> <span class="n">n</span>
<span class="n">inputs</span><span class="p">,</span> <span class="n">states</span> <span class="o">=</span> <span class="n">cell</span><span class="o">.</span><span class="n">unroll</span><span class="p">(</span><span class="n">length</span><span class="p">,</span> <span class="n">inputs</span><span class="o">=</span><span class="n">inputs</span><span class="p">,</span> <span class="n">begin_state</span><span class="o">=</span><span class="n">states</span><span class="p">,</span> <span class="n">layout</span><span class="o">=</span><span class="n">layout</span><span class="p">,</span>
<span class="n">merge_outputs</span><span class="o">=</span><span class="bp">None</span> <span class="k">if</span> <span class="n">i</span> <span class="o"><</span> <span class="n">num_cells</span><span class="o">-</span><span class="mi">1</span> <span class="k">else</span> <span class="n">merge_outputs</span><span class="p">)</span>
<span class="n">next_states</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">states</span><span class="p">)</span>
<span class="k">return</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">next_states</span></div>
<div class="viewcode-block" id="DropoutCell"><a class="viewcode-back" href="../../../api/python/symbol/rnn.html#mxnet.rnn.DropoutCell">[docs]</a><span class="k">class</span> <span class="nc">DropoutCell</span><span class="p">(</span><span class="n">BaseRNNCell</span><span class="p">):</span>
<span class="sd">"""Apply dropout on input.</span>
<span class="sd"> Parameters</span>
<span class="sd"> ----------</span>
<span class="sd"> dropout : float</span>
<span class="sd"> Percentage of elements to drop out, which</span>
<span class="sd"> is 1 - percentage to retain.</span>
<span class="sd"> prefix : str, default 'dropout_'</span>
<span class="sd"> Prefix for names of layers</span>
<span class="sd"> (this prefix is also used for names of weights if `params` is None</span>
<span class="sd"> i.e. if `params` are being created and not reused)</span>
<span class="sd"> params : RNNParams, default None</span>
<span class="sd"> Container for weight sharing between cells. Created if None.</span>
<span class="sd"> """</span>
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">dropout</span><span class="p">,</span> <span class="n">prefix</span><span class="o">=</span><span class="s1">'dropout_'</span><span class="p">,</span> <span class="n">params</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
<span class="nb">super</span><span class="p">(</span><span class="n">DropoutCell</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">prefix</span><span class="p">,</span> <span class="n">params</span><span class="p">)</span>
<span class="k">assert</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">dropout</span><span class="p">,</span> <span class="n">numeric_types</span><span class="p">),</span> <span class="s2">"dropout probability must be a number"</span>
<span class="bp">self</span><span class="o">.</span><span class="n">dropout</span> <span class="o">=</span> <span class="n">dropout</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">state_info</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">return</span> <span class="p">[]</span>
<span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">states</span><span class="p">):</span>
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">dropout</span> <span class="o">></span> <span class="mi">0</span><span class="p">:</span>
<span class="n">inputs</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">data</span><span class="o">=</span><span class="n">inputs</span><span class="p">,</span> <span class="n">p</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">dropout</span><span class="p">)</span>
<span class="k">return</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">states</span>
<span class="k">def</span> <span class="nf">unroll</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">length</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">begin_state</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">layout</span><span class="o">=</span><span class="s1">'NTC'</span><span class="p">,</span> <span class="n">merge_outputs</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
<span class="bp">self</span><span class="o">.</span><span class="n">reset</span><span class="p">()</span>
<span class="n">inputs</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">_normalize_sequence</span><span class="p">(</span><span class="n">length</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">layout</span><span class="p">,</span> <span class="n">merge_outputs</span><span class="p">)</span>
<span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Symbol</span><span class="p">):</span>
<span class="k">return</span> <span class="bp">self</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="p">[])</span>
<span class="k">else</span><span class="p">:</span>
<span class="k">return</span> <span class="nb">super</span><span class="p">(</span><span class="n">DropoutCell</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="n">unroll</span><span class="p">(</span>
<span class="n">length</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">begin_state</span><span class="o">=</span><span class="n">begin_state</span><span class="p">,</span> <span class="n">layout</span><span class="o">=</span><span class="n">layout</span><span class="p">,</span>
<span class="n">merge_outputs</span><span class="o">=</span><span class="n">merge_outputs</span><span class="p">)</span></div>
<span class="k">class</span> <span class="nc">ModifierCell</span><span class="p">(</span><span class="n">BaseRNNCell</span><span class="p">):</span>
<span class="sd">"""Base class for modifier cells. A modifier</span>
<span class="sd"> cell takes a base cell, apply modifications</span>
<span class="sd"> on it (e.g. Zoneout), and returns a new cell.</span>
<span class="sd"> After applying modifiers the base cell should</span>
<span class="sd"> no longer be called directly. The modifer cell</span>
<span class="sd"> should be used instead.</span>
<span class="sd"> """</span>
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">base_cell</span><span class="p">):</span>
<span class="nb">super</span><span class="p">(</span><span class="n">ModifierCell</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
<span class="n">base_cell</span><span class="o">.</span><span class="n">_modified</span> <span class="o">=</span> <span class="bp">True</span>
<span class="bp">self</span><span class="o">.</span><span class="n">base_cell</span> <span class="o">=</span> <span class="n">base_cell</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">params</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_own_params</span> <span class="o">=</span> <span class="bp">False</span>
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_cell</span><span class="o">.</span><span class="n">params</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">state_info</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_cell</span><span class="o">.</span><span class="n">state_info</span>
<span class="k">def</span> <span class="nf">begin_state</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">init_sym</span><span class="o">=</span><span class="n">symbol</span><span class="o">.</span><span class="n">zeros</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span> <span class="c1"># pylint: disable=arguments-differ</span>
<span class="k">assert</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_modified</span><span class="p">,</span> \
<span class="s2">"After applying modifier cells (e.g. DropoutCell) the base "</span> \
<span class="s2">"cell cannot be called directly. Call the modifier cell instead."</span>
<span class="bp">self</span><span class="o">.</span><span class="n">base_cell</span><span class="o">.</span><span class="n">_modified</span> <span class="o">=</span> <span class="bp">False</span>
<span class="n">begin</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_cell</span><span class="o">.</span><span class="n">begin_state</span><span class="p">(</span><span class="n">init_sym</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">base_cell</span><span class="o">.</span><span class="n">_modified</span> <span class="o">=</span> <span class="bp">True</span>
<span class="k">return</span> <span class="n">begin</span>
<span class="k">def</span> <span class="nf">unpack_weights</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">args</span><span class="p">):</span>
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_cell</span><span class="o">.</span><span class="n">unpack_weights</span><span class="p">(</span><span class="n">args</span><span class="p">)</span>
<span class="k">def</span> <span class="nf">pack_weights</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">args</span><span class="p">):</span>
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_cell</span><span class="o">.</span><span class="n">pack_weights</span><span class="p">(</span><span class="n">args</span><span class="p">)</span>
<span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">states</span><span class="p">):</span>
<span class="k">raise</span> <span class="ne">NotImplementedError</span>
<div class="viewcode-block" id="ZoneoutCell"><a class="viewcode-back" href="../../../api/python/symbol/rnn.html#mxnet.rnn.ZoneoutCell">[docs]</a><span class="k">class</span> <span class="nc">ZoneoutCell</span><span class="p">(</span><span class="n">ModifierCell</span><span class="p">):</span>
<span class="sd">"""Apply Zoneout on base cell.</span>
<span class="sd"> Parameters</span>
<span class="sd"> ----------</span>
<span class="sd"> base_cell : BaseRNNCell</span>
<span class="sd"> Cell on whose states to perform zoneout.</span>
<span class="sd"> zoneout_outputs : float, default 0.</span>
<span class="sd"> Fraction of the output that gets dropped out during training time.</span>
<span class="sd"> zoneout_states : float, default 0.</span>
<span class="sd"> Fraction of the states that gets dropped out during training time.</span>
<span class="sd"> """</span>
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">base_cell</span><span class="p">,</span> <span class="n">zoneout_outputs</span><span class="o">=</span><span class="mf">0.</span><span class="p">,</span> <span class="n">zoneout_states</span><span class="o">=</span><span class="mf">0.</span><span class="p">):</span>
<span class="k">assert</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">base_cell</span><span class="p">,</span> <span class="n">FusedRNNCell</span><span class="p">),</span> \
<span class="s2">"FusedRNNCell doesn't support zoneout. "</span> \
<span class="s2">"Please unfuse first."</span>
<span class="k">assert</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">base_cell</span><span class="p">,</span> <span class="n">BidirectionalCell</span><span class="p">),</span> \
<span class="s2">"BidirectionalCell doesn't support zoneout since it doesn't support step. "</span> \
<span class="s2">"Please add ZoneoutCell to the cells underneath instead."</span>
<span class="k">assert</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">base_cell</span><span class="p">,</span> <span class="n">SequentialRNNCell</span><span class="p">)</span> <span class="ow">or</span> <span class="ow">not</span> <span class="n">base_cell</span><span class="o">.</span><span class="n">_bidirectional</span><span class="p">,</span> \
<span class="s2">"Bidirectional SequentialRNNCell doesn't support zoneout. "</span> \
<span class="s2">"Please add ZoneoutCell to the cells underneath instead."</span>
<span class="nb">super</span><span class="p">(</span><span class="n">ZoneoutCell</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">base_cell</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">zoneout_outputs</span> <span class="o">=</span> <span class="n">zoneout_outputs</span>
<span class="bp">self</span><span class="o">.</span><span class="n">zoneout_states</span> <span class="o">=</span> <span class="n">zoneout_states</span>
<span class="bp">self</span><span class="o">.</span><span class="n">prev_output</span> <span class="o">=</span> <span class="bp">None</span>
<span class="k">def</span> <span class="nf">reset</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="nb">super</span><span class="p">(</span><span class="n">ZoneoutCell</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="n">reset</span><span class="p">()</span>
<span class="bp">self</span><span class="o">.</span><span class="n">prev_output</span> <span class="o">=</span> <span class="bp">None</span>
<span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">states</span><span class="p">):</span>
<span class="n">cell</span><span class="p">,</span> <span class="n">p_outputs</span><span class="p">,</span> <span class="n">p_states</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_cell</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">zoneout_outputs</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">zoneout_states</span>
<span class="n">next_output</span><span class="p">,</span> <span class="n">next_states</span> <span class="o">=</span> <span class="n">cell</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="n">states</span><span class="p">)</span>
<span class="n">mask</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">p</span><span class="p">,</span> <span class="n">like</span><span class="p">:</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">symbol</span><span class="o">.</span><span class="n">ones_like</span><span class="p">(</span><span class="n">like</span><span class="p">),</span> <span class="n">p</span><span class="o">=</span><span class="n">p</span><span class="p">)</span>
<span class="n">prev_output</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">prev_output</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">prev_output</span> <span class="ow">is</span> <span class="ow">not</span> <span class="bp">None</span> <span class="k">else</span> <span class="n">symbol</span><span class="o">.</span><span class="n">zeros</span><span class="p">((</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">))</span>
<span class="n">output</span> <span class="o">=</span> <span class="p">(</span><span class="n">symbol</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">mask</span><span class="p">(</span><span class="n">p_outputs</span><span class="p">,</span> <span class="n">next_output</span><span class="p">),</span> <span class="n">next_output</span><span class="p">,</span> <span class="n">prev_output</span><span class="p">)</span>
<span class="k">if</span> <span class="n">p_outputs</span> <span class="o">!=</span> <span class="mf">0.</span> <span class="k">else</span> <span class="n">next_output</span><span class="p">)</span>
<span class="n">states</span> <span class="o">=</span> <span class="p">([</span><span class="n">symbol</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">mask</span><span class="p">(</span><span class="n">p_states</span><span class="p">,</span> <span class="n">new_s</span><span class="p">),</span> <span class="n">new_s</span><span class="p">,</span> <span class="n">old_s</span><span class="p">)</span> <span class="k">for</span> <span class="n">new_s</span><span class="p">,</span> <span class="n">old_s</span> <span class="ow">in</span>
<span class="nb">zip</span><span class="p">(</span><span class="n">next_states</span><span class="p">,</span> <span class="n">states</span><span class="p">)]</span> <span class="k">if</span> <span class="n">p_states</span> <span class="o">!=</span> <span class="mf">0.</span> <span class="k">else</span> <span class="n">next_states</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">prev_output</span> <span class="o">=</span> <span class="n">output</span>
<span class="k">return</span> <span class="n">output</span><span class="p">,</span> <span class="n">states</span></div>
<div class="viewcode-block" id="ResidualCell"><a class="viewcode-back" href="../../../api/python/symbol/rnn.html#mxnet.rnn.ResidualCell">[docs]</a><span class="k">class</span> <span class="nc">ResidualCell</span><span class="p">(</span><span class="n">ModifierCell</span><span class="p">):</span>
<span class="sd">"""Adds residual connection as described in Wu et al, 2016</span>
<span class="sd"> (https://arxiv.org/abs/1609.08144).</span>
<span class="sd"> Output of the cell is output of the base cell plus input.</span>
<span class="sd"> Parameters</span>
<span class="sd"> ----------</span>
<span class="sd"> base_cell : BaseRNNCell</span>
<span class="sd"> Cell on whose outputs to add residual connection.</span>
<span class="sd"> """</span>
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">base_cell</span><span class="p">):</span>
<span class="nb">super</span><span class="p">(</span><span class="n">ResidualCell</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">base_cell</span><span class="p">)</span>
<span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">states</span><span class="p">):</span>
<span class="n">output</span><span class="p">,</span> <span class="n">states</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_cell</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="n">states</span><span class="p">)</span>
<span class="n">output</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">elemwise_add</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">name</span><span class="o">=</span><span class="s2">"</span><span class="si">%s</span><span class="s2">_plus_residual"</span> <span class="o">%</span> <span class="n">output</span><span class="o">.</span><span class="n">name</span><span class="p">)</span>
<span class="k">return</span> <span class="n">output</span><span class="p">,</span> <span class="n">states</span>
<span class="k">def</span> <span class="nf">unroll</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">length</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">begin_state</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">layout</span><span class="o">=</span><span class="s1">'NTC'</span><span class="p">,</span> <span class="n">merge_outputs</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
<span class="bp">self</span><span class="o">.</span><span class="n">reset</span><span class="p">()</span>
<span class="bp">self</span><span class="o">.</span><span class="n">base_cell</span><span class="o">.</span><span class="n">_modified</span> <span class="o">=</span> <span class="bp">False</span>
<span class="n">outputs</span><span class="p">,</span> <span class="n">states</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_cell</span><span class="o">.</span><span class="n">unroll</span><span class="p">(</span><span class="n">length</span><span class="p">,</span> <span class="n">inputs</span><span class="o">=</span><span class="n">inputs</span><span class="p">,</span> <span class="n">begin_state</span><span class="o">=</span><span class="n">begin_state</span><span class="p">,</span>
<span class="n">layout</span><span class="o">=</span><span class="n">layout</span><span class="p">,</span> <span class="n">merge_outputs</span><span class="o">=</span><span class="n">merge_outputs</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">base_cell</span><span class="o">.</span><span class="n">_modified</span> <span class="o">=</span> <span class="bp">True</span>
<span class="n">merge_outputs</span> <span class="o">=</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Symbol</span><span class="p">)</span> <span class="k">if</span> <span class="n">merge_outputs</span> <span class="ow">is</span> <span class="bp">None</span> <span class="k">else</span> \
<span class="n">merge_outputs</span>
<span class="n">inputs</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">_normalize_sequence</span><span class="p">(</span><span class="n">length</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">layout</span><span class="p">,</span> <span class="n">merge_outputs</span><span class="p">)</span>
<span class="k">if</span> <span class="n">merge_outputs</span><span class="p">:</span>
<span class="n">outputs</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">elemwise_add</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">name</span><span class="o">=</span><span class="s2">"</span><span class="si">%s</span><span class="s2">_plus_residual"</span> <span class="o">%</span> <span class="n">outputs</span><span class="o">.</span><span class="n">name</span><span class="p">)</span>
<span class="k">else</span><span class="p">:</span>
<span class="n">outputs</span> <span class="o">=</span> <span class="p">[</span><span class="n">symbol</span><span class="o">.</span><span class="n">elemwise_add</span><span class="p">(</span><span class="n">output_sym</span><span class="p">,</span> <span class="n">input_sym</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s2">"</span><span class="si">%s</span><span class="s2">_plus_residual"</span> <span class="o">%</span> <span class="n">output_sym</span><span class="o">.</span><span class="n">name</span><span class="p">)</span>
<span class="k">for</span> <span class="n">output_sym</span><span class="p">,</span> <span class="n">input_sym</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="n">inputs</span><span class="p">)]</span>
<span class="k">return</span> <span class="n">outputs</span><span class="p">,</span> <span class="n">states</span></div>
<div class="viewcode-block" id="BidirectionalCell"><a class="viewcode-back" href="../../../api/python/symbol/rnn.html#mxnet.rnn.BidirectionalCell">[docs]</a><span class="k">class</span> <span class="nc">BidirectionalCell</span><span class="p">(</span><span class="n">BaseRNNCell</span><span class="p">):</span>
<span class="sd">"""Bidirectional RNN cell.</span>
<span class="sd"> Parameters</span>
<span class="sd"> ----------</span>
<span class="sd"> l_cell : BaseRNNCell</span>
<span class="sd"> cell for forward unrolling</span>
<span class="sd"> r_cell : BaseRNNCell</span>
<span class="sd"> cell for backward unrolling</span>
<span class="sd"> params : RNNParams, default None.</span>
<span class="sd"> Container for weight sharing between cells.</span>
<span class="sd"> A new RNNParams container is created if `params` is None.</span>
<span class="sd"> output_prefix : str, default 'bi_'</span>
<span class="sd"> prefix for name of output</span>
<span class="sd"> """</span>
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">l_cell</span><span class="p">,</span> <span class="n">r_cell</span><span class="p">,</span> <span class="n">params</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">output_prefix</span><span class="o">=</span><span class="s1">'bi_'</span><span class="p">):</span>
<span class="nb">super</span><span class="p">(</span><span class="n">BidirectionalCell</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="s1">''</span><span class="p">,</span> <span class="n">params</span><span class="o">=</span><span class="n">params</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_output_prefix</span> <span class="o">=</span> <span class="n">output_prefix</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_override_cell_params</span> <span class="o">=</span> <span class="n">params</span> <span class="ow">is</span> <span class="ow">not</span> <span class="bp">None</span>
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_override_cell_params</span><span class="p">:</span>
<span class="k">assert</span> <span class="n">l_cell</span><span class="o">.</span><span class="n">_own_params</span> <span class="ow">and</span> <span class="n">r_cell</span><span class="o">.</span><span class="n">_own_params</span><span class="p">,</span> \
<span class="s2">"Either specify params for BidirectionalCell "</span> \
<span class="s2">"or child cells, not both."</span>
<span class="n">l_cell</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">_params</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">_params</span><span class="p">)</span>
<span class="n">r_cell</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">_params</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">_params</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">_params</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">l_cell</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">_params</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">_params</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">r_cell</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">_params</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_cells</span> <span class="o">=</span> <span class="p">[</span><span class="n">l_cell</span><span class="p">,</span> <span class="n">r_cell</span><span class="p">]</span>
<span class="k">def</span> <span class="nf">unpack_weights</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">args</span><span class="p">):</span>
<span class="k">return</span> <span class="n">_cells_unpack_weights</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_cells</span><span class="p">,</span> <span class="n">args</span><span class="p">)</span>
<span class="k">def</span> <span class="nf">pack_weights</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">args</span><span class="p">):</span>
<span class="k">return</span> <span class="n">_cells_pack_weights</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_cells</span><span class="p">,</span> <span class="n">args</span><span class="p">)</span>
<span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">states</span><span class="p">):</span>
<span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="s2">"Bidirectional cannot be stepped. Please use unroll"</span><span class="p">)</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">state_info</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">return</span> <span class="n">_cells_state_info</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_cells</span><span class="p">)</span>
<span class="k">def</span> <span class="nf">begin_state</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span> <span class="c1"># pylint: disable=arguments-differ</span>
<span class="k">assert</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_modified</span><span class="p">,</span> \
<span class="s2">"After applying modifier cells (e.g. DropoutCell) the base "</span> \
<span class="s2">"cell cannot be called directly. Call the modifier cell instead."</span>
<span class="k">return</span> <span class="n">_cells_begin_state</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_cells</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
<span class="k">def</span> <span class="nf">unroll</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">length</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">begin_state</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">layout</span><span class="o">=</span><span class="s1">'NTC'</span><span class="p">,</span> <span class="n">merge_outputs</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
<span class="bp">self</span><span class="o">.</span><span class="n">reset</span><span class="p">()</span>
<span class="n">inputs</span><span class="p">,</span> <span class="n">axis</span> <span class="o">=</span> <span class="n">_normalize_sequence</span><span class="p">(</span><span class="n">length</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">layout</span><span class="p">,</span> <span class="bp">False</span><span class="p">)</span>
<span class="k">if</span> <span class="n">begin_state</span> <span class="ow">is</span> <span class="bp">None</span><span class="p">:</span>
<span class="n">begin_state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">begin_state</span><span class="p">()</span>
<span class="n">states</span> <span class="o">=</span> <span class="n">begin_state</span>
<span class="n">l_cell</span><span class="p">,</span> <span class="n">r_cell</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_cells</span>
<span class="n">l_outputs</span><span class="p">,</span> <span class="n">l_states</span> <span class="o">=</span> <span class="n">l_cell</span><span class="o">.</span><span class="n">unroll</span><span class="p">(</span><span class="n">length</span><span class="p">,</span> <span class="n">inputs</span><span class="o">=</span><span class="n">inputs</span><span class="p">,</span>
<span class="n">begin_state</span><span class="o">=</span><span class="n">states</span><span class="p">[:</span><span class="nb">len</span><span class="p">(</span><span class="n">l_cell</span><span class="o">.</span><span class="n">state_info</span><span class="p">)],</span>
<span class="n">layout</span><span class="o">=</span><span class="n">layout</span><span class="p">,</span> <span class="n">merge_outputs</span><span class="o">=</span><span class="n">merge_outputs</span><span class="p">)</span>
<span class="n">r_outputs</span><span class="p">,</span> <span class="n">r_states</span> <span class="o">=</span> <span class="n">r_cell</span><span class="o">.</span><span class="n">unroll</span><span class="p">(</span><span class="n">length</span><span class="p">,</span>
<span class="n">inputs</span><span class="o">=</span><span class="nb">list</span><span class="p">(</span><span class="nb">reversed</span><span class="p">(</span><span class="n">inputs</span><span class="p">)),</span>
<span class="n">begin_state</span><span class="o">=</span><span class="n">states</span><span class="p">[</span><span class="nb">len</span><span class="p">(</span><span class="n">l_cell</span><span class="o">.</span><span class="n">state_info</span><span class="p">):],</span>
<span class="n">layout</span><span class="o">=</span><span class="n">layout</span><span class="p">,</span> <span class="n">merge_outputs</span><span class="o">=</span><span class="n">merge_outputs</span><span class="p">)</span>
<span class="k">if</span> <span class="n">merge_outputs</span> <span class="ow">is</span> <span class="bp">None</span><span class="p">:</span>
<span class="n">merge_outputs</span> <span class="o">=</span> <span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">l_outputs</span><span class="p">,</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Symbol</span><span class="p">)</span>
<span class="ow">and</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">r_outputs</span><span class="p">,</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Symbol</span><span class="p">))</span>
<span class="k">if</span> <span class="ow">not</span> <span class="n">merge_outputs</span><span class="p">:</span>
<span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">l_outputs</span><span class="p">,</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Symbol</span><span class="p">):</span>
<span class="n">l_outputs</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">symbol</span><span class="o">.</span><span class="n">SliceChannel</span><span class="p">(</span><span class="n">l_outputs</span><span class="p">,</span> <span class="n">axis</span><span class="o">=</span><span class="n">axis</span><span class="p">,</span>
<span class="n">num_outputs</span><span class="o">=</span><span class="n">length</span><span class="p">,</span> <span class="n">squeeze_axis</span><span class="o">=</span><span class="mi">1</span><span class="p">))</span>
<span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">r_outputs</span><span class="p">,</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Symbol</span><span class="p">):</span>
<span class="n">r_outputs</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">symbol</span><span class="o">.</span><span class="n">SliceChannel</span><span class="p">(</span><span class="n">r_outputs</span><span class="p">,</span> <span class="n">axis</span><span class="o">=</span><span class="n">axis</span><span class="p">,</span>
<span class="n">num_outputs</span><span class="o">=</span><span class="n">length</span><span class="p">,</span> <span class="n">squeeze_axis</span><span class="o">=</span><span class="mi">1</span><span class="p">))</span>
<span class="k">if</span> <span class="n">merge_outputs</span><span class="p">:</span>
<span class="n">l_outputs</span> <span class="o">=</span> <span class="p">[</span><span class="n">l_outputs</span><span class="p">]</span>
<span class="n">r_outputs</span> <span class="o">=</span> <span class="p">[</span><span class="n">symbol</span><span class="o">.</span><span class="n">reverse</span><span class="p">(</span><span class="n">r_outputs</span><span class="p">,</span> <span class="n">axis</span><span class="o">=</span><span class="n">axis</span><span class="p">)]</span>
<span class="k">else</span><span class="p">:</span>
<span class="n">r_outputs</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="nb">reversed</span><span class="p">(</span><span class="n">r_outputs</span><span class="p">))</span>
<span class="n">outputs</span> <span class="o">=</span> <span class="p">[</span><span class="n">symbol</span><span class="o">.</span><span class="n">Concat</span><span class="p">(</span><span class="n">l_o</span><span class="p">,</span> <span class="n">r_o</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="o">+</span><span class="n">merge_outputs</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="p">(</span><span class="s1">'</span><span class="si">%s</span><span class="s1">out'</span><span class="o">%</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_output_prefix</span><span class="p">)</span> <span class="k">if</span> <span class="n">merge_outputs</span>
<span class="k">else</span> <span class="s1">'</span><span class="si">%s</span><span class="s1">t</span><span class="si">%d</span><span class="s1">'</span><span class="o">%</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_output_prefix</span><span class="p">,</span> <span class="n">i</span><span class="p">)))</span>
<span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">l_o</span><span class="p">,</span> <span class="n">r_o</span> <span class="ow">in</span>
<span class="nb">zip</span><span class="p">(</span><span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">l_outputs</span><span class="p">)),</span> <span class="n">l_outputs</span><span class="p">,</span> <span class="n">r_outputs</span><span class="p">)]</span>
<span class="k">if</span> <span class="n">merge_outputs</span><span class="p">:</span>
<span class="n">outputs</span> <span class="o">=</span> <span class="n">outputs</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
<span class="n">states</span> <span class="o">=</span> <span class="p">[</span><span class="n">l_states</span><span class="p">,</span> <span class="n">r_states</span><span class="p">]</span>
<span class="k">return</span> <span class="n">outputs</span><span class="p">,</span> <span class="n">states</span></div>
<span class="k">class</span> <span class="nc">BaseConvRNNCell</span><span class="p">(</span><span class="n">BaseRNNCell</span><span class="p">):</span>
<span class="sd">"""Abstract base class for Convolutional RNN cells"""</span>
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">input_shape</span><span class="p">,</span> <span class="n">num_hidden</span><span class="p">,</span>
<span class="n">h2h_kernel</span><span class="p">,</span> <span class="n">h2h_dilate</span><span class="p">,</span>
<span class="n">i2h_kernel</span><span class="p">,</span> <span class="n">i2h_stride</span><span class="p">,</span>
<span class="n">i2h_pad</span><span class="p">,</span> <span class="n">i2h_dilate</span><span class="p">,</span>
<span class="n">i2h_weight_initializer</span><span class="p">,</span> <span class="n">h2h_weight_initializer</span><span class="p">,</span>
<span class="n">i2h_bias_initializer</span><span class="p">,</span> <span class="n">h2h_bias_initializer</span><span class="p">,</span>
<span class="n">activation</span><span class="p">,</span> <span class="n">prefix</span><span class="o">=</span><span class="s1">''</span><span class="p">,</span> <span class="n">params</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">conv_layout</span><span class="o">=</span><span class="s1">'NCHW'</span><span class="p">):</span>
<span class="nb">super</span><span class="p">(</span><span class="n">BaseConvRNNCell</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">prefix</span><span class="o">=</span><span class="n">prefix</span><span class="p">,</span> <span class="n">params</span><span class="o">=</span><span class="n">params</span><span class="p">)</span>
<span class="c1"># Convolution setting</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_h2h_kernel</span> <span class="o">=</span> <span class="n">h2h_kernel</span>
<span class="k">assert</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_h2h_kernel</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">%</span> <span class="mi">2</span> <span class="o">==</span> <span class="mi">1</span><span class="p">)</span> <span class="ow">and</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_h2h_kernel</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">%</span> <span class="mi">2</span> <span class="o">==</span> <span class="mi">1</span><span class="p">),</span> \
<span class="s2">"Only support odd number, get h2h_kernel= </span><span class="si">%s</span><span class="s2">"</span> <span class="o">%</span> <span class="nb">str</span><span class="p">(</span><span class="n">h2h_kernel</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_h2h_pad</span> <span class="o">=</span> <span class="p">(</span><span class="n">h2h_dilate</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="p">(</span><span class="n">h2h_kernel</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> <span class="o">//</span> <span class="mi">2</span><span class="p">,</span>
<span class="n">h2h_dilate</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="p">(</span><span class="n">h2h_kernel</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> <span class="o">//</span> <span class="mi">2</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_h2h_dilate</span> <span class="o">=</span> <span class="n">h2h_dilate</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_i2h_kernel</span> <span class="o">=</span> <span class="n">i2h_kernel</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_i2h_stride</span> <span class="o">=</span> <span class="n">i2h_stride</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_i2h_pad</span> <span class="o">=</span> <span class="n">i2h_pad</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_i2h_dilate</span> <span class="o">=</span> <span class="n">i2h_dilate</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span> <span class="o">=</span> <span class="n">num_hidden</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_input_shape</span> <span class="o">=</span> <span class="n">input_shape</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_conv_layout</span> <span class="o">=</span> <span class="n">conv_layout</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_activation</span> <span class="o">=</span> <span class="n">activation</span>
<span class="c1"># Infer state shape</span>
<span class="n">data</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Variable</span><span class="p">(</span><span class="s1">'data'</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_state_shape</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Convolution</span><span class="p">(</span><span class="n">data</span><span class="o">=</span><span class="n">data</span><span class="p">,</span>
<span class="n">num_filter</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span><span class="p">,</span>
<span class="n">kernel</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_i2h_kernel</span><span class="p">,</span>
<span class="n">stride</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_i2h_stride</span><span class="p">,</span>
<span class="n">pad</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_i2h_pad</span><span class="p">,</span>
<span class="n">dilate</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_i2h_dilate</span><span class="p">,</span>
<span class="n">layout</span><span class="o">=</span><span class="n">conv_layout</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_state_shape</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_state_shape</span><span class="o">.</span><span class="n">infer_shape</span><span class="p">(</span><span class="n">data</span><span class="o">=</span><span class="n">input_shape</span><span class="p">)[</span><span class="mi">1</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_state_shape</span> <span class="o">=</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="p">)</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">_state_shape</span><span class="p">[</span><span class="mi">1</span><span class="p">:]</span>
<span class="c1"># Get params</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_iW</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'i2h_weight'</span><span class="p">,</span> <span class="n">init</span><span class="o">=</span><span class="n">i2h_weight_initializer</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_hW</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'h2h_weight'</span><span class="p">,</span> <span class="n">init</span><span class="o">=</span><span class="n">h2h_weight_initializer</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_iB</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'i2h_bias'</span><span class="p">,</span> <span class="n">init</span><span class="o">=</span><span class="n">i2h_bias_initializer</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_hB</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">params</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'h2h_bias'</span><span class="p">,</span> <span class="n">init</span><span class="o">=</span><span class="n">h2h_bias_initializer</span><span class="p">)</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">_num_gates</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_gate_names</span><span class="p">)</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">state_info</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">return</span> <span class="p">[{</span><span class="s1">'shape'</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">_state_shape</span><span class="p">,</span> <span class="s1">'__layout__'</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">_conv_layout</span><span class="p">},</span>
<span class="p">{</span><span class="s1">'shape'</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">_state_shape</span><span class="p">,</span> <span class="s1">'__layout__'</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">_conv_layout</span><span class="p">}]</span>
<span class="k">def</span> <span class="nf">_conv_forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">states</span><span class="p">,</span> <span class="n">name</span><span class="p">):</span>
<span class="n">i2h</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Convolution</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">i2h'</span><span class="o">%</span><span class="n">name</span><span class="p">,</span>
<span class="n">data</span><span class="o">=</span><span class="n">inputs</span><span class="p">,</span>
<span class="n">num_filter</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span><span class="o">*</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_gates</span><span class="p">,</span>
<span class="n">kernel</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_i2h_kernel</span><span class="p">,</span>
<span class="n">stride</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_i2h_stride</span><span class="p">,</span>
<span class="n">pad</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_i2h_pad</span><span class="p">,</span>
<span class="n">dilate</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_i2h_dilate</span><span class="p">,</span>
<span class="n">weight</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_iW</span><span class="p">,</span>
<span class="n">bias</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_iB</span><span class="p">,</span>
<span class="n">layout</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_conv_layout</span><span class="p">)</span>
<span class="n">h2h</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Convolution</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">h2h'</span><span class="o">%</span><span class="n">name</span><span class="p">,</span>
<span class="n">data</span><span class="o">=</span><span class="n">states</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span>
<span class="n">num_filter</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_hidden</span><span class="o">*</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_gates</span><span class="p">,</span>
<span class="n">kernel</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_h2h_kernel</span><span class="p">,</span>
<span class="n">dilate</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_h2h_dilate</span><span class="p">,</span>
<span class="n">pad</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_h2h_pad</span><span class="p">,</span>
<span class="n">stride</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
<span class="n">weight</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_hW</span><span class="p">,</span>
<span class="n">bias</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_hB</span><span class="p">,</span>
<span class="n">layout</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_conv_layout</span><span class="p">)</span>
<span class="k">return</span> <span class="n">i2h</span><span class="p">,</span> <span class="n">h2h</span>
<span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">states</span><span class="p">):</span>
<span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="s2">"BaseConvRNNCell is abstract class for convolutional RNN"</span><span class="p">)</span>
<span class="k">class</span> <span class="nc">ConvRNNCell</span><span class="p">(</span><span class="n">BaseConvRNNCell</span><span class="p">):</span>
<span class="sd">"""Convolutional RNN cells</span>
<span class="sd"> Parameters</span>
<span class="sd"> ----------</span>
<span class="sd"> input_shape : tuple of int</span>
<span class="sd"> Shape of input in single timestep.</span>
<span class="sd"> num_hidden : int</span>
<span class="sd"> Number of units in output symbol.</span>
<span class="sd"> h2h_kernel : tuple of int, default (3, 3)</span>
<span class="sd"> Kernel of Convolution operator in state-to-state transitions.</span>
<span class="sd"> h2h_dilate : tuple of int, default (1, 1)</span>
<span class="sd"> Dilation of Convolution operator in state-to-state transitions.</span>
<span class="sd"> i2h_kernel : tuple of int, default (3, 3)</span>
<span class="sd"> Kernel of Convolution operator in input-to-state transitions.</span>
<span class="sd"> i2h_stride : tuple of int, default (1, 1)</span>
<span class="sd"> Stride of Convolution operator in input-to-state transitions.</span>
<span class="sd"> i2h_pad : tuple of int, default (1, 1)</span>
<span class="sd"> Pad of Convolution operator in input-to-state transitions.</span>
<span class="sd"> i2h_dilate : tuple of int, default (1, 1)</span>
<span class="sd"> Dilation of Convolution operator in input-to-state transitions.</span>
<span class="sd"> i2h_weight_initializer : str or Initializer</span>
<span class="sd"> Initializer for the input weights matrix, used for the convolution</span>
<span class="sd"> transformation of the inputs.</span>
<span class="sd"> h2h_weight_initializer : str or Initializer</span>
<span class="sd"> Initializer for the recurrent weights matrix, used for the convolution</span>
<span class="sd"> transformation of the recurrent state.</span>
<span class="sd"> i2h_bias_initializer : str or Initializer, default zeros</span>
<span class="sd"> Initializer for the bias vector.</span>
<span class="sd"> h2h_bias_initializer : str or Initializer, default zeros</span>
<span class="sd"> Initializer for the bias vector.</span>
<span class="sd"> activation : str or Symbol,</span>
<span class="sd"> default functools.partial(symbol.LeakyReLU, act_type='leaky', slope=0.2)</span>
<span class="sd"> Type of activation function.</span>
<span class="sd"> prefix : str, default 'ConvRNN_'</span>
<span class="sd"> Prefix for name of layers (and name of weight if params is None).</span>
<span class="sd"> params : RNNParams, default None</span>
<span class="sd"> Container for weight sharing between cells. Created if None.</span>
<span class="sd"> conv_layout : str, , default 'NCHW'</span>
<span class="sd"> Layout of ConvolutionOp</span>
<span class="sd"> """</span>
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">input_shape</span><span class="p">,</span> <span class="n">num_hidden</span><span class="p">,</span>
<span class="n">h2h_kernel</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">h2h_dilate</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
<span class="n">i2h_kernel</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">i2h_stride</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
<span class="n">i2h_pad</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span> <span class="n">i2h_dilate</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
<span class="n">i2h_weight_initializer</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">h2h_weight_initializer</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span>
<span class="n">i2h_bias_initializer</span><span class="o">=</span><span class="s1">'zeros'</span><span class="p">,</span> <span class="n">h2h_bias_initializer</span><span class="o">=</span><span class="s1">'zeros'</span><span class="p">,</span>
<span class="n">activation</span><span class="o">=</span><span class="n">functools</span><span class="o">.</span><span class="n">partial</span><span class="p">(</span><span class="n">symbol</span><span class="o">.</span><span class="n">LeakyReLU</span><span class="p">,</span> <span class="n">act_type</span><span class="o">=</span><span class="s1">'leaky'</span><span class="p">,</span> <span class="n">slope</span><span class="o">=</span><span class="mf">0.2</span><span class="p">),</span>
<span class="n">prefix</span><span class="o">=</span><span class="s1">'ConvRNN_'</span><span class="p">,</span> <span class="n">params</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">conv_layout</span><span class="o">=</span><span class="s1">'NCHW'</span><span class="p">):</span>
<span class="nb">super</span><span class="p">(</span><span class="n">ConvRNNCell</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">input_shape</span><span class="o">=</span><span class="n">input_shape</span><span class="p">,</span> <span class="n">num_hidden</span><span class="o">=</span><span class="n">num_hidden</span><span class="p">,</span>
<span class="n">h2h_kernel</span><span class="o">=</span><span class="n">h2h_kernel</span><span class="p">,</span> <span class="n">h2h_dilate</span><span class="o">=</span><span class="n">h2h_dilate</span><span class="p">,</span>
<span class="n">i2h_kernel</span><span class="o">=</span><span class="n">i2h_kernel</span><span class="p">,</span> <span class="n">i2h_stride</span><span class="o">=</span><span class="n">i2h_stride</span><span class="p">,</span>
<span class="n">i2h_pad</span><span class="o">=</span><span class="n">i2h_pad</span><span class="p">,</span> <span class="n">i2h_dilate</span><span class="o">=</span><span class="n">i2h_dilate</span><span class="p">,</span>
<span class="n">i2h_weight_initializer</span><span class="o">=</span><span class="n">i2h_weight_initializer</span><span class="p">,</span>
<span class="n">h2h_weight_initializer</span><span class="o">=</span><span class="n">h2h_weight_initializer</span><span class="p">,</span>
<span class="n">i2h_bias_initializer</span><span class="o">=</span><span class="n">i2h_bias_initializer</span><span class="p">,</span>
<span class="n">h2h_bias_initializer</span><span class="o">=</span><span class="n">h2h_bias_initializer</span><span class="p">,</span>
<span class="n">activation</span><span class="o">=</span><span class="n">activation</span><span class="p">,</span> <span class="n">prefix</span><span class="o">=</span><span class="n">prefix</span><span class="p">,</span>
<span class="n">params</span><span class="o">=</span><span class="n">params</span><span class="p">,</span> <span class="n">conv_layout</span><span class="o">=</span><span class="n">conv_layout</span><span class="p">)</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">_gate_names</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">return</span> <span class="p">(</span><span class="s1">''</span><span class="p">,)</span>
<span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">states</span><span class="p">):</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_counter</span> <span class="o">+=</span> <span class="mi">1</span>
<span class="n">name</span> <span class="o">=</span> <span class="s1">'</span><span class="si">%s</span><span class="s1">t</span><span class="si">%d</span><span class="s1">_'</span><span class="o">%</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_counter</span><span class="p">)</span>
<span class="n">i2h</span><span class="p">,</span> <span class="n">h2h</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_conv_forward</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="n">states</span><span class="p">,</span> <span class="n">name</span><span class="p">)</span>
<span class="n">output</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_activation</span><span class="p">(</span><span class="n">i2h</span> <span class="o">+</span> <span class="n">h2h</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_activation</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">out'</span><span class="o">%</span><span class="n">name</span><span class="p">)</span>
<span class="k">return</span> <span class="n">output</span><span class="p">,</span> <span class="p">[</span><span class="n">output</span><span class="p">]</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">state_info</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">return</span> <span class="p">[{</span><span class="s1">'shape'</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">_state_shape</span><span class="p">,</span> <span class="s1">'__layout__'</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">_conv_layout</span><span class="p">}]</span>
<span class="k">class</span> <span class="nc">ConvLSTMCell</span><span class="p">(</span><span class="n">BaseConvRNNCell</span><span class="p">):</span>
<span class="sd">"""Convolutional LSTM network cell.</span>
<span class="sd"> Reference:</span>
<span class="sd"> Xingjian et al. NIPS2015</span>
<span class="sd"> Parameters</span>
<span class="sd"> ----------</span>
<span class="sd"> input_shape : tuple of int</span>
<span class="sd"> Shape of input in single timestep.</span>
<span class="sd"> num_hidden : int</span>
<span class="sd"> Number of units in output symbol.</span>
<span class="sd"> h2h_kernel : tuple of int, default (3, 3)</span>
<span class="sd"> Kernel of Convolution operator in state-to-state transitions.</span>
<span class="sd"> h2h_dilate : tuple of int, default (1, 1)</span>
<span class="sd"> Dilation of Convolution operator in state-to-state transitions.</span>
<span class="sd"> i2h_kernel : tuple of int, default (3, 3)</span>
<span class="sd"> Kernel of Convolution operator in input-to-state transitions.</span>
<span class="sd"> i2h_stride : tuple of int, default (1, 1)</span>
<span class="sd"> Stride of Convolution operator in input-to-state transitions.</span>
<span class="sd"> i2h_pad : tuple of int, default (1, 1)</span>
<span class="sd"> Pad of Convolution operator in input-to-state transitions.</span>
<span class="sd"> i2h_dilate : tuple of int, default (1, 1)</span>
<span class="sd"> Dilation of Convolution operator in input-to-state transitions.</span>
<span class="sd"> i2h_weight_initializer : str or Initializer</span>
<span class="sd"> Initializer for the input weights matrix, used for the convolution</span>
<span class="sd"> transformation of the inputs.</span>
<span class="sd"> h2h_weight_initializer : str or Initializer</span>
<span class="sd"> Initializer for the recurrent weights matrix, used for the convolution</span>
<span class="sd"> transformation of the recurrent state.</span>
<span class="sd"> i2h_bias_initializer : str or Initializer, default zeros</span>
<span class="sd"> Initializer for the bias vector.</span>
<span class="sd"> h2h_bias_initializer : str or Initializer, default zeros</span>
<span class="sd"> Initializer for the bias vector.</span>
<span class="sd"> activation : str or Symbol</span>
<span class="sd"> default functools.partial(symbol.LeakyReLU, act_type='leaky', slope=0.2)</span>
<span class="sd"> Type of activation function.</span>
<span class="sd"> prefix : str, default 'ConvLSTM_'</span>
<span class="sd"> Prefix for name of layers (and name of weight if params is None).</span>
<span class="sd"> params : RNNParams, default None</span>
<span class="sd"> Container for weight sharing between cells. Created if None.</span>
<span class="sd"> conv_layout : str, , default 'NCHW'</span>
<span class="sd"> Layout of ConvolutionOp</span>
<span class="sd"> """</span>
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">input_shape</span><span class="p">,</span> <span class="n">num_hidden</span><span class="p">,</span>
<span class="n">h2h_kernel</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">h2h_dilate</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
<span class="n">i2h_kernel</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">i2h_stride</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
<span class="n">i2h_pad</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span> <span class="n">i2h_dilate</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
<span class="n">i2h_weight_initializer</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">h2h_weight_initializer</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span>
<span class="n">i2h_bias_initializer</span><span class="o">=</span><span class="s1">'zeros'</span><span class="p">,</span> <span class="n">h2h_bias_initializer</span><span class="o">=</span><span class="s1">'zeros'</span><span class="p">,</span>
<span class="n">activation</span><span class="o">=</span><span class="n">functools</span><span class="o">.</span><span class="n">partial</span><span class="p">(</span><span class="n">symbol</span><span class="o">.</span><span class="n">LeakyReLU</span><span class="p">,</span> <span class="n">act_type</span><span class="o">=</span><span class="s1">'leaky'</span><span class="p">,</span> <span class="n">slope</span><span class="o">=</span><span class="mf">0.2</span><span class="p">),</span>
<span class="n">prefix</span><span class="o">=</span><span class="s1">'ConvLSTM_'</span><span class="p">,</span> <span class="n">params</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span>
<span class="n">conv_layout</span><span class="o">=</span><span class="s1">'NCHW'</span><span class="p">):</span>
<span class="nb">super</span><span class="p">(</span><span class="n">ConvLSTMCell</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">input_shape</span><span class="o">=</span><span class="n">input_shape</span><span class="p">,</span> <span class="n">num_hidden</span><span class="o">=</span><span class="n">num_hidden</span><span class="p">,</span>
<span class="n">h2h_kernel</span><span class="o">=</span><span class="n">h2h_kernel</span><span class="p">,</span> <span class="n">h2h_dilate</span><span class="o">=</span><span class="n">h2h_dilate</span><span class="p">,</span>
<span class="n">i2h_kernel</span><span class="o">=</span><span class="n">i2h_kernel</span><span class="p">,</span> <span class="n">i2h_stride</span><span class="o">=</span><span class="n">i2h_stride</span><span class="p">,</span>
<span class="n">i2h_pad</span><span class="o">=</span><span class="n">i2h_pad</span><span class="p">,</span> <span class="n">i2h_dilate</span><span class="o">=</span><span class="n">i2h_dilate</span><span class="p">,</span>
<span class="n">i2h_weight_initializer</span><span class="o">=</span><span class="n">i2h_weight_initializer</span><span class="p">,</span>
<span class="n">h2h_weight_initializer</span><span class="o">=</span><span class="n">h2h_weight_initializer</span><span class="p">,</span>
<span class="n">i2h_bias_initializer</span><span class="o">=</span><span class="n">i2h_bias_initializer</span><span class="p">,</span>
<span class="n">h2h_bias_initializer</span><span class="o">=</span><span class="n">h2h_bias_initializer</span><span class="p">,</span>
<span class="n">activation</span><span class="o">=</span><span class="n">activation</span><span class="p">,</span> <span class="n">prefix</span><span class="o">=</span><span class="n">prefix</span><span class="p">,</span>
<span class="n">params</span><span class="o">=</span><span class="n">params</span><span class="p">,</span> <span class="n">conv_layout</span><span class="o">=</span><span class="n">conv_layout</span><span class="p">)</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">_gate_names</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">return</span> <span class="p">[</span><span class="s1">'_i'</span><span class="p">,</span> <span class="s1">'_f'</span><span class="p">,</span> <span class="s1">'_c'</span><span class="p">,</span> <span class="s1">'_o'</span><span class="p">]</span>
<span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">states</span><span class="p">):</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_counter</span> <span class="o">+=</span> <span class="mi">1</span>
<span class="n">name</span> <span class="o">=</span> <span class="s1">'</span><span class="si">%s</span><span class="s1">t</span><span class="si">%d</span><span class="s1">_'</span><span class="o">%</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_counter</span><span class="p">)</span>
<span class="n">i2h</span><span class="p">,</span> <span class="n">h2h</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_conv_forward</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="n">states</span><span class="p">,</span> <span class="n">name</span><span class="p">)</span>
<span class="n">gates</span> <span class="o">=</span> <span class="n">i2h</span> <span class="o">+</span> <span class="n">h2h</span>
<span class="n">slice_gates</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">SliceChannel</span><span class="p">(</span><span class="n">gates</span><span class="p">,</span> <span class="n">num_outputs</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span> <span class="n">axis</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_conv_layout</span><span class="o">.</span><span class="n">find</span><span class="p">(</span><span class="s1">'C'</span><span class="p">),</span>
<span class="n">name</span><span class="o">=</span><span class="s2">"</span><span class="si">%s</span><span class="s2">slice"</span><span class="o">%</span><span class="n">name</span><span class="p">)</span>
<span class="n">in_gate</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Activation</span><span class="p">(</span><span class="n">slice_gates</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">act_type</span><span class="o">=</span><span class="s2">"sigmoid"</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">i'</span><span class="o">%</span><span class="n">name</span><span class="p">)</span>
<span class="n">forget_gate</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Activation</span><span class="p">(</span><span class="n">slice_gates</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">act_type</span><span class="o">=</span><span class="s2">"sigmoid"</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">f'</span><span class="o">%</span><span class="n">name</span><span class="p">)</span>
<span class="n">in_transform</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_activation</span><span class="p">(</span><span class="n">slice_gates</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">_activation</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">c'</span><span class="o">%</span><span class="n">name</span><span class="p">)</span>
<span class="n">out_gate</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Activation</span><span class="p">(</span><span class="n">slice_gates</span><span class="p">[</span><span class="mi">3</span><span class="p">],</span> <span class="n">act_type</span><span class="o">=</span><span class="s2">"sigmoid"</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">o'</span><span class="o">%</span><span class="n">name</span><span class="p">)</span>
<span class="n">next_c</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">_internal</span><span class="o">.</span><span class="n">_plus</span><span class="p">(</span><span class="n">forget_gate</span> <span class="o">*</span> <span class="n">states</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">in_gate</span> <span class="o">*</span> <span class="n">in_transform</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">state'</span><span class="o">%</span><span class="n">name</span><span class="p">)</span>
<span class="n">next_h</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">_internal</span><span class="o">.</span><span class="n">_mul</span><span class="p">(</span><span class="n">out_gate</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_activation</span><span class="p">(</span><span class="n">next_c</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_activation</span><span class="p">),</span>
<span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">out'</span><span class="o">%</span><span class="n">name</span><span class="p">)</span>
<span class="k">return</span> <span class="n">next_h</span><span class="p">,</span> <span class="p">[</span><span class="n">next_h</span><span class="p">,</span> <span class="n">next_c</span><span class="p">]</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">state_info</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">return</span> <span class="p">[{</span><span class="s1">'shape'</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">_state_shape</span><span class="p">,</span> <span class="s1">'__layout__'</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">_conv_layout</span><span class="p">},</span>
<span class="p">{</span><span class="s1">'shape'</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">_state_shape</span><span class="p">,</span> <span class="s1">'__layout__'</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">_conv_layout</span><span class="p">}]</span>
<span class="k">class</span> <span class="nc">ConvGRUCell</span><span class="p">(</span><span class="n">BaseConvRNNCell</span><span class="p">):</span>
<span class="sd">"""Convolutional Gated Rectified Unit (GRU) network cell.</span>
<span class="sd"> Parameters</span>
<span class="sd"> ----------</span>
<span class="sd"> input_shape : tuple of int</span>
<span class="sd"> Shape of input in single timestep.</span>
<span class="sd"> num_hidden : int</span>
<span class="sd"> Number of units in output symbol.</span>
<span class="sd"> h2h_kernel : tuple of int, default (3, 3)</span>
<span class="sd"> Kernel of Convolution operator in state-to-state transitions.</span>
<span class="sd"> h2h_dilate : tuple of int, default (1, 1)</span>
<span class="sd"> Dilation of Convolution operator in state-to-state transitions.</span>
<span class="sd"> i2h_kernel : tuple of int, default (3, 3)</span>
<span class="sd"> Kernel of Convolution operator in input-to-state transitions.</span>
<span class="sd"> i2h_stride : tuple of int, default (1, 1)</span>
<span class="sd"> Stride of Convolution operator in input-to-state transitions.</span>
<span class="sd"> i2h_pad : tuple of int, default (1, 1)</span>
<span class="sd"> Pad of Convolution operator in input-to-state transitions.</span>
<span class="sd"> i2h_dilate : tuple of int, default (1, 1)</span>
<span class="sd"> Dilation of Convolution operator in input-to-state transitions.</span>
<span class="sd"> i2h_weight_initializer : str or Initializer</span>
<span class="sd"> Initializer for the input weights matrix, used for the convolution</span>
<span class="sd"> transformation of the inputs.</span>
<span class="sd"> h2h_weight_initializer : str or Initializer</span>
<span class="sd"> Initializer for the recurrent weights matrix, used for the convolution</span>
<span class="sd"> transformation of the recurrent state.</span>
<span class="sd"> i2h_bias_initializer : str or Initializer, default zeros</span>
<span class="sd"> Initializer for the bias vector.</span>
<span class="sd"> h2h_bias_initializer : str or Initializer, default zeros</span>
<span class="sd"> Initializer for the bias vector.</span>
<span class="sd"> activation : str or Symbol,</span>
<span class="sd"> default functools.partial(symbol.LeakyReLU, act_type='leaky', slope=0.2)</span>
<span class="sd"> Type of activation function.</span>
<span class="sd"> prefix : str, default 'ConvGRU_'</span>
<span class="sd"> Prefix for name of layers (and name of weight if params is None).</span>
<span class="sd"> params : RNNParams, default None</span>
<span class="sd"> Container for weight sharing between cells. Created if None.</span>
<span class="sd"> conv_layout : str, , default 'NCHW'</span>
<span class="sd"> Layout of ConvolutionOp</span>
<span class="sd"> """</span>
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">input_shape</span><span class="p">,</span> <span class="n">num_hidden</span><span class="p">,</span>
<span class="n">h2h_kernel</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">h2h_dilate</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
<span class="n">i2h_kernel</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">i2h_stride</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
<span class="n">i2h_pad</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span> <span class="n">i2h_dilate</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
<span class="n">i2h_weight_initializer</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">h2h_weight_initializer</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span>
<span class="n">i2h_bias_initializer</span><span class="o">=</span><span class="s1">'zeros'</span><span class="p">,</span> <span class="n">h2h_bias_initializer</span><span class="o">=</span><span class="s1">'zeros'</span><span class="p">,</span>
<span class="n">activation</span><span class="o">=</span><span class="n">functools</span><span class="o">.</span><span class="n">partial</span><span class="p">(</span><span class="n">symbol</span><span class="o">.</span><span class="n">LeakyReLU</span><span class="p">,</span> <span class="n">act_type</span><span class="o">=</span><span class="s1">'leaky'</span><span class="p">,</span> <span class="n">slope</span><span class="o">=</span><span class="mf">0.2</span><span class="p">),</span>
<span class="n">prefix</span><span class="o">=</span><span class="s1">'ConvGRU_'</span><span class="p">,</span> <span class="n">params</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">conv_layout</span><span class="o">=</span><span class="s1">'NCHW'</span><span class="p">):</span>
<span class="nb">super</span><span class="p">(</span><span class="n">ConvGRUCell</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">input_shape</span><span class="o">=</span><span class="n">input_shape</span><span class="p">,</span> <span class="n">num_hidden</span><span class="o">=</span><span class="n">num_hidden</span><span class="p">,</span>
<span class="n">h2h_kernel</span><span class="o">=</span><span class="n">h2h_kernel</span><span class="p">,</span> <span class="n">h2h_dilate</span><span class="o">=</span><span class="n">h2h_dilate</span><span class="p">,</span>
<span class="n">i2h_kernel</span><span class="o">=</span><span class="n">i2h_kernel</span><span class="p">,</span> <span class="n">i2h_stride</span><span class="o">=</span><span class="n">i2h_stride</span><span class="p">,</span>
<span class="n">i2h_pad</span><span class="o">=</span><span class="n">i2h_pad</span><span class="p">,</span> <span class="n">i2h_dilate</span><span class="o">=</span><span class="n">i2h_dilate</span><span class="p">,</span>
<span class="n">i2h_weight_initializer</span><span class="o">=</span><span class="n">i2h_weight_initializer</span><span class="p">,</span>
<span class="n">h2h_weight_initializer</span><span class="o">=</span><span class="n">h2h_weight_initializer</span><span class="p">,</span>
<span class="n">i2h_bias_initializer</span><span class="o">=</span><span class="n">i2h_bias_initializer</span><span class="p">,</span>
<span class="n">h2h_bias_initializer</span><span class="o">=</span><span class="n">h2h_bias_initializer</span><span class="p">,</span>
<span class="n">activation</span><span class="o">=</span><span class="n">activation</span><span class="p">,</span> <span class="n">prefix</span><span class="o">=</span><span class="n">prefix</span><span class="p">,</span>
<span class="n">params</span><span class="o">=</span><span class="n">params</span><span class="p">,</span> <span class="n">conv_layout</span><span class="o">=</span><span class="n">conv_layout</span><span class="p">)</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">_gate_names</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">return</span> <span class="p">[</span><span class="s1">'_r'</span><span class="p">,</span> <span class="s1">'_z'</span><span class="p">,</span> <span class="s1">'_o'</span><span class="p">]</span>
<span class="nd">@property</span>
<span class="k">def</span> <span class="nf">state_info</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">return</span> <span class="p">[{</span><span class="s1">'shape'</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">_state_shape</span><span class="p">,</span> <span class="s1">'__layout__'</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">_conv_layout</span><span class="p">}]</span>
<span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">states</span><span class="p">):</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_counter</span> <span class="o">+=</span> <span class="mi">1</span>
<span class="n">seq_idx</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_counter</span>
<span class="n">name</span> <span class="o">=</span> <span class="s1">'</span><span class="si">%s</span><span class="s1">t</span><span class="si">%d</span><span class="s1">_'</span> <span class="o">%</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prefix</span><span class="p">,</span> <span class="n">seq_idx</span><span class="p">)</span>
<span class="n">i2h</span><span class="p">,</span> <span class="n">h2h</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_conv_forward</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="n">states</span><span class="p">,</span> <span class="n">name</span><span class="p">)</span>
<span class="n">i2h_r</span><span class="p">,</span> <span class="n">i2h_z</span><span class="p">,</span> <span class="n">i2h</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">SliceChannel</span><span class="p">(</span><span class="n">i2h</span><span class="p">,</span> <span class="n">num_outputs</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">name</span><span class="o">=</span><span class="s2">"</span><span class="si">%s</span><span class="s2">_i2h_slice"</span> <span class="o">%</span> <span class="n">name</span><span class="p">)</span>
<span class="n">h2h_r</span><span class="p">,</span> <span class="n">h2h_z</span><span class="p">,</span> <span class="n">h2h</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">SliceChannel</span><span class="p">(</span><span class="n">h2h</span><span class="p">,</span> <span class="n">num_outputs</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">name</span><span class="o">=</span><span class="s2">"</span><span class="si">%s</span><span class="s2">_h2h_slice"</span> <span class="o">%</span> <span class="n">name</span><span class="p">)</span>
<span class="n">reset_gate</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Activation</span><span class="p">(</span><span class="n">i2h_r</span> <span class="o">+</span> <span class="n">h2h_r</span><span class="p">,</span> <span class="n">act_type</span><span class="o">=</span><span class="s2">"sigmoid"</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s2">"</span><span class="si">%s</span><span class="s2">_r_act"</span> <span class="o">%</span> <span class="n">name</span><span class="p">)</span>
<span class="n">update_gate</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">Activation</span><span class="p">(</span><span class="n">i2h_z</span> <span class="o">+</span> <span class="n">h2h_z</span><span class="p">,</span> <span class="n">act_type</span><span class="o">=</span><span class="s2">"sigmoid"</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s2">"</span><span class="si">%s</span><span class="s2">_z_act"</span> <span class="o">%</span> <span class="n">name</span><span class="p">)</span>
<span class="n">next_h_tmp</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_activation</span><span class="p">(</span><span class="n">i2h</span> <span class="o">+</span> <span class="n">reset_gate</span> <span class="o">*</span> <span class="n">h2h</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_activation</span><span class="p">,</span>
<span class="n">name</span><span class="o">=</span><span class="s2">"</span><span class="si">%s</span><span class="s2">_h_act"</span> <span class="o">%</span> <span class="n">name</span><span class="p">)</span>
<span class="n">next_h</span> <span class="o">=</span> <span class="n">symbol</span><span class="o">.</span><span class="n">_internal</span><span class="o">.</span><span class="n">_plus</span><span class="p">((</span><span class="mf">1.</span> <span class="o">-</span> <span class="n">update_gate</span><span class="p">)</span> <span class="o">*</span> <span class="n">next_h_tmp</span><span class="p">,</span> <span class="n">update_gate</span> <span class="o">*</span> <span class="n">states</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span>
<span class="n">name</span><span class="o">=</span><span class="s1">'</span><span class="si">%s</span><span class="s1">out'</span> <span class="o">%</span> <span class="n">name</span><span class="p">)</span>
<span class="k">return</span> <span class="n">next_h</span><span class="p">,</span> <span class="p">[</span><span class="n">next_h</span><span class="p">]</span>
</pre></div>
</div>
</div>
<div aria-label="main navigation" class="sphinxsidebar rightsidebar" role="navigation">
<div class="sphinxsidebarwrapper">
</div>
</div>
</div><div class="footer">
<div class="section-disclaimer">
<div class="container">
<div>
<img height="60" src="https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/apache_incubator_logo.png"/>
<p>
Apache MXNet is an effort undergoing incubation at The Apache Software Foundation (ASF), <strong>sponsored by the <i>Apache Incubator</i></strong>. Incubation is required of all newly accepted projects until a further review indicates that the infrastructure, communications, and decision making process have stabilized in a manner consistent with other successful ASF projects. While incubation status is not necessarily a reflection of the completeness or stability of the code, it does indicate that the project has yet to be fully endorsed by the ASF.
</p>
<p>
"Copyright © 2017, The Apache Software Foundation
Apache MXNet, MXNet, Apache, the Apache feather, and the Apache MXNet project logo are either registered trademarks or trademarks of the Apache Software Foundation."
</p>
</div>
</div>
</div>
</div> <!-- pagename != index -->
</div>
<script crossorigin="anonymous" integrity="sha384-0mSbJDEHialfmuBBQP6A4Qrprq5OVfW37PRR3j5ELqxss1yVqOtnepnHVP9aJ7xS" src="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.6/js/bootstrap.min.js"></script>
<script src="../../../_static/js/sidebar.js" type="text/javascript"></script>
<script src="../../../_static/js/search.js" type="text/javascript"></script>
<script src="../../../_static/js/navbar.js" type="text/javascript"></script>
<script src="../../../_static/js/clipboard.min.js" type="text/javascript"></script>
<script src="../../../_static/js/copycode.js" type="text/javascript"></script>
<script src="../../../_static/js/page.js" type="text/javascript"></script>
<script type="text/javascript">
$('body').ready(function () {
$('body').css('visibility', 'visible');
});
</script>
</body>
</html>