Google Git
Sign in
apache / singa-site / be983bc44db8673b80f36c133d1ceb12926645fd / . / content / en / develop / build.html
blob: 3bd1ea4f8297733d63816fab8b6e010991427ebd [file] [log] [blame]
<!DOCTYPE html>
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>Build SINGA from Source &mdash; singa 2.0.0 documentation</title>
<script type="text/javascript" src="../_static/js/modernizr.min.js"></script>
<script type="text/javascript">
var DOCUMENTATION_OPTIONS = {
URL_ROOT:'../',
VERSION:'2.0.0',
LANGUAGE:'None',
COLLAPSE_INDEX:false,
FILE_SUFFIX:'.html',
HAS_SOURCE: true,
SOURCELINK_SUFFIX: '.txt'
};
</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="../_static/js/theme.js"></script>
<link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
<link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
<link rel="index" title="Index" href="../genindex.html" />
<link rel="search" title="Search" href="../search.html" />
<link href="../_static/style.css" rel="stylesheet" type="text/css">
<!--link href="../_static/fontawesome-all.min.css" rel="stylesheet" type="text/css"-->
<link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.0.13/css/all.css"
integrity="sha384-DNOHZ68U8hZfKXOrtjWvjxusGo9WQnrNx2sqG0tfsghAvtVlRW3tvkXWZh58N9jp" crossorigin="anonymous">
<style>
.fa:hover {
opacity: 0.7;
}
.fab:hover {
opacity: 0.7;
}
</style>
</head>
<body class="wy-body-for-nav">
<div class="wy-grid-for-nav">
<nav data-toggle="wy-nav-shift" class="wy-nav-side">
<div class="wy-side-scroll">
<div class="wy-side-nav-search" >
<a href="../index.html" class="icon icon-home"> singa
<img src="../_static/singa.png" class="logo" alt="Logo"/>
</a>
<div class="version">
latest
</div>
<div role="search">
<form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
<input type="text" name="q" placeholder="Search docs" />
<input type="hidden" name="check_keywords" value="yes" />
<input type="hidden" name="area" value="default" />
</form>
</div>
</div>
<div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
<ul>
<li class="toctree-l1"><a class="reference internal" href="../docs/index.html">Documentation</a></li>
<li class="toctree-l1"><a class="reference internal" href="../downloads.html">Download SINGA</a></li>
<li class="toctree-l1"><a class="reference internal" href="../security.html">Security</a></li>
</ul>
<p class="caption"><span class="caption-text">Development</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="contribute-code.html">How to Contribute Code</a></li>
<li class="toctree-l1"><a class="reference internal" href="contribute-docs.html">How to Contribute to Documentation</a></li>
<li class="toctree-l1"><a class="reference internal" href="how-to-release.html">How to prepare a release</a></li>
</ul>
<p class="caption"><span class="caption-text">Community</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../community/source-repository.html">Source Repository</a></li>
<li class="toctree-l1"><a class="reference internal" href="../community/mail-lists.html">Project Mailing Lists</a></li>
<li class="toctree-l1"><a class="reference internal" href="../community/issue-tracking.html">Issue Tracking</a></li>
<li class="toctree-l1"><a class="reference internal" href="../community/team-list.html">The SINGA Team</a></li>
</ul>
</div>
</div>
</nav>
<section data-toggle="wy-nav-shift" class="wy-nav-content-wrap">
<nav class="wy-nav-top" aria-label="top navigation">
<i data-toggle="wy-nav-top" class="fa fa-bars"></i>
<a href="../index.html">singa</a>
</nav>
<div class="wy-nav-content">
<div class="rst-content">
<div role="navigation" aria-label="breadcrumbs navigation">
<ul class="wy-breadcrumbs">
<li><a href="../index.html">Docs</a> &raquo;</li>
<li>Build SINGA from Source</li>
<li class="wy-breadcrumbs-aside">
</li>
</ul>
<hr/>
</div>
<div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
<div itemprop="articleBody">
<!--
Licensed to the Apache Software Foundation (ASF) under one
or more contributor license agreements. See the NOTICE file
distributed with this work for additional information
regarding copyright ownership. The ASF licenses this file
to you under the Apache License, Version 2.0 (the
"License"); you may not use this file except in compliance
with the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing,
software distributed under the License is distributed on an
"AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
KIND, either express or implied. See the License for the
specific language governing permissions and limitations
under the License.
--><div class="section" id="build-singa-from-source">
<h1>Build SINGA from Source<a class="headerlink" href="#build-singa-from-source" title="Permalink to this headline"></a></h1>
<p>The source files could be downloaded either as a
<a class="reference external" href="https://dist.apache.org/repos/dist/dev/singa/">tar.gz file</a>, or as a git repo</p>
<div class="highlight-default"><div class="highlight"><pre><span></span>$ git clone https://github.com/apache/singa.git
$ cd singa/
</pre></div>
</div>
<p>If you want to contribute code to SINGA, refer to <a class="reference external" href="#">this page</a> for the steps and requirements.</p>
<div class="section" id="use-conda-to-build-singa">
<h2>Use Conda to build SINGA<a class="headerlink" href="#use-conda-to-build-singa" title="Permalink to this headline"></a></h2>
<p>Conda-build is a building tool that installs the dependent libraries from anaconda cloud and
executes the building scripts.</p>
<p>To install conda-build (after installing conda)</p>
<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">conda</span> <span class="n">install</span> <span class="n">conda</span><span class="o">-</span><span class="n">build</span>
</pre></div>
</div>
<div class="section" id="build-cpu-version">
<h3>Build CPU Version<a class="headerlink" href="#build-cpu-version" title="Permalink to this headline"></a></h3>
<p>To build the CPU version of SINGA</p>
<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">conda</span> <span class="n">build</span> <span class="n">tool</span><span class="o">/</span><span class="n">conda</span><span class="o">/</span><span class="n">singa</span><span class="o">/</span>
</pre></div>
</div>
<p>The above commands have been tested on Ubuntu (14.04, 16.04 and 18.04) and macOS 10.11.
Refer to the <a class="reference external" href="https://travis-ci.org/apache/singa">Travis-CI page</a> for more information.</p>
</div>
<div class="section" id="build-gpu-version">
<h3>Build GPU Version<a class="headerlink" href="#build-gpu-version" title="Permalink to this headline"></a></h3>
<p>To build the GPU version of SINGA, the building machine must have Nvida GPU, and the CUDA driver (&gt;= 384.81), CUDA toolkit (&gt;=9) and cuDNN (&gt;=7) must have be installed. The following two Docker images provide the building environment:</p>
<ol class="simple">
<li>apache/singa:conda-cuda9.0</li>
<li>apache/singa:conda-cuda10.0</li>
</ol>
<p>Once the building environment is ready, you need to export the CUDA version first, and then run conda command to build SINGA</p>
<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">export</span> <span class="n">CUDA</span><span class="o">=</span><span class="n">x</span><span class="o">.</span><span class="n">y</span> <span class="p">(</span><span class="n">e</span><span class="o">.</span><span class="n">g</span><span class="o">.</span> <span class="mf">9.0</span><span class="p">)</span>
<span class="n">conda</span> <span class="n">build</span> <span class="n">tool</span><span class="o">/</span><span class="n">conda</span><span class="o">/</span><span class="n">singa</span><span class="o">/</span>
</pre></div>
</div>
</div>
<div class="section" id="post-processing">
<h3>Post Processing<a class="headerlink" href="#post-processing" title="Permalink to this headline"></a></h3>
<p>The location of the generated package file (<code class="docutils literal"><span class="pre">.tar.gz</span></code>) is shown on the screen.
The generated package can be installed directly,</p>
<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">conda</span> <span class="n">install</span> <span class="o">-</span><span class="n">c</span> <span class="n">conda</span><span class="o">-</span><span class="n">forge</span> <span class="o">--</span><span class="n">use</span><span class="o">-</span><span class="n">local</span> <span class="o">&lt;</span><span class="n">path</span> <span class="n">to</span> <span class="n">the</span> <span class="n">package</span> <span class="n">file</span><span class="o">&gt;</span>
</pre></div>
</div>
<p>or uploaded to anaconda cloud for others to download and install. You need to register an account on anaconda for <a class="reference external" href="https://docs.anaconda.com/anaconda-cloud/user-guide/getting-started/">uploading the package</a>.</p>
<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">conda</span> <span class="n">install</span> <span class="n">anaconda</span><span class="o">-</span><span class="n">client</span>
<span class="n">anaconda</span> <span class="n">login</span>
<span class="n">anaconda</span> <span class="n">upload</span> <span class="o">-</span><span class="n">l</span> <span class="n">main</span> <span class="o">&lt;</span><span class="n">path</span> <span class="n">to</span> <span class="n">the</span> <span class="n">package</span> <span class="n">file</span><span class="o">&gt;</span>
</pre></div>
</div>
<p>After uploading the package to the cloud, you can see it on <a class="reference external" href="https://anaconda.org/">Anaconda Cloud</a> website or via the following command</p>
<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">conda</span> <span class="n">search</span> <span class="o">-</span><span class="n">c</span> <span class="o">&lt;</span><span class="n">anaconda</span> <span class="n">username</span><span class="o">&gt;</span> <span class="n">singa</span>
</pre></div>
</div>
<p>Each specific SINGA package is identified by the version and build string. To install a specific SINGA package, you need to provide all the information, e.g.,</p>
<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">conda</span> <span class="n">install</span> <span class="o">-</span><span class="n">c</span> <span class="o">&lt;</span><span class="n">anaconda</span> <span class="n">username</span><span class="o">&gt;</span> <span class="o">-</span><span class="n">c</span> <span class="n">conda</span><span class="o">-</span><span class="n">forge</span> <span class="n">singa</span><span class="o">=</span><span class="mf">2.1</span><span class="o">.</span><span class="mf">0.</span><span class="n">dev</span><span class="o">=</span><span class="n">cpu_py36</span>
</pre></div>
</div>
<p>To make the installation command simple, you can create the following additional packages which depend on the latest CPU and GPU SINGA packages.</p>
<div class="highlight-default"><div class="highlight"><pre><span></span><span class="c1"># for singa-cpu</span>
<span class="n">conda</span> <span class="n">build</span> <span class="n">tool</span><span class="o">/</span><span class="n">conda</span><span class="o">/</span><span class="n">cpu</span><span class="o">/</span> <span class="o">--</span><span class="n">python</span><span class="o">=</span><span class="mf">3.6</span>
<span class="n">conda</span> <span class="n">build</span> <span class="n">tool</span><span class="o">/</span><span class="n">conda</span><span class="o">/</span><span class="n">cpu</span><span class="o">/</span> <span class="o">--</span><span class="n">python</span><span class="o">=</span><span class="mf">3.7</span>
<span class="c1"># for singa-gpu</span>
<span class="n">conda</span> <span class="n">build</span> <span class="n">tool</span><span class="o">/</span><span class="n">conda</span><span class="o">/</span><span class="n">gpu</span><span class="o">/</span> <span class="o">--</span><span class="n">python</span><span class="o">=</span><span class="mf">3.6</span>
<span class="n">conda</span> <span class="n">build</span> <span class="n">tool</span><span class="o">/</span><span class="n">conda</span><span class="o">/</span><span class="n">gpu</span><span class="o">/</span> <span class="o">--</span><span class="n">python</span><span class="o">=</span><span class="mf">3.7</span>
</pre></div>
</div>
<p>Therefore, when you run</p>
<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">conda</span> <span class="n">install</span> <span class="o">-</span><span class="n">c</span> <span class="o">&lt;</span><span class="n">anaconda</span> <span class="n">username</span><span class="o">&gt;</span> <span class="o">-</span><span class="n">c</span> <span class="n">conda</span><span class="o">-</span><span class="n">forge</span> <span class="n">singa</span><span class="o">-</span><span class="n">xpu</span>
</pre></div>
</div>
<p>(<code class="docutils literal"><span class="pre">xpu</span></code> is either &#8216;cpu&#8216; or &#8216;gpu&#8216;), the corresponding real SINGA package is installed as the dependent library.</p>
</div>
</div>
<div class="section" id="use-native-tools-to-build-singa-on-ubuntu">
<h2>Use native tools to build SINGA on Ubuntu<a class="headerlink" href="#use-native-tools-to-build-singa-on-ubuntu" title="Permalink to this headline"></a></h2>
<p>Refer to SINGA <a class="reference external" href="https://github.com/apache/singa/blob/master/tool/docker/devel/ubuntu/cuda9/Dockerfile#L30">Dockerfiles</a>
for the instructions of installing the dependent libraries on Ubuntu 16.04. You can also create a Docker container using the <a class="reference external" href="#">devel images</a> and build SINGA inside the container.
To build SINGA with GPU, MKLDNN, Python and unit tests, run the following instructions</p>
<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">mkdir</span> <span class="n">build</span> <span class="c1"># at the root of singa folder</span>
<span class="n">cd</span> <span class="n">build</span>
<span class="n">cmake</span> <span class="o">-</span><span class="n">DENABLE_TEST</span><span class="o">=</span><span class="n">ON</span> <span class="o">-</span><span class="n">DUSE_CUDA</span><span class="o">=</span><span class="n">ON</span> <span class="o">-</span><span class="n">DUSE_MKLDNN</span><span class="o">=</span><span class="n">ON</span> <span class="o">-</span><span class="n">DUSE_PYTHON3</span><span class="o">=</span><span class="n">ON</span> <span class="o">..</span>
<span class="n">make</span>
<span class="n">cd</span> <span class="n">python</span>
<span class="n">pip</span> <span class="n">install</span> <span class="o">.</span>
</pre></div>
</div>
<p>The details of the CMake options are explained in the last section of this page.
The last command install the Python package. You can also run <code class="docutils literal"><span class="pre">pip</span> <span class="pre">install</span> <span class="pre">-e</span> <span class="pre">.</span></code>, which creates symlinks instead of copying the Python files into the site-package folder.</p>
<p>If SINGA is compiled with ENABLE_TEST=ON, you can run the unit tests by</p>
<div class="highlight-default"><div class="highlight"><pre><span></span>$ ./bin/test_singa
</pre></div>
</div>
<p>You can see all the testing cases with testing results. If SINGA passes all
tests, then you have successfully installed SINGA.</p>
</div>
<div class="section" id="use-native-tools-to-build-singa-on-centos7">
<h2>Use native tools to Build SINGA on Centos7<a class="headerlink" href="#use-native-tools-to-build-singa-on-centos7" title="Permalink to this headline"></a></h2>
<p>Building from source will be different for Centos7 as package names differ.Follow the instructions given below.</p>
<div class="section" id="installing-dependencies">
<h3>Installing dependencies<a class="headerlink" href="#installing-dependencies" title="Permalink to this headline"></a></h3>
<p>Basic packages/libraries</p>
<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">sudo</span> <span class="n">yum</span> <span class="n">install</span> <span class="n">freetype</span><span class="o">-</span><span class="n">devel</span> <span class="n">libXft</span><span class="o">-</span><span class="n">devel</span> <span class="n">ncurses</span><span class="o">-</span><span class="n">devel</span> <span class="n">openblas</span><span class="o">-</span><span class="n">devel</span> <span class="n">blas</span><span class="o">-</span><span class="n">devel</span> <span class="n">lapack</span> <span class="n">devel</span> <span class="n">atlas</span><span class="o">-</span><span class="n">devel</span> <span class="n">kernel</span><span class="o">-</span><span class="n">headers</span> <span class="n">unzip</span> <span class="n">wget</span> <span class="n">pkgconfig</span> <span class="nb">zip</span> <span class="n">zlib</span><span class="o">-</span><span class="n">devel</span> <span class="n">libcurl</span><span class="o">-</span><span class="n">devel</span> <span class="n">cmake</span> <span class="n">curl</span> <span class="n">unzip</span> <span class="n">dh</span><span class="o">-</span><span class="n">autoreconf</span> <span class="n">git</span> <span class="n">python</span><span class="o">-</span><span class="n">devel</span> <span class="n">glog</span><span class="o">-</span><span class="n">devel</span> <span class="n">protobuf</span><span class="o">-</span><span class="n">devel</span>
</pre></div>
</div>
<p>For build-essential</p>
<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">sudo</span> <span class="n">yum</span> <span class="n">group</span> <span class="n">install</span> <span class="s2">&quot;Development Tools&quot;</span>
</pre></div>
</div>
<p>For installing swig</p>
<div class="highlight-default"><div class="highlight"><pre><span></span>sudo yum install pcre-devel
wget http://prdownloads.sourceforge.net/swig/swig-3.0.10.tar.gz
tar xvzf swig-3.0.10.tar.gz
cd swig-3.0.10.tar.gz
./configure --prefix=${RUN}
make
make install
</pre></div>
</div>
<p>For installing gfortran</p>
<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">sudo</span> <span class="n">yum</span> <span class="n">install</span> <span class="n">centos</span><span class="o">-</span><span class="n">release</span><span class="o">-</span><span class="n">scl</span><span class="o">-</span><span class="n">rh</span>
<span class="n">sudo</span> <span class="n">yum</span> <span class="o">--</span><span class="n">enablerepo</span><span class="o">=</span><span class="n">centos</span><span class="o">-</span><span class="n">sclo</span><span class="o">-</span><span class="n">rh</span><span class="o">-</span><span class="n">testing</span> <span class="n">install</span> <span class="n">devtoolset</span><span class="o">-</span><span class="mi">7</span><span class="o">-</span><span class="n">gcc</span><span class="o">-</span><span class="n">gfortran</span>
</pre></div>
</div>
<p>For installing pip and other packages</p>
<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">sudo</span> <span class="n">yum</span> <span class="n">install</span> <span class="n">epel</span><span class="o">-</span><span class="n">release</span>
<span class="n">sudo</span> <span class="n">yum</span> <span class="n">install</span> <span class="n">python</span><span class="o">-</span><span class="n">pip</span>
<span class="n">pip</span> <span class="n">install</span> <span class="n">matplotlib</span> <span class="n">numpy</span> <span class="n">pandas</span> <span class="n">scikit</span><span class="o">-</span><span class="n">learn</span> <span class="n">pydot</span>
</pre></div>
</div>
</div>
<div class="section" id="installation">
<h3>Installation<a class="headerlink" href="#installation" title="Permalink to this headline"></a></h3>
<p>Follow steps 1-5 of <em>Use native tools to build SINGA on Ubuntu</em></p>
</div>
<div class="section" id="testing">
<h3>Testing<a class="headerlink" href="#testing" title="Permalink to this headline"></a></h3>
<p>You can run the unit tests by,</p>
<div class="highlight-default"><div class="highlight"><pre><span></span>$ ./bin/test_singa
</pre></div>
</div>
<p>You can see all the testing cases with testing results. If SINGA passes all
tests, then you have successfully installed SINGA.</p>
</div>
</div>
<div class="section" id="compile-singa-on-windows">
<h2>Compile SINGA on Windows<a class="headerlink" href="#compile-singa-on-windows" title="Permalink to this headline"></a></h2>
<p>Instructions for building on Windows with Python support can be found <a class="reference external" href="install_win.html">here</a>.</p>
</div>
<div class="section" id="more-details-about-the-compilation-options">
<h2>More details about the compilation options<a class="headerlink" href="#more-details-about-the-compilation-options" title="Permalink to this headline"></a></h2>
<div class="section" id="use-modules-deprecated">
<h3>USE_MODULES (deprecated)<a class="headerlink" href="#use-modules-deprecated" title="Permalink to this headline"></a></h3>
<p>If protobuf and openblas are not installed, you can compile SINGA together with them</p>
<div class="highlight-default"><div class="highlight"><pre><span></span>$ In SINGA ROOT folder
$ mkdir build
$ cd build
$ cmake -DUSE_MODULES=ON ..
$ make
</pre></div>
</div>
<p>cmake would download OpenBlas and Protobuf (2.6.1) and compile them together
with SINGA.</p>
<p>You can use <code class="docutils literal"><span class="pre">ccmake</span> <span class="pre">..</span></code> to configure the compilation options.
If some dependent libraries are not in the system default paths, you need to export
the following environment variables</p>
<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">export</span> <span class="n">CMAKE_INCLUDE_PATH</span><span class="o">=&lt;</span><span class="n">path</span> <span class="n">to</span> <span class="n">the</span> <span class="n">header</span> <span class="n">file</span> <span class="n">folder</span><span class="o">&gt;</span>
<span class="n">export</span> <span class="n">CMAKE_LIBRARY_PATH</span><span class="o">=&lt;</span><span class="n">path</span> <span class="n">to</span> <span class="n">the</span> <span class="n">lib</span> <span class="n">file</span> <span class="n">folder</span><span class="o">&gt;</span>
</pre></div>
</div>
</div>
<div class="section" id="use-python">
<h3>USE_PYTHON<a class="headerlink" href="#use-python" title="Permalink to this headline"></a></h3>
<p>Option for compiling the Python wrapper for SINGA,</p>
<div class="highlight-default"><div class="highlight"><pre><span></span>$ cmake -DUSE_PYTHON=ON ..
$ make
$ cd python
$ pip install .
</pre></div>
</div>
</div>
<div class="section" id="use-cuda">
<h3>USE_CUDA<a class="headerlink" href="#use-cuda" title="Permalink to this headline"></a></h3>
<p>Users are encouraged to install the CUDA and
<a class="reference external" href="https://developer.nvidia.com/cudnn">cuDNN</a> for running SINGA on GPUs to
get better performance.</p>
<p>SINGA has been tested over CUDA 9/10, and cuDNN 7. If cuDNN is
installed into non-system folder, e.g. /home/bob/local/cudnn/, the following
commands should be executed for cmake and the runtime to find it</p>
<div class="highlight-default"><div class="highlight"><pre><span></span>$ export CMAKE_INCLUDE_PATH=/home/bob/local/cudnn/include:$CMAKE_INCLUDE_PATH
$ export CMAKE_LIBRARY_PATH=/home/bob/local/cudnn/lib64:$CMAKE_LIBRARY_PATH
$ export LD_LIBRARY_PATH=/home/bob/local/cudnn/lib64:$LD_LIBRARY_PATH
</pre></div>
</div>
<p>The cmake options for CUDA and cuDNN should be switched on</p>
<div class="highlight-default"><div class="highlight"><pre><span></span># Dependent libs are install already
$ cmake -DUSE_CUDA=ON ..
$ make
</pre></div>
</div>
</div>
<div class="section" id="use-mkldnn">
<h3>USE_MKLDNN<a class="headerlink" href="#use-mkldnn" title="Permalink to this headline"></a></h3>
<p>User can enable MKL-DNN to enhance the performance of CPU computation.</p>
<p>Installation guide of MKL-DNN could be found <a class="reference external" href="https://github.com/intel/mkl-dnn#installation">here</a>.</p>
<p>SINGA has been tested over MKL-DNN v0.17.2.</p>
<p>To build SINGA with MKL-DNN support:</p>
<div class="highlight-default"><div class="highlight"><pre><span></span># Dependent libs are installed already
$ cmake -DUSE_MKLDNN=ON ..
$ make
</pre></div>
</div>
</div>
<div class="section" id="use-opencl">
<h3>USE_OPENCL<a class="headerlink" href="#use-opencl" title="Permalink to this headline"></a></h3>
<p>SINGA uses opencl-headers and viennacl (version 1.7.1 or newer) for OpenCL support, which
can be installed using via</p>
<div class="highlight-default"><div class="highlight"><pre><span></span># On Ubuntu 16.04
$ sudo apt-get install opencl-headers, libviennacl-dev
# On Fedora
$ sudo yum install opencl-headers, viennacl
</pre></div>
</div>
<p>Additionally, you will need the OpenCL Installable Client Driver (ICD) for the platforms that you want to run OpenCL on.</p>
<ul class="simple">
<li>For AMD and nVidia GPUs, the driver package should also install the correct OpenCL ICD.</li>
<li>For Intel CPUs and/or GPUs, get the driver from the <a class="reference external" href="https://software.intel.com/en-us/articles/opencl-drivers">Intel website.</a> Note that the drivers provided on that website only supports recent CPUs and Iris GPUs.</li>
<li>For older Intel CPUs, you can use the <code class="docutils literal"><span class="pre">beignet-opencl-icd</span></code> package.</li>
</ul>
<p>Note that running OpenCL on CPUs is not currently recommended because it is slow.
Memory transfer is on the order of whole seconds (1000&#8216;s of ms on CPUs as compared to 1&#8216;s of ms on GPUs).</p>
<p>More information on setting up a working OpenCL environment may be found <a class="reference external" href="https://wiki.tiker.net/OpenCLHowTo">here</a>.</p>
<p>If the package version of ViennaCL is not at least 1.7.1, you will need to build it from source:</p>
<p>Clone <a class="reference external" href="https://github.com/viennacl/viennacl-dev">the repository from here</a>, checkout the <code class="docutils literal"><span class="pre">release-1.7.1</span></code> tag and build it.
Remember to add its directory to <code class="docutils literal"><span class="pre">PATH</span></code> and the built libraries to <code class="docutils literal"><span class="pre">LD_LIBRARY_PATH</span></code>.</p>
<p>To build SINGA with OpenCL support (tested on SINGA 1.1):</p>
<div class="highlight-default"><div class="highlight"><pre><span></span>$ cmake -DUSE_OPENCL=ON ..
$ make
</pre></div>
</div>
</div>
<div class="section" id="package">
<h3>PACKAGE<a class="headerlink" href="#package" title="Permalink to this headline"></a></h3>
<p>This setting is used to build the Debian package. Set PACKAGE=ON and build the package with make command like this:</p>
<div class="highlight-default"><div class="highlight"><pre><span></span>$ cmake -DPACKAGE=ON
$ make package
</pre></div>
</div>
</div>
</div>
<div class="section" id="faq">
<h2>FAQ<a class="headerlink" href="#faq" title="Permalink to this headline"></a></h2>
<ul>
<li><p class="first">Q: Error from &#8216;import singa&#8216;</p>
<p>A: Please check the detailed error from <code class="docutils literal"><span class="pre">python</span> <span class="pre">-c</span>&#160; <span class="pre">&quot;from</span> <span class="pre">singa</span> <span class="pre">import</span> <span class="pre">_singa_wrap&quot;</span></code>. Sometimes it is caused by the dependent libraries, e.g. there are multiple versions of protobuf, missing of cudnn, numpy version mismatch. Following steps show the solutions for different cases</p>
<ol>
<li><p class="first">Check the cudnn and cuda. If cudnn is missing or not match with the wheel version, you can download the correct version of cudnn into ~/local/cudnn/ and</p>
<div class="highlight-default"><div class="highlight"><pre><span></span> $ echo &quot;export LD_LIBRARY_PATH=/home/&lt;yourname&gt;/local/cudnn/lib64:$LD_LIBRARY_PATH&quot; &gt;&gt; ~/.bashrc
</pre></div>
</div>
</li>
<li><p class="first">If it is the problem related to protobuf. You can install protobuf (3.6.1) from source into a local folder, say ~/local/; Decompress the tar file, and then</p>
<div class="highlight-default"><div class="highlight"><pre><span></span> $ ./configure --prefix=/home/&lt;yourname&gt;local
$ make &amp;&amp; make install
$ echo &quot;export LD_LIBRARY_PATH=/home/&lt;yourname&gt;/local/lib:$LD_LIBRARY_PATH&quot; &gt;&gt; ~/.bashrc
$ source ~/.bashrc
</pre></div>
</div>
</li>
<li><p class="first">If it cannot find other libs including python, then create virtual env using pip or conda;</p>
</li>
<li><p class="first">If it is not caused by the above reasons, go to the folder of <code class="docutils literal"><span class="pre">_singa_wrap.so</span></code>,</p>
<div class="highlight-default"><div class="highlight"><pre><span></span> $ python
&gt;&gt; import importlib
&gt;&gt; importlib.import_module(&#39;_singa_wrap&#39;)
</pre></div>
</div>
</li>
</ol>
<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">Check</span> <span class="n">the</span> <span class="n">error</span> <span class="n">message</span><span class="o">.</span> <span class="n">For</span> <span class="n">example</span><span class="p">,</span> <span class="k">if</span> <span class="n">the</span> <span class="n">numpy</span> <span class="n">version</span> <span class="n">mismatches</span><span class="p">,</span> <span class="n">the</span> <span class="n">error</span> <span class="n">message</span> <span class="n">would</span> <span class="n">be</span><span class="p">,</span>
<span class="ne">RuntimeError</span><span class="p">:</span> <span class="n">module</span> <span class="n">compiled</span> <span class="n">against</span> <span class="n">API</span> <span class="n">version</span> <span class="mh">0xb</span> <span class="n">but</span> <span class="n">this</span> <span class="n">version</span> <span class="n">of</span> <span class="n">numpy</span> <span class="ow">is</span> <span class="mh">0xa</span>
<span class="n">Then</span> <span class="n">you</span> <span class="n">need</span> <span class="n">to</span> <span class="n">upgrade</span> <span class="n">the</span> <span class="n">numpy</span><span class="o">.</span>
</pre></div>
</div>
</li>
<li><p class="first">Q: Error from running <code class="docutils literal"><span class="pre">cmake</span> <span class="pre">..</span></code>, which cannot find the dependent libraries.</p>
<p>A: If you haven&#8216;t installed the libraries, install them. If you installed
the libraries in a folder that is outside of the system folder, e.g. /usr/local,
you need to export the following variables</p>
<div class="highlight-default"><div class="highlight"><pre><span></span> $ export CMAKE_INCLUDE_PATH=&lt;path to your header file folder&gt;
$ export CMAKE_LIBRARY_PATH=&lt;path to your lib file folder&gt;
</pre></div>
</div>
</li>
<li><p class="first">Q: Error from <code class="docutils literal"><span class="pre">make</span></code>, e.g. the linking phase</p>
<p>A: If your libraries are in other folders than system default paths, you need
to export the following varaibles</p>
<div class="highlight-default"><div class="highlight"><pre><span></span> $ export LIBRARY_PATH=&lt;path to your lib file folder&gt;
$ export LD_LIBRARY_PATH=&lt;path to your lib file folder&gt;
</pre></div>
</div>
</li>
<li><p class="first">Q: Error from header files, e.g. &#8216;cblas.h no such file or directory exists&#8216;</p>
<p>A: You need to include the folder of the cblas.h into CPLUS_INCLUDE_PATH,
e.g.,</p>
<div class="highlight-default"><div class="highlight"><pre><span></span> $ export CPLUS_INCLUDE_PATH=/opt/OpenBLAS/include:$CPLUS_INCLUDE_PATH
</pre></div>
</div>
</li>
<li><p class="first">Q:While compiling SINGA, I get error <code class="docutils literal"><span class="pre">SSE2</span> <span class="pre">instruction</span> <span class="pre">set</span> <span class="pre">not</span> <span class="pre">enabled</span></code></p>
<p>A:You can try following command:</p>
<div class="highlight-default"><div class="highlight"><pre><span></span> $ make CFLAGS=&#39;-msse2&#39; CXXFLAGS=&#39;-msse2&#39;
</pre></div>
</div>
</li>
<li><p class="first">Q:I get <code class="docutils literal"><span class="pre">ImportError:</span> <span class="pre">cannot</span> <span class="pre">import</span> <span class="pre">name</span> <span class="pre">enum_type_wrapper</span></code> from google.protobuf.internal when I try to import .py files.</p>
<p>A: You need to install the python binding of protobuf, which could be installed via</p>
<div class="highlight-default"><div class="highlight"><pre><span></span> $ sudo apt-get install protobuf
</pre></div>
</div>
<p>or from source</p>
<div class="highlight-default"><div class="highlight"><pre><span></span> $ cd /PROTOBUF/SOURCE/FOLDER
$ cd python
$ python setup.py build
$ python setup.py install
</pre></div>
</div>
</li>
<li><p class="first">Q: When I build OpenBLAS from source, I am told that I need a Fortran compiler.</p>
<p>A: You can compile OpenBLAS by</p>
<div class="highlight-default"><div class="highlight"><pre><span></span> $ make ONLY_CBLAS=1
</pre></div>
</div>
<p>or install it using</p>
<div class="highlight-default"><div class="highlight"><pre><span></span> $ sudo apt-get install libopenblas-dev
</pre></div>
</div>
</li>
<li><p class="first">Q: When I build protocol buffer, it reports that GLIBC++_3.4.20 not found in /usr/lib64/libstdc++.so.6.</p>
<p>A: This means the linker found libstdc++.so.6 but that library
belongs to an older version of GCC than was used to compile and link the
program. The program depends on code defined in
the newer libstdc++ that belongs to the newer version of GCC, so the linker
must be told how to find the newer libstdc++ shared library.
The simplest way to fix this is to find the correct libstdc++ and export it to
LD_LIBRARY_PATH. For example, if GLIBC++_3.4.20 is listed in the output of the
following command,</p>
<div class="highlight-default"><div class="highlight"><pre><span></span> $ strings /usr/local/lib64/libstdc++.so.6|grep GLIBC++
</pre></div>
</div>
<p>then you just set your environment variable as</p>
<div class="highlight-default"><div class="highlight"><pre><span></span> $ export LD_LIBRARY_PATH=/usr/local/lib64:$LD_LIBRARY_PATH
</pre></div>
</div>
</li>
<li><p class="first">Q: When I build glog, it reports that &#8220;src/logging_unittest.cc:83:20: error: ‘gflags’ is not a namespace-name&#8220;</p>
<p>A: It maybe that you have installed gflags with a different namespace such as &#8220;google&#8220;. so glog can&#8216;t find &#8216;gflags&#8216; namespace.
Because it is not necessary to have gflags to build glog. So you can change the configure.ac file to ignore gflags.</p>
<div class="highlight-default"><div class="highlight"><pre><span></span> <span class="mf">1.</span> <span class="n">cd</span> <span class="n">to</span> <span class="n">glog</span> <span class="n">src</span> <span class="n">directory</span>
<span class="mf">2.</span> <span class="n">change</span> <span class="n">line</span> <span class="mi">125</span> <span class="n">of</span> <span class="n">configure</span><span class="o">.</span><span class="n">ac</span> <span class="n">to</span> <span class="s2">&quot;AC_CHECK_LIB(gflags, main, ac_cv_have_libgflags=0, ac_cv_have_libgflags=0)&quot;</span>
<span class="mf">3.</span> <span class="n">autoreconf</span>
</pre></div>
</div>
<p>After this, you can build glog again.</p>
</li>
<li><p class="first">Q: When using virtual environment, every time I run pip install, it would reinstall numpy. However, the numpy would not be used when I <code class="docutils literal"><span class="pre">import</span> <span class="pre">numpy</span></code></p>
<p>A: It could be caused by the <code class="docutils literal"><span class="pre">PYTHONPATH</span></code> which should be set to empty when you are using virtual environment to avoid the conflicts with the path of
the virtual environment.</p>
</li>
<li><p class="first">Q: When compiling PySINGA from source, there is a compilation error due to the missing of &lt;numpy/objectarray.h&gt;</p>
<p>A: Please install numpy and export the path of numpy header files as</p>
<div class="highlight-default"><div class="highlight"><pre><span></span> $ export CPLUS_INCLUDE_PATH=`python -c &quot;import numpy; print numpy.get_include()&quot;`:$CPLUS_INCLUDE_PATH
</pre></div>
</div>
</li>
<li><p class="first">Q: When I run SINGA in Mac OS X, I got the error &#8220;Fatal Python error: PyThreadState_Get: no current thread Abort trap: 6&#8220;</p>
<p>A: This error happens typically when you have multiple version of Python on your system and you installed SINGA via pip (this problem is resolved for installation via conda),
e.g, the one comes with the OS and the one installed by Homebrew. The Python linked by PySINGA must be the same as the Python interpreter.
You can check your interpreter by <code class="docutils literal"><span class="pre">which</span> <span class="pre">python</span></code> and check the Python linked by PySINGA via <code class="docutils literal"><span class="pre">otool</span> <span class="pre">-L</span> <span class="pre">&lt;path</span> <span class="pre">to</span> <span class="pre">_singa_wrap.so&gt;</span></code>.
To fix this error, compile SINGA with the correct version of Python.
In particular, if you build PySINGA from source, you need to specify the paths when invoking <a class="reference external" href="http://stackoverflow.com/questions/15291500/i-have-2-versions-of-python-installed-but-cmake-is-using-older-version-how-do">cmake</a></p>
<div class="highlight-default"><div class="highlight"><pre><span></span> $ cmake -DPYTHON_LIBRARY=`python-config --prefix`/lib/libpython2.7.dylib -DPYTHON_INCLUDE_DIR=`python-config --prefix`/include/python2.7/ ..
</pre></div>
</div>
<p>If installed PySINGA from binary packages, e.g. debian or wheel, then you need to change the python interpreter, e.g., reset the $PATH to put the correct path of Python at the front position.</p>
</li>
</ul>
</div>
</div>
</div>
</div>
<footer>
<hr/>
<div role="contentinfo">
<p>
&copy; Copyright 2019 The Apache Software Foundation. All rights reserved. Apache SINGA, Apache, the Apache feather logo, and the Apache SINGA project logos are trademarks of The Apache Software Foundation. All other marks mentioned may be trademarks or registered trademarks of their respective owners.
</p>
</div>
Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
</footer>
</div>
</div>
</section>
</div>
<script type="text/javascript">
jQuery(function () {
SphinxRtdTheme.Navigation.enable(true);
});
</script>
<div class="rst-versions" data-toggle="rst-versions" role="note" aria-label="versions">
<span class="rst-current-version" data-toggle="rst-current-version">
<span class="fa fa-book"> singa </span>
v: latest
<span class="fa fa-caret-down"></span>
</span>
<div class="rst-other-versions">
<dl>
<dt>Languages</dt>
<dd><a href=".././index.html">English</a></dd>
<dd><a href=".././zh/index.html">中文</a></dd>
</dl>
<dl>
<dt>Versions</dt>
<dd><a href="http://singa.apache.org/v0.3.0/">0.3</a></dd>
<dd><a href="http://singa.apache.org/v1.1.0/">1.1</a></dd>
</dl>
</div>
<a href="http://www.apache.org"
style="color:lightblue;padding: 5px; font-size: 10px; text-align: center; text-decoration: none; margin: 5px 2px;">Foundation</a>
<a href="http://www.apache.org/events/current-event"
style="color:lightblue;padding: 5px; font-size: 10px; text-align: center; text-decoration: none; margin: 5px 2px;">Events</a>
<a href="http://www.apache.org/foundation/thanks.html"
style="color:lightblue;padding: 5px; font-size: 10px; text-align: center; text-decoration: none; margin: 5px 2px;">Thanks</a>
<a href="http://www.apache.org/foundation/sponsorship.html"
style="color:lightblue;padding: 5px; font-size: 10px; text-align: center; text-decoration: none; margin: 5px 2px;">Sponsorship</a>
<a href="http://www.apache.org/licenses/"
style="color:lightblue;padding: 5px; font-size: 10px; text-align: center; text-decoration: none; margin: 5px 2px;">License</a>
<br>
<a href="https://github.com/apache/singa" class="fa fa-github"
style="padding: 10px; font-size: 20px; width: 30px; text-align: center; text-decoration: none; margin: 5px 2px;"></a>
<a href="https://aws.amazon.com/marketplace/seller-profile?id=5bcac385-12c4-4802-aec7-351e09b77b4c"
class="fab fa-aws"
style="padding: 10px; font-size: 20px; width: 30px; text-align: center; text-decoration: none; margin: 5px 2px;"></a>
<a href="https://hub.docker.com/r/apache/singa/" class="fab fa-docker"
style="padding: 10px; font-size: 20px; width: 30px; text-align: center; text-decoration: none; margin: 5px 2px;"></a>
<a href="https://www.linkedin.com/groups/13550034" class="fa fa-linkedin"
style="padding: 10px; font-size: 20px; width: 30px; text-align: center; text-decoration: none; margin: 5px 2px;"></a>
<a href="https://twitter.com/ApacheSinga" class="fa fa-twitter"
style="padding: 10px; font-size: 20px; width: 30px; text-align: center; text-decoration: none; margin: 5px 2px;"></a>
<a href="https://www.facebook.com/Apache-SINGA-347284219056544/" class="fa fa-facebook"
style="padding: 10px; font-size: 20px; width: 30px; text-align: center; text-decoration: none; margin: 5px 2px;"></a>
<a href="https://www.researchgate.net/project/Apache-SINGA" class="fab fa-researchgate"
style="padding: 10px; font-size: 20px; width: 30px; text-align: center; text-decoration: none; margin: 5px 2px;"></a>
</div>
<a href="https://github.com/apache/singa">
<img style="position: absolute; top: 0; right: 0; border: 0; z-index: 10000;"
src="https://s3.amazonaws.com/github/ribbons/forkme_right_orange_ff7600.png" alt="Fork me on GitHub">
</a>
</body>
</html>