You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@mxnet.apache.org by GitBox <gi...@apache.org> on 2018/06/11 16:12:44 UTC
[GitHub] marcoabreu closed pull request #11211: docs build update for v1.2.1
marcoabreu closed pull request #11211: docs build update for v1.2.1
URL: https://github.com/apache/incubator-mxnet/pull/11211
This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:
As this is a foreign pull request (from a fork), the diff is supplied
below (as it won't show otherwise due to GitHub magic):
diff --git a/ci/docker/install/ubuntu_caffe.sh b/ci/docker/install/ubuntu_caffe.sh
index 50e257b2ef3..8c13612114b 100755
--- a/ci/docker/install/ubuntu_caffe.sh
+++ b/ci/docker/install/ubuntu_caffe.sh
@@ -18,10 +18,19 @@
# under the License.
set -ex
-apt-get install -y libprotobuf-dev libleveldb-dev \
- libsnappy-dev libopencv-dev libhdf5-serial-dev protobuf-compiler \
- libatlas-base-dev python-dev libgflags-dev libgoogle-glog-dev liblmdb-dev \
- python-numpy python-opencv
+apt-get install -y \
+ libgflags-dev \
+ libgoogle-glog-dev \
+ libhdf5-serial-dev \
+ libleveldb-dev \
+ liblmdb-dev \
+ libopencv-dev \
+ libprotobuf-dev \
+ libsnappy-dev \
+ protobuf-compiler \
+ python-dev \
+ python-numpy \
+ python-opencv
apt-get install -y --no-install-recommends libboost-all-dev
@@ -46,4 +55,3 @@ make all pycaffe -j$(nproc)
cd python
for req in $(cat requirements.txt); do pip2 install $req; done
-
diff --git a/ci/docker/install/ubuntu_core.sh b/ci/docker/install/ubuntu_core.sh
index f588d06ab57..65ad4726597 100755
--- a/ci/docker/install/ubuntu_core.sh
+++ b/ci/docker/install/ubuntu_core.sh
@@ -23,19 +23,21 @@
set -ex
apt-get update
apt-get install -y \
+ apt-transport-https \
build-essential \
+ ca-certificates \
+ cmake \
+ curl \
git \
- libopenblas-dev \
+ libatlas-base-dev \
+ libcurl4-openssl-dev \
+ libjemalloc-dev \
liblapack-dev \
+ libopenblas-dev \
libopencv-dev \
- libcurl4-openssl-dev \
- cmake \
- wget \
- unzip \
- sudo \
libzmq3-dev \
+ ninja-build \
software-properties-common \
- ninja-build
-
-# Link Openblas to Cblas as this link does not exist on ubuntu16.04
-ln -s /usr/lib/libopenblas.so /usr/lib/libcblas.so
+ sudo \
+ unzip \
+ wget
diff --git a/ci/docker/install/ubuntu_docs.sh b/ci/docker/install/ubuntu_docs.sh
index bb8026f0daa..45a6768becb 100755
--- a/ci/docker/install/ubuntu_docs.sh
+++ b/ci/docker/install/ubuntu_docs.sh
@@ -21,8 +21,21 @@
# the whole docker cache for the image
set -ex
-wget http://downloads.lightbend.com/scala/2.11.8/scala-2.11.8.deb && \
- dpkg -i scala-2.11.8.deb && rm scala-2.11.8.deb
+# Install dependencies
+echo 'Installing dependencies...'
+apt-get install -y \
+ doxygen \
+ pandoc
-apt-get install -y doxygen libatlas-base-dev graphviz pandoc
-pip install sphinx==1.3.5 CommonMark==0.5.4 breathe mock recommonmark pypandoc beautifulsoup4
+echo 'Installing python packages...'
+pip install --upgrade pip && pip install \
+ beautifulsoup4 \
+ breathe \
+ CommonMark==0.5.4 \
+ h5py \
+ mock==1.0.1 \
+ pypandoc \
+ recommonmark==0.4.0 \
+ sphinx==1.5.6
+
+echo 'Dependency installation complete.'
diff --git a/ci/docker/install/ubuntu_scala.sh b/ci/docker/install/ubuntu_scala.sh
index ee5554d218a..8dc788c12f6 100755
--- a/ci/docker/install/ubuntu_scala.sh
+++ b/ci/docker/install/ubuntu_scala.sh
@@ -22,10 +22,16 @@
set -ex
# install libraries for mxnet's scala package on ubuntu
+echo 'Installing Scala...'
apt-get install -y software-properties-common
add-apt-repository -y ppa:webupd8team/java
apt-get update
-echo "oracle-java8-installer shared/accepted-oracle-license-v1-1 select true" | debconf-set-selections
-apt-get install -y oracle-java8-installer
-apt-get install -y oracle-java8-set-default
-apt-get update && apt-get install -y maven
\ No newline at end of file
+apt-get install -y openjdk-8-jdk
+apt-get install -y openjdk-8-jre
+
+echo "deb https://dl.bintray.com/sbt/debian /" | tee -a /etc/apt/sources.list.d/sbt.list
+apt-key adv --keyserver hkp://keyserver.ubuntu.com:80 --recv 2EE0EA64E40A89B84B2DF73499E82A75642AC823
+apt-get update && apt-get install -y \
+ maven \
+ sbt \
+ scala
diff --git a/ci/docker/runtime_functions.sh b/ci/docker/runtime_functions.sh
index 44de137b6a8..77ffad126da 100755
--- a/ci/docker/runtime_functions.sh
+++ b/ci/docker/runtime_functions.sh
@@ -542,6 +542,17 @@ test_ubuntu_cpu_python3() {
popd
}
+build_docs() {
+ set -ex
+ pushd .
+ cd /work/mxnet/docs/build_version_doc
+ ./build_all_version.sh $1
+ ./update_all_version.sh $2 $3 $4
+ cd VersionedWeb
+ tar -zcvf ../artifacts.tgz .
+ popd
+}
+
# Deploy
deploy_docs() {
diff --git a/docs/Jenkinsfile b/docs/Jenkinsfile
new file mode 100644
index 00000000000..618511b7bce
--- /dev/null
+++ b/docs/Jenkinsfile
@@ -0,0 +1,81 @@
+// -*- mode: groovy -*-
+
+// 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.
+
+// Jenkins pipeline
+// See documents at https://jenkins.io/doc/book/pipeline/jenkinsfile/
+
+// timeout in minutes
+max_time = 60
+// assign any caught errors here
+err = null
+
+// initialize source code
+def init_git() {
+ deleteDir()
+ retry(5) {
+ try {
+ // Make sure wait long enough for api.github.com request quota. Important: Don't increase the amount of
+ // retries as this will increase the amount of requests and worsen the throttling
+ timeout(time: 15, unit: 'MINUTES') {
+ checkout scm
+ sh 'git submodule update --init --recursive'
+ sh 'git clean -d -f'
+ }
+ } catch (exc) {
+ deleteDir()
+ error "Failed to fetch source codes with ${exc}"
+ sleep 2
+ }
+ }
+}
+
+try {
+ stage('Build Docs') {
+ node('restricted-mxnetlinux-cpu') {
+ ws('workspace/docs') {
+ init_git()
+ timeout(time: max_time, unit: 'MINUTES') {
+ sh "ci/build.py -p ubuntu_cpu /work/runtime_functions.sh build_docs ${params.tags_to_build} ${params.tag_list} ${params.tag_default} ${params.domain}"
+ archiveArtifacts 'docs/build_version_doc/artifacts.tgz'
+ build 'restricted-website-publish'
+ }
+ }
+ }
+ }
+
+ // set build status to success at the end
+ currentBuild.result = "SUCCESS"
+} catch (caughtError) {
+ node("restricted-mxnetlinux-cpu") {
+ sh "echo caught ${caughtError}"
+ err = caughtError
+ currentBuild.result = "FAILURE"
+ }
+} finally {
+ node("restricted-mxnetlinux-cpu") {
+ // Only send email if master failed
+ if (currentBuild.result == "FAILURE" && env.BRANCH_NAME == "master") {
+ emailext body: 'Build for MXNet branch ${BRANCH_NAME} has broken. Please view the build at ${BUILD_URL}', replyTo: '${EMAIL}', subject: '[BUILD FAILED] Branch ${BRANCH_NAME} build ${BUILD_NUMBER}', to: '${EMAIL}'
+ }
+ // Remember to rethrow so the build is marked as failing
+ if (err) {
+ throw err
+ }
+ }
+}
diff --git a/docs/build_version_doc/README.md b/docs/build_version_doc/README.md
index 221618b0aa5..3509adf50c7 100644
--- a/docs/build_version_doc/README.md
+++ b/docs/build_version_doc/README.md
@@ -7,86 +7,70 @@ This folder contains a variety of scripts to generate the MXNet.io website as we
* [AddVersion.py](AddVersion.py) - MXNet.io site data massaging; injects the versions dropdown menu in the navigation bar
* [build_site_tag.sh](build_site_tag.sh) - takes version tags as input and generates static html; calls `build_all_version.sh` and `update_all_version.sh`
* [build_all_version.sh](build_all_version.sh) - takes version tags as input and builds the basic static html for MXNet.io
-* [build_doc.sh](build_doc.sh) - used by the CI system to generate MXNet.io; only triggered by new tags; not meant for manual runs or custom outputs
* [Dockerfile](Dockerfile) - has all dependencies needed to build and update MXNet.io's static html
* [update_all_version.sh](update_all_version.sh) - takes the output of `build_all_version.sh` then uses `AddVersion.py` and `AddPackageLink.py` to update the static html
-## CI Flow
-1. Calls `build_doc.sh`.
-2. `VersionedWeb` folder generated with static html of site; old versions are in `VersionedWeb/versions`.
-3. `asf-site` branch from the [incubator-mxnet-site](https://github.com/apache/incubator-mxnet-site) project is checked out and contents are deleted.
-4. New site content from `VersionedWeb` is copied into `asf-site` branch and then committed with `git`.
-5. [MXNet.io](http://mxnet.io) should then show the new content.
+## Setting Up a Docs Dev Server
-## Manual Generation
+For these instructions, you will use an Ubuntu machine. This flow has been tested on a [Deep Learning Base AMI](https://aws.amazon.com/marketplace/pp/B077GCZ4GR), although you may use the full Deep Learning Base AMI or any other Ubuntu 16.04 system with some minor adjustments.
-Use Ubuntu and the setup defined below, or use the Dockerfile provided in this folder to spin up an Ubuntu image with all of the dependencies. Further info on Docker is provided later in this document. For a cloud image, this was tested on [Deep Learning AMI v5](https://aws.amazon.com/marketplace/pp/B077GCH38C?qid=1520359179176).
+**Step 1:** Spin up your Ubuntu server and SSH in.
-**Note**: for AMI users or if you already have Conda, you might be stuck with the latest version and the docs build will have a conflict. To fix this, run `/home/ubuntu/anaconda3/bin/pip uninstall sphinx` and follow this with `pip install --user sphinx==1.5.6`.
+**Step 2:** Create a Python 2.7 virtual environment (see note).
-If you need to build <= v0.12.0, then use a Python 2 environment to avoid errors with `mxdoc.py`. This is a sphinx extension, that was not Python 3 compatible in the old versions. On the Deep Learning AMI, use `source activate mxnet_p27`, and then install the following dependencies.
+```bash
+sudo apt install virtualenv
+virtualenv -p python2.7 mxnet_docs
+source mxnet_docs/bin/activate
+```
+**Note:** Using a Python 2.7 environment is required to build older versions of the docs that have Python 3 incompatibilities. If you're only building the latest or version 1.0.0+, then you may use a Python 3 environment.
-### Dependencies
+**Step 3:** Clone the repo.
-These are the dependencies for docs generation for Ubuntu 16.04.
+```bash
+git clone --recursive https://github.com/apache/incubator-mxnet.git
+```
-This script is available for you to run directly on Ubuntu from the source repository.
-Run `./setup_docs_ubuntu.sh`.
+**Step 4:** Install dependencies.
+This script will install dependencies for you.
+
+```bash
+./incubator-mxnet/docs/build_version_doc/setup_docs_ubuntu.sh
```
-sudo apt-get update
-sudo apt-get install -y \
- apt-transport-https \
- ca-certificates \
- curl \
- doxygen \
- git \
- libjemalloc-dev \
- pandoc \
- software-properties-common
-# You made need to run `/home/ubuntu/anaconda3/bin/pip uninstall sphinx`
-# Recommonmark/Sphinx errors: https://github.com/sphinx-doc/sphinx/issues/3800
-# Recommonmark should be replaced so Sphinx can be upgraded
-# For now we remove other versions of Sphinx and pin it to v1.5.6
+**Step 4:** Make the docs.
-pip install --user \
- beautifulsoup4 \
- breathe \
- CommonMark==0.5.4 \
- h5py \
- mock==1.0.1 \
- pypandoc \
- recommonmark==0.4.0 \
- sphinx==1.5.6
+Here you have two options:
-# Setup scala
-echo "deb https://dl.bintray.com/sbt/debian /" | sudo tee -a /etc/apt/sources.list.d/sbt.list
-sudo apt-key adv --keyserver hkp://keyserver.ubuntu.com:80 --recv 2EE0EA64E40A89B84B2DF73499E82A75642AC823
-sudo apt-get update
-sudo apt-get install -y \
- sbt \
- scala
+* Build this current version (master) with the following:
-# Optionally setup Apache2
-sudo apt-get install -y apache2
-sudo ufw allow 'Apache Full'
-# turn on mod_rewrite
-sudo a2enmod rewrite
+```bash
+cd incubator-mxnet
+make docs USE_OPENMP=1
+```
-echo 'To enable redirects you need to edit /etc/apache2/apache2.conf '
-echo '--> Change directives for Directory for /var/www/html using the following: '
-echo ' AllowOverride all '
-echo '--> Then restart apache with: '
-echo ' sudo systemctl restart apache2'
+* Build all versions as what is seen in the production site. This will have the versions dropdown and any post-build processing that generates site artifacts and other requirements for the production site.
-# Cleanup
-sudo apt autoremove -y
+The following script will build all of the latest versions, set the default site to be `master` and use your dev server's IP or DNS for navigation items.
+
+```bash
+./build_site_tag.sh '1.2.0;1.1.0;1.0.0;0.12.0;0.11.0;master' master http://your-ip-or-dns
```
-### Full Website Build
+**Final Step:** Serve and test.
+
+Refer to [Serving Your Development Version](#serving-your-development-version) for detailed instructions.
+
+
+**Troubleshooting**: for AMI users or if you already have Conda, you might be stuck with the latest version and the docs build will have a conflict. To fix this, run `/home/ubuntu/anaconda3/bin/pip uninstall sphinx` and follow this with `pip install --user sphinx==1.5.6`.
+
+If you need to build <= v0.12.0, then use a Python 2 environment to avoid errors with `mxdoc.py`. This is a sphinx extension, that was not Python 3 compatible in the old versions. On the Deep Learning AMI, use `source activate mxnet_p27`, and then install the following dependencies.
+
+
+## Full Website Build
The following three scripts will help you build multiple version tags and deploy a full site build that with each available API version. If you just want to run master or your current fork's branch you should skip ahead to the [Developer Instructions](#developer-instructions).
The full site build scripts can be run stand-alone or in conjunction, but `build_all_version.sh` should be run first.
@@ -95,10 +79,10 @@ The full site build scripts can be run stand-alone or in conjunction, but `build
This will checkout each tag provided as an argument and build the docs in the `apache_mxnet` folder. The output is copied to the `VersionedWeb` folder and each version will have a subfolder in `VersionedWeb/versions/`.
Takes one argument:
-* **tag list** - space delimited list of Github tags; Example: "1.1.0 1.0.0 master"
+* **tag list** - semicolon delimited list of Github tags; Example: "1.1.0;1.0.0;master"
**Example Usage**:
-`./build_all_version.sh "1.1.0 1.0.0 0.12.1 0.12.0 0.11.0 master"`
+`./build_all_version.sh "1.1.0;1.0.0;0.12.1;0.12.0;0.11.0;master"`
### update_all_version.sh
This uses the output of `build_all_version.sh`. If you haven't built the specific tag yet, then you cannot update it.
@@ -111,7 +95,7 @@ Takes three arguments:
Each subfolder in `VersionedWeb/versions` will be processed with `AddVersion.py` and `AddPackageLink.py` to update the static html. Finally, the tag specified as the default tag, will be copied to the root of `VersionedWeb` to serve as MXNet.io's home (default) website. The other versions are accessible via the versions dropdown menu in the top level navigation of the website.
**Example Usage**:
-`./update_all_version.sh "1.1.0 1.0.0 0.12.1 0.12.0 0.11.0 master" 1.1.0 http://mxnet.incubator.apache.org/`
+`./update_all_version.sh "1.1.0;1.0.0;0.12.1;0.12.0;0.11.0;master" 1.1.0 http://mxnet.incubator.apache.org/`
### build_site_tag.sh
This one is useful for Docker, or to easily chain the two prior scripts. When you run the image you can call this script as a command a pass the tags, default tag, and root url.
@@ -120,7 +104,7 @@ Takes the same three arguments that update_all_version.sh takes.
It will execute `build_all_version.sh` first, then execute `update_all_version.sh` next.
**Example Usage**:
-./build_site_tag.sh "1.1.0 master" 1.1.0 http://mxnet.incubator.apache.org/
+./build_site_tag.sh "1.1.0;master" 1.1.0 http://mxnet.incubator.apache.org/
Then run a web server on the outputted `VersionedWeb` folder.
@@ -235,4 +219,63 @@ There are several manual and semi-automatic processes to be aware of, but the bo
1. The root should have the current `.htaccess` file from master in `/docs/`. Make sure you've updated this in master and included the most recent version in your PR.
2. The css file from master `/docs/_static/` will be needed. Be sure that the different versions of the site work. They might need the old version, but the newer version might fix bugs that were in the tags from the legacy versions.
3. Pay attention to `mxdocs.py` as some docs modifications are happening there.
-4. Review Any other modifications to the legacy versions can be seen in
+
+
+## Dependencies
+
+These are the dependencies for docs generation for Ubuntu 16.04.
+
+This script is available for you to run directly on Ubuntu from the source repository.
+Run `./setup_docs_ubuntu.sh`.
+
+```
+sudo apt-get update
+sudo apt-get install -y \
+ apt-transport-https \
+ ca-certificates \
+ curl \
+ doxygen \
+ git \
+ libjemalloc-dev \
+ pandoc \
+ software-properties-common
+
+# You made need to run `/home/ubuntu/anaconda3/bin/pip uninstall sphinx`
+# Recommonmark/Sphinx errors: https://github.com/sphinx-doc/sphinx/issues/3800
+# Recommonmark should be replaced so Sphinx can be upgraded
+# For now we remove other versions of Sphinx and pin it to v1.5.6
+
+pip install \
+ beautifulsoup4 \
+ breathe \
+ CommonMark==0.5.4 \
+ h5py \
+ mock==1.0.1 \
+ pypandoc \
+ recommonmark==0.4.0 \
+ sphinx==1.5.6
+
+# Setup scala
+echo "deb https://dl.bintray.com/sbt/debian /" | sudo tee -a /etc/apt/sources.list.d/sbt.list
+sudo apt-key adv --keyserver hkp://keyserver.ubuntu.com:80 --recv 2EE0EA64E40A89B84B2DF73499E82A75642AC823
+sudo apt-get update
+sudo apt-get install -y \
+ maven \
+ sbt \
+ scala
+
+# Optionally setup Apache2
+sudo apt-get install -y apache2
+sudo ufw allow 'Apache Full'
+# turn on mod_rewrite
+sudo a2enmod rewrite
+
+echo 'To enable redirects you need to edit /etc/apache2/apache2.conf '
+echo '--> Change directives for Directory for /var/www/html using the following: '
+echo ' AllowOverride all '
+echo '--> Then restart apache with: '
+echo ' sudo systemctl restart apache2'
+
+# Cleanup
+sudo apt autoremove -y
+```
diff --git a/docs/.htaccess b/docs/build_version_doc/artifacts/.htaccess
similarity index 100%
rename from docs/.htaccess
rename to docs/build_version_doc/artifacts/.htaccess
diff --git a/docs/build_version_doc/artifacts/0.12.0.index.html b/docs/build_version_doc/artifacts/0.12.0.index.html
new file mode 100644
index 00000000000..1852ae53dff
--- /dev/null
+++ b/docs/build_version_doc/artifacts/0.12.0.index.html
@@ -0,0 +1,1241 @@
+<!DOCTYPE html>
+
+<html lang="en">
+ <head>
+ <meta charset="utf-8">
+ <meta http-equiv="X-UA-Compatible" content="IE=edge">
+ <meta name="viewport" content="width=device-width, initial-scale=1">
+
+
+
+ <title>Installing MXNet — mxnet documentation</title>
+
+ <link rel="stylesheet" href="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.6/css/bootstrap.min.css" integrity="sha384-1q8mTJOASx8j1Au+a5WDVnPi2lkFfwwEAa8hDDdjZlpLegxhjVME1fgjWPGmkzs7" crossorigin="anonymous">
+ <link rel="stylesheet" href="https://maxcdn.bootstrapcdn.com/font-awesome/4.5.0/css/font-awesome.min.css">
+
+
+ <link rel="stylesheet" href="../_static/basic.css" type="text/css" />
+ <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+
+ <link rel="stylesheet" href="../_static/mxnet.css" type="text/css">
+
+ <script type="text/javascript">
+ var DOCUMENTATION_OPTIONS = {
+ URL_ROOT: '../',
+ VERSION: '',
+ COLLAPSE_INDEX: false,
+ FILE_SUFFIX: '.html',
+ HAS_SOURCE: true,
+ SOURCELINK_SUFFIX: '.txt'
+ };
+ </script>
+
+ <script type="text/javascript" src="../_static/jquery-1.11.1.js"></script>
+
+ <script type="text/javascript" src="../_static/underscore.js"></script>
+
+ <script type="text/javascript" src="../_static/searchtools_custom.js"></script>
+
+ <script type="text/javascript" src="../_static/doctools.js"></script>
+
+ <script type="text/javascript" src="../_static/selectlang.js"></script>
+
+
+ <script type="text/javascript" src="https://cdn.mathjax.org/mathjax/latest/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></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://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.0/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script> -->
+ <!-- -->
+ <link rel="index" title="Index" href="../genindex.html" />
+ <link rel="search" title="Search" href="../search.html" />
+
+ <link rel="icon" type="image/png" href="https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/mxnet-icon.png">
+ </head>
+ <body role="document" background="https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/mxnet-background.png">
+ <div class='content-block'><div class="navbar navbar-fixed-top">
+ <div class="container" id="navContainer">
+ <div id="header-inner" class="innder">
+ <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 id="main-nav" class='nav-bar'>
+ <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 href="#" class="main-nav-link dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">Gluon <span class="caret"></span></a>
+ <ul id="package-dropdown-menu" class="dropdown-menu navbar-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 href="#" class="main-nav-link dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">API <span class="caret"></span></a>
+ <ul id="package-dropdown-menu" class="dropdown-menu navbar-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 href="#" class="main-nav-link dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">Docs <span class="caret"></span></a>
+ <ul id="package-dropdown-menu-docs" class="dropdown-menu navbar-menu">
+ <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/master/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 href="#" class="main-nav-link dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">Community <span class="caret"></span></a>
+ <ul id="package-dropdown-menu-community" class="dropdown-menu navbar-menu">
+ <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>
+ </nav>
+ <script> function getRootPath(){ return "../" } </script>
+ <div class="burgerIcon dropdown">
+ <a href="#" class="dropdown-toggle" data-toggle="dropdown" role="button">☰</a>
+ <ul id="burgerMenu" class="dropdown-menu">
+ <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 tabindex="-1" href="../community/index.html">Community</a></li>
+ <li><a tabindex="-1" href="../community/contribute.html">Contribute</a></li>
+ <li><a tabindex="-1" href="../community/powered_by.html">Powered By</a></li>
+ </ul>
+ </li>
+
+ <li class="dropdown-submenu">
+ <a href="#" tabindex="-1">API</a>
+ <ul class="dropdown-menu">
+
+ <li><a tabindex="-1" href="../api/python/index.html">Python</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/scala/index.html">Scala</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/r/index.html">R</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/julia/index.html">Julia</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/c++/index.html">C++</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/perl/index.html">Perl</a>
+ </li>
+
+ </ul>
+ </li>
+
+ <li class="dropdown-submenu">
+ <a href="#" tabindex="-1">Docs</a>
+ <ul class="dropdown-menu">
+ <li><a tabindex="-1" href="../tutorials/index.html">Tutorials</a></li>
+ <li><a tabindex="-1" href="../faq/index.html">FAQ</a></li>
+ <li><a tabindex="-1" href="../architecture/index.html">Architecture</a></li>
+ <li><a tabindex="-1" href="https://github.com/apache/incubator-mxnet/tree/master/example">Examples</a></li>
+ <li><a tabindex="-1" href="../model_zoo/index.html">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>
+ </ul>
+ </div>
+ <div class="plusIcon dropdown">
+ <a href="#" class="dropdown-toggle" data-toggle="dropdown" role="button"><span class="glyphicon glyphicon-plus" aria-hidden="true"></span></a>
+ <ul id="plusMenu" class="dropdown-menu dropdown-menu-right"></ul>
+ </div>
+ <div id="search-input-wrap">
+
+<form class="" role="search" action="../search.html" method="get" autocomplete="off">
+ <div class="form-group inner-addon left-addon">
+ <i class="glyphicon glyphicon-search"></i>
+ <input type="text" name="q" class="form-control" placeholder="Search">
+ </div>
+ <input type="hidden" name="check_keywords" value="yes" />
+ <input type="hidden" name="area" value="default" />
+
+</form>
+ <div id="search-preview"></div>
+ </div>
+ <div id='searchIcon'>
+ <span class="glyphicon glyphicon-search" aria-hidden="true"></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 class="sphinxsidebar leftsidebar" role="navigation" aria-label="main 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>
+
+ <div class="section" id="installing-mxnet">
+<span id="installing-mxnet"></span><h1>Installing MXNet<a class="headerlink" href="#installing-mxnet" title="Permalink to this headline">¶</a></h1>
+<p>Indicate your preferred configuration. Then, follow the customized commands to install <em>MXNet</em>.</p>
+<script type="text/javascript" src='../_static/js/options.js'></script><!-- START - OS Menu --><div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">Linux</button>
+ <button type="button" class="btn btn-default opt">MacOS</button>
+ <button type="button" class="btn btn-default opt">Windows</button>
+ <button type="button" class="btn btn-default opt">Cloud</button>
+ <button type="button" class="btn btn-default opt">Devices</button>
+</div><!-- START - Language Menu --><div class="linux macos windows">
+<div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">Python</button>
+ <button type="button" class="btn btn-default opt">Scala</button>
+ <button type="button" class="btn btn-default opt">R</button>
+ <button type="button" class="btn btn-default opt">Julia</button>
+ <button type="button" class="btn btn-default opt">Perl</button>
+</div>
+</div><!-- No CPU GPU for other Devices -->
+<div class="linux macos windows cloud">
+<div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">CPU</button>
+ <button type="button" class="btn btn-default opt">GPU</button>
+</div>
+</div><!-- other devices -->
+<div class="devices">
+<div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">Raspberry Pi</button>
+ <button type="button" class="btn btn-default opt">NVIDIA Jetson TX2</button>
+</div>
+</div><!-- Linux Python GPU Options --><div class="linux macos">
+<div class="python">
+<div class="cpu gpu">
+<div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">Pip</button>
+ <button type="button" class="btn btn-default opt">Virtualenv</button>
+ <button type="button" class="btn btn-default opt">Docker</button>
+ <button type="button" class="btn btn-default opt">Build from Source</button>
+</div>
+</div>
+</div>
+</div><!-- END - Main Menu --><!-- START - Linux Python CPU Installation Instructions --><div class="linux">
+ <div class="python">
+ <div class="cpu"><p>The following installation instructions have been tested on Ubuntu 14.04 and 16.04.</p>
+<div class="virtualenv">
+<br/><p><strong>Step 1</strong> Install virtualenv for Ubuntu.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y python-dev python-virtualenv
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Create and activate virtualenv environment for MXNet.</p>
+<p>Following command creates a virtualenv environment at <code class="docutils literal"><span class="pre">~/mxnet</span></code> directory. However, you can choose any directory by replacing <code class="docutils literal"><span class="pre">~/mxnet</span></code> with a directory of your choice.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ virtualenv --system-site-packages ~/mxnet
+</pre></div>
+</div>
+<p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install MXNet in the active virtualenv environment.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install --upgrade pip
+</pre></div>
+</div>
+<p>Install <em>MXNet</em> with OpenBLAS acceleration.</p>
+<p>To install MXNet 0.12.1 use the following:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet==0.12.1 </pre></div></div>
+<p>To install MXNet 0.12.0 use the following:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet==0.12.0 </pre></div>
+</div>
+<p><strong>Step 4</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 5</strong> Validate the installation by running simple <em>MXNet</em> code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Note</strong> You can read more about virtualenv <a class="reference external" href="https://virtualenv.pypa.io/en/stable/userguide/">here</a>.</p>
+</div><div class="pip">
+<br/><p><strong>Step 1</strong> Install prerequisites - wget and latest pip.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command in the terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y wget python gcc
+$ wget https://bootstrap.pypa.io/get-pip.py <span class="o">&&</span> sudo python get-pip.py
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install MXNet with OpenBLAS acceleration.</p>
+<p>To install MXNet 0.12.1 use the following: </p><div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet==0.12.1</pre></div> </div>
+<p>To install MXNet 0.12.0 use the following: </p><div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet==0.12.0</pre></div> </div>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Experimental Choice</strong> If You would like to install mxnet with Intel MKL, try the experimental pip package with MKL:</p>
+<p>To install MXNet 0.12.1 with MKL use the following: </p><div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet-mkl==0.12.1</pre></div> </div>
+<p>To install MXNet 0.12.0 with MKL use the following: </p><div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet-mkl==0.12.0</pre></div> </div>
+</div><div class="docker">
+<br/><p>Docker images with <em>MXNet</em> are available at <a class="reference external" href="https://hub.docker.com/r/mxnet/">Docker Hub</a>.</p>
+<p><strong>Step 1</strong> Install Docker on your machine by following the <a class="reference external" href="https://docs.docker.com/engine/installation/linux/ubuntu/#install-using-the-repository">docker installation instructions</a>.</p>
+<p><em>Note</em> - You can install Community Edition (CE) to get started with <em>MXNet</em>.</p>
+<p><strong>Step 2</strong> [Optional] Post installation steps to manage Docker as a non-root user.</p>
+<p>Follow the four steps in this <a class="reference external" href="https://docs.docker.com/engine/installation/linux/linux-postinstall/#manage-docker-as-a-non-root-user">docker documentation</a> to allow managing docker containers without <em>sudo</em>.</p>
+<p>If you skip this step, you need to use <em>sudo</em> each time you invoke Docker.</p>
+<p><strong>Step 2</strong> Pull the MXNet docker image.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker pull mxnet/python <span class="c1"># Use sudo if you skip Step 2</span>
+</pre></div>
+</div>
+<p>You can list docker images to see if mxnet/python docker image pull was successful.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker images <span class="c1"># Use sudo if you skip Step 2</span>
+
+REPOSITORY TAG IMAGE ID CREATED SIZE
+mxnet/python latest 00d026968b3c <span class="m">3</span> weeks ago <span class="m">1</span>.41 GB
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div><div class="build-from-source">
+<br/><p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings. Example - Python bindings, Scala bindings.</li>
+</ol>
+<p><strong>Minimum Requirements</strong></p>
+<ol class="simple">
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">GCC 4.8</a> or later to compile C++ 11.</li>
+<li><a class="reference external" href="https://www.gnu.org/software/make/">GNU Make</a></li>
+</ol>
+<p><br/></p>
+<p><strong>Build the MXNet core shared library</strong></p>
+<p><strong>Step 1</strong> Install build tools and git.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y build-essential git
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install OpenBLAS.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="https://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprograms">BLAS</a> and <a class="reference external" href="https://en.wikipedia.org/wiki/LAPACK">LAPACK</a> libraries for accelerated numerical computations on CPU machine. There are several flavors of BLAS/LAPACK libraries - <a class="reference external" href="http://www.openblas.net/">OpenBLAS</a>, <a class="reference external" href="http://math-atlas.sourceforge.net/">ATLAS</a> and <a class="reference external" href="https://software.intel.com/en-us/intel-mkl">MKL</a>. In this step we install OpenBLAS. You can choose to install ATLAS or MKL.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopenblas-dev liblapack-dev
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install OpenCV.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="http://opencv.org/">OpenCV</a> for efficient image loading and augmentation operations.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopencv-dev
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Download MXNet sources and build MXNet core shared library. You can clone the repository as described in the following code block, or you may try the <a href="download.html">download links</a> for your desired MXNet version.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ git clone --recursive https://github.com/dmlc/mxnet
+$ <span class="nb">cd</span> mxnet
+$ make -j <span class="k">$(</span>nproc<span class="k">)</span> <span class="nv">USE_OPENCV</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_BLAS</span><span class="o">=</span>openblas
+</pre></div>
+</div>
+<p><em>Note</em> - USE_OPENCV and USE_BLAS are make file flags to set compilation options to use OpenCV and BLAS library. You can explore and use more compilation options in <code class="docutils literal"><span class="pre">make/config.mk</span></code>.</p>
+<p><br/></p>
+<p><strong>Build the MXNet Python binding</strong></p>
+<p><strong>Step 1</strong> Install prerequisites - python, setup-tools, python-pip and numpy.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y python-dev python-setuptools python-numpy python-pip
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install the MXNet Python binding.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">cd</span> python
+$ pip install --upgrade pip
+$ pip install -e .
+</pre></div>
+</div>
+<p>Note that the <code class="docutils literal"><span class="pre">-e</span></code> flag is optional. It is equivalent to <code class="docutils literal"><span class="pre">--editable</span></code> and means that if you edit the source files, these changes will be reflected in the package installed.</p>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div></div>
+</div>
+</div>
+<!-- END - Linux Python CPU Installation Instructions --><!-- START - Linux Python GPU Installation Instructions --><div class="linux">
+<div class="python">
+<div class="gpu"><p>The following installation instructions have been tested on Ubuntu 14.04 and 16.04.</p>
+<p><strong>Prerequisites</strong></p>
+<p>Install the following NVIDIA libraries to setup <em>MXNet</em> with GPU support:</p>
+<ol class="simple">
+<li>Install CUDA 8.0 following the NVIDIA’s <a class="reference external" href="http://docs.nvidia.com/cuda/cuda-installation-guide-linux/">installation guide</a>.</li>
+<li>Install cuDNN 5 for CUDA 8.0 following the NVIDIA’s <a class="reference external" href="https://developer.nvidia.com/cudnn">installation guide</a>. You may need to register with NVIDIA for downloading the cuDNN library.</li>
+</ol>
+<p><strong>Note:</strong> Make sure to add CUDA install path to <code class="docutils literal"><span class="pre">LD_LIBRARY_PATH</span></code>.</p>
+<p>Example - <em>export LD_LIBRARY_PATH=/usr/local/cuda/lib64/:$LD_LIBRARY_PATH</em></p>
+<div class="pip">
+<br/><p><strong>Step 1</strong> Install prerequisites - wget and latest pip.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command in the terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y wget python
+$ wget https://bootstrap.pypa.io/get-pip.py <span class="o">&&</span> sudo python get-pip.py
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install <em>MXNet</em> with GPU support using CUDA 8.0</p>
+<p>To install MXNet 0.12.1 use the following: </p><div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet-cu80==0.12.1</pre></div> </div>
+<p>To install MXNet 0.12.0 use the following: </p><div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet-cu80==0.12.0</pre></div> </div>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Experimental Choice</strong> If You would like to install mxnet with Intel MKL, try the experimental pip package with MKL:</p>
+<p>To install MXNet 0.12.1 with MKL use the following: </p><div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet-cumkl==0.12.1</pre></div> </div>
+<p>To install MXNet 0.12.0 with MKL use the following: </p><div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet-cumkl==0.12.0</pre></div> </div>
+</div><div class="virtualenv"><p><br/></p>
+<p><strong>Step 1</strong> Install virtualenv for Ubuntu.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y python-dev python-virtualenv
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Create and activate virtualenv environment for MXNet.</p>
+<p>Following command creates a virtualenv environment at <code class="docutils literal"><span class="pre">~/mxnet</span></code> directory. However, you can choose any directory by replacing <code class="docutils literal"><span class="pre">~/mxnet</span></code> with a directory of your choice.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ virtualenv --system-site-packages ~/mxnet
+</pre></div>
+</div>
+<p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install MXNet in the active virtualenv environment.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install --upgrade pip
+</pre></div>
+</div>
+<p>Install <em>MXNet</em> with GPU support using CUDA 8.0.</p>
+<p>To install MXNet 0.12.1 use the following: </p><div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet-cu80==0.12.1</pre></div> </div>
+<p>To install MXNet 0.12.0 use the following: </p><div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet-cu80==0.12.0</pre></div> </div>
+<p><strong>Step 4</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 5</strong> Validate the installation by running simple <em>MXNet</em> code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Note</strong> You can read more about virtualenv <a class="reference external" href="https://virtualenv.pypa.io/en/stable/userguide/">here</a>.</p>
+</div><div class="docker"><p><br/></p>
+<p>Docker images with <em>MXNet</em> are available at <a class="reference external" href="https://hub.docker.com/r/mxnet/">Docker Hub</a>.</p>
+<p><strong>Step 1</strong> Install Docker on your machine by following the <a class="reference external" href="https://docs.docker.com/engine/installation/linux/ubuntu/#install-using-the-repository">docker installation instructions</a>.</p>
+<p><em>Note</em> - You can install Community Edition (CE) to get started with <em>MXNet</em>.</p>
+<p><strong>Step 2</strong> [Optional] Post installation steps to manage Docker as a non-root user.</p>
+<p>Follow the four steps in this <a class="reference external" href="https://docs.docker.com/engine/installation/linux/linux-postinstall/#manage-docker-as-a-non-root-user">docker documentation</a> to allow managing docker containers without <em>sudo</em>.</p>
+<p>If you skip this step, you need to use <em>sudo</em> each time you invoke Docker.</p>
+<p><strong>Step 3</strong> Install <em>nvidia-docker-plugin</em> following the <a class="reference external" href="https://github.com/NVIDIA/nvidia-docker/wiki/Installation">installation instructions</a>. <em>nvidia-docker-plugin</em> is required to enable the usage of GPUs from the docker containers.</p>
+<p><strong>Step 4</strong> Pull the MXNet docker image.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker pull mxnet/python:gpu <span class="c1"># Use sudo if you skip Step 2</span>
+</pre></div>
+</div>
+<p>You can list docker images to see if mxnet/python docker image pull was successful.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker images <span class="c1"># Use sudo if you skip Step 2</span>
+
+REPOSITORY TAG IMAGE ID CREATED SIZE
+mxnet/python gpu 493b2683c269 <span class="m">3</span> weeks ago <span class="m">4</span>.77 GB
+</pre></div>
+</div>
+<p><strong>Step 5</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div><div class="build-from-source"><p><br/></p>
+<p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings. Example - Python bindings, Scala bindings.</li>
+</ol>
+<p><strong>Minimum Requirements</strong></p>
+<ol class="simple">
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">GCC 4.8</a> or later to compile C++ 11.</li>
+<li><a class="reference external" href="https://www.gnu.org/software/make/">GNU Make</a></li>
+</ol>
+<p><br/></p>
+<p><strong>Build the MXNet core shared library</strong></p>
+<p><strong>Step 1</strong> Install build tools and git.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y build-essential git
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install OpenBLAS.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="https://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprograms">BLAS</a> and <a class="reference external" href="https://en.wikipedia.org/wiki/LAPACK">LAPACK</a> libraries for accelerated numerical computations on CPU machine. There are several flavors of BLAS/LAPACK libraries - <a class="reference external" href="http://www.openblas.net/">OpenBLAS</a>, <a class="reference external" href="http://math-atlas.sourceforge.net/">ATLAS</a> and <a class="reference external" href="https://software.intel.com/en-us/intel-mkl">MKL</a>. In this step we install OpenBLAS. You can choose to install ATLAS or MKL.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopenblas-dev liblapack-dev
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install OpenCV.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="http://opencv.org/">OpenCV</a> for efficient image loading and augmentation operations.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopencv-dev
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Download MXNet sources and build MXNet core shared library. You can clone the repository as described in the following code block, or you may try the <a href="download.html">download links</a> for your desired MXNet version.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ git clone --recursive https://github.com/dmlc/mxnet
+$ <span class="nb">cd</span> mxnet
+$ make -j <span class="k">$(</span>nproc<span class="k">)</span> <span class="nv">USE_OPENCV</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_BLAS</span><span class="o">=</span>openblas <span class="nv">USE_CUDA</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_CUDA_PATH</span><span class="o">=</span>/usr/local/cuda <span class="nv">USE_CUDNN</span><span class="o">=</span><span class="m">1</span>
+</pre></div>
+</div>
+<p><em>Note</em> - USE_OPENCV, USE_BLAS, USE_CUDA, USE_CUDA_PATH AND USE_CUDNN are make file flags to set compilation options to use OpenCV, OpenBLAS, CUDA and cuDNN libraries. You can explore and use more compilation options in <code class="docutils literal"><span class="pre">make/config.mk</span></code>. Make sure to set USE_CUDA_PATH to right CUDA installation path. In most cases it is - <em>/usr/local/cuda</em>.</p>
+<p><br/></p>
+<p><strong>Install the MXNet Python binding</strong></p>
+<p><strong>Step 1</strong> Install prerequisites - python, setup-tools, python-pip and numpy.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y python-dev python-setuptools python-numpy python-pip
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install the MXNet Python binding.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">cd</span> python
+$ pip install --upgrade pip
+$ pip install -e .
+</pre></div>
+</div>
+<p>Note that the <code class="docutils literal"><span class="pre">-e</span></code> flag is optional. It is equivalent to <code class="docutils literal"><span class="pre">--editable</span></code> and means that if you edit the source files, these changes will be reflected in the package installed.</p>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div></div>
+</div>
+</div>
+<!-- END - Linux Python GPU Installation Instructions --><!-- START - MacOS Python CPU Installation Instructions --><div class="macos">
+ <div class="python">
+ <div class="cpu"><p>The following installation instructions have been tested on OSX Sierra and El Capitan.</p>
+<div class="virtualenv">
+<br/><p><strong>Step 1</strong> Install prerequisites - Homebrew, python development tools.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="c1"># Install Homebrew</span>
+$ /usr/bin/ruby -e <span class="s2">"</span><span class="k">$(</span>curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install<span class="k">)</span><span class="s2">"</span>
+$ <span class="nb">export</span> <span class="nv">PATH</span><span class="o">=</span>/usr/local/bin:/usr/local/sbin:<span class="nv">$PATH</span>
+
+<span class="c1"># Install python development tools - python2.7, pip, python-setuptools</span>
+$ brew install python
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install virtualenv for macOS.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install virtualenv
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Create and activate virtualenv environment for MXNet.</p>
+<p>Following command creates a virtualenv environment at <code class="docutils literal"><span class="pre">~/mxnet</span></code> directory. However, you can choose any directory by replacing <code class="docutils literal"><span class="pre">~/mxnet</span></code> with a directory of your choice.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ virtualenv --system-site-packages ~/mxnet
+</pre></div>
+</div>
+<p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Install MXNet in the active virtualenv environment.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install --upgrade pip
+<span class="o">(</span>mxnet<span class="o">)</span>$ pip install --upgrade setuptools
+</pre></div>
+</div>
+<p>Install <em>MXNet</em> with OpenBLAS acceleration.</p>
+<p>To install MXNet 0.12.1 use the following: </p><div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet==0.12.1</pre></div> </div>
+<p>To install MXNet 0.12.0 use the following: </p><div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet==0.12.0</pre></div> </div>
+<p><strong>Step 5</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ brew install graphviz
+<span class="o">(</span>mxnet<span class="o">)</span>$ pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 6</strong> Validate the installation by running simple <em>MXNet</em> code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Note</strong> You can read more about virtualenv <a class="reference external" href="https://virtualenv.pypa.io/en/stable/userguide/">here</a>.</p>
+</div><div class="pip">
+<br/><p><strong>Step 1</strong> Install prerequisites - Homebrew, python development tools.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="c1"># Install Homebrew</span>
+$ /usr/bin/ruby -e <span class="s2">"</span><span class="k">$(</span>curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install<span class="k">)</span><span class="s2">"</span>
+$ <span class="nb">export</span> <span class="nv">PATH</span><span class="o">=</span>/usr/local/bin:/usr/local/sbin:<span class="nv">$PATH</span>
+
+<span class="c1"># Install python development tools - python2.7, pip, python-setuptools</span>
+$ brew install python
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install MXNet with OpenBLAS acceleration.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install --upgrade pip
+$ pip install --upgrade setuptools
+</pre></div>
+</div>
+<p>To install MXNet 0.12.1 use the following: </p><div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet==0.12.1</pre></div> </div>
+<p>To install MXNet 0.12.0 use the following: </p><div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet==0.12.0</pre></div> </div>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ brew install graphviz
+$ pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div><div class="docker">
+<br/><p>Docker images with <em>MXNet</em> are available at <a class="reference external" href="https://hub.docker.com/r/mxnet/">Docker Hub</a>.</p>
+<p><strong>Step 1</strong> Install Docker on your machine by following the <a class="reference external" href="https://docs.docker.com/docker-for-mac/install/#install-and-run-docker-for-mac">docker installation instructions</a>.</p>
+<p><em>Note</em> - You can install Community Edition (CE) to get started with <em>MXNet</em>.</p>
+<p><strong>Step 2</strong> Pull the MXNet docker image.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker pull mxnet/python
+</pre></div>
+</div>
+<p>You can list docker images to see if mxnet/python docker image pull was successful.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker images
+
+REPOSITORY TAG IMAGE ID CREATED SIZE
+mxnet/python latest 00d026968b3c <span class="m">3</span> weeks ago <span class="m">1</span>.41 GB
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div><div class="build-from-source">
+<br/><p><strong>Prerequisites</strong></p>
+<p>If not already installed, <a class="reference external" href="https://developer.apple.com/xcode/">download and install Xcode</a> (or <a class="reference external" href="https://itunes.apple.com/us/app/xcode/id497799835">insall it from the App Store</a>) for macOS. <a class="reference external" href="https://en.wikipedia.org/wiki/Xcode">Xcode</a> is an integrated development environment for macOS containing a suite of software development tools like C/C++ compilers, BLAS library and more.</p>
+<p><br/></p>
+<p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings. Example - Python bindings, Scala bindings.</li>
+</ol>
+<p>Make sure you have installed Xcode before proceeding further.</p>
+<p><br/></p>
+<p>All the instructions to build <em>MXNet</em> core shared library and <em>MXNet</em> Python bindings are compiled as one helper <em>bash</em> script. You can use <a class="reference external" href="https://raw.githubusercontent.com/dmlc/mxnet/master/setup-utils/install-mxnet-osx-python.sh">this bash script</a> to build <em>MXNet</em> for Python, from source, on macOS.</p>
+<p><strong>Step 1</strong> Download the bash script for building MXNet from source.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ curl -O https://raw.githubusercontent.com/dmlc/mxnet/master/setup-utils/install-mxnet-osx-python.sh
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Run the script to get latest MXNet source and build.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="c1"># Make the script executable</span>
+$ chmod <span class="m">744</span> install-mxnet-osx-python.sh
+
+<span class="c1"># Run the script. It takes around 5 mins.</span>
+$ bash install-mxnet-osx-python.sh
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div> <!-- End of source build --></div>
+</div>
+</div><!-- END - Mac OS Python CPU Installation Instructions --><!-- START - Mac OS Python GPU Installation Instructions --><div class="macos">
+ <div class="python">
+ <div class="gpu"><p>More details and verified installation instructions for macOS, with GPUs, coming soon.</p>
+<p><em>MXNet</em> is expected to be compatible on macOS with NVIDIA GPUs. Please install CUDA 8.0 and cuDNN 5.0, prior to installing GPU version of <em>MXNet</em>.</p>
+</div>
+</div>
+</div><!-- END - Mac OS Python GPU Installation Instructions --><!-- START - Cloud Python Installation Instructions --><div class="cloud"><p>AWS Marketplace distributes AMIs (Amazon Machine Image) with MXNet pre-installed. You can launch an Amazon EC2 instance with one of the below AMIs:</p>
+<ol class="simple">
+<li>Deep Learning AMI (Amazon Machine Image) for <a class="reference external" href="https://aws.amazon.com/marketplace/pp/B06VSPXKDX">Ubuntu</a></li>
+<li>Deep Learning AMI for <a class="reference external" href="https://aws.amazon.com/marketplace/pp/B01M0AXXQB">Amazon Linux</a></li>
+</ol>
+<p>You could also run distributed deeplearning with <em>MXNet</em> on AWS using <a class="reference external" href="https://github.com/awslabs/deeplearning-cfn/blob/master/README.md">Cloudformation Template</a>.</p>
+</div><!-- END - Cloud Python Installation Instructions --><!-- START - MacOS R CPU Installation Instructions --><div class="macos">
+ <div class="r">
+ <div class="cpu"><p>The CPU version of MXNet R package can be installed in R like other packages</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span>cran <span class="o"><-</span> <span class="kp">getOption</span><span class="p">(</span><span class="s">"repos"</span><span class="p">)</span>
+cran<span class="p">[</span><span class="s">"dmlc"</span><span class="p">]</span> <span class="o"><-</span> <span class="s">"https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/R/CRAN/"</span>
+<span class="kp">options</span><span class="p">(</span>repos <span class="o">=</span> cran<span class="p">)</span>
+install.packages<span class="p">(</span><span class="s">"mxnet"</span><span class="p">)</span>
+</pre></div>
+</div>
+</div><div class="gpu"><p>Will be available soon.</p>
+</div></div>
+</div>
+<!-- END - MacOS R CPU Installation Instructions --><div class="linux">
+ <div class="r">
+ <div class="cpu">
+<br/><p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings.</li>
+</ol>
+<p><strong>Minimum Requirements</strong></p>
+<ol class="simple">
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">GCC 4.8</a> or later to compile C++ 11.</li>
+<li><a class="reference external" href="https://www.gnu.org/software/make/">GNU Make</a></li>
+</ol>
+<p><br/></p>
+<p><strong>Build the MXNet core shared library</strong></p>
+<p><strong>Step 1</strong> Install build tools and git.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y build-essential git
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install OpenBLAS.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="https://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprograms">BLAS</a> and <a class="reference external" href="https://en.wikipedia.org/wiki/LAPACK">LAPACK</a> libraries for accelerated numerical computations on CPU machine. There are several flavors of BLAS/LAPACK libraries - <a class="reference external" href="http://www.openblas.net/">OpenBLAS</a>, <a class="reference external" href="http://math-atlas.sourceforge.net/">ATLAS</a> and <a class="reference external" href="https://software.intel.com/en-us/intel-mkl">MKL</a>. In this step we install OpenBLAS. You can choose to install ATLAS or MKL.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopenblas-dev liblapack-dev
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install OpenCV.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="http://opencv.org/">OpenCV</a> for efficient image loading and augmentation operations.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopencv-dev
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Download MXNet sources and build MXNet core shared library. You can clone the repository as described in the following code block, or you may try the <a href="download.html">download links</a> for your desired MXNet version.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ git clone --recursive https://github.com/dmlc/mxnet
+$ <span class="nb">cd</span> mxnet
+$ make -j <span class="k">$(</span>nproc<span class="k">)</span> <span class="nv">USE_OPENCV</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_BLAS</span><span class="o">=</span>openblas
+</pre></div>
+</div>
+<p><em>Note</em> - USE_OPENCV and USE_BLAS are make file flags to set compilation options to use OpenCV and BLAS library. You can explore and use more compilation options in <code class="docutils literal"><span class="pre">make/config.mk</span></code>.</p>
+<p><br/></p>
+<p><strong>Build and install the MXNet R binding</strong></p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ make rpkg
+$ R CMD INSTALL mxnet_current_r.tar.gz
+</pre></div>
+</div>
+</div><div class="gpu"><p>The following installation instructions have been tested on Ubuntu 14.04 and 16.04.</p>
+<p><strong>Prerequisites</strong></p>
+<p>Install the following NVIDIA libraries to setup <em>MXNet</em> with GPU support:</p>
+<ol class="simple">
+<li>Install CUDA 8.0 following the NVIDIA’s <a class="reference external" href="http://docs.nvidia.com/cuda/cuda-installation-guide-linux/">installation guide</a>.</li>
+<li>Install cuDNN 5 for CUDA 8.0 following the NVIDIA’s <a class="reference external" href="https://developer.nvidia.com/cudnn">installation guide</a>. You may need to register with NVIDIA for downloading the cuDNN library.</li>
+</ol>
+<p><strong>Note:</strong> Make sure to add CUDA install path to <code class="docutils literal"><span class="pre">LD_LIBRARY_PATH</span></code>.</p>
+<p>Example - <em>export LD_LIBRARY_PATH=/usr/local/cuda/lib64/:$LD_LIBRARY_PATH</em></p>
+<p><br/></p>
+<p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings.</li>
+</ol>
+<p><strong>Minimum Requirements</strong></p>
+<ol class="simple">
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">GCC 4.8</a> or later to compile C++ 11.</li>
+<li><a class="reference external" href="https://www.gnu.org/software/make/">GNU Make</a></li>
+</ol>
+<p><br/></p>
+<p><strong>Build the MXNet core shared library</strong></p>
+<p><strong>Step 1</strong> Install build tools and git.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y build-essential git
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install OpenBLAS.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="https://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprograms">BLAS</a> and <a class="reference external" href="https://en.wikipedia.org/wiki/LAPACK">LAPACK</a> libraries for accelerated numerical computations on CPU machine. There are several flavors of BLAS/LAPACK libraries - <a class="reference external" href="http://www.openblas.net/">OpenBLAS</a>, <a class="reference external" href="http://math-atlas.sourceforge.net/">ATLAS</a> and <a class="reference external" href="https://software.intel.com/en-us/intel-mkl">MKL</a>. In this step we install OpenBLAS. You can choose to install ATLAS or MKL.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopenblas-dev liblapack-dev
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install OpenCV.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="http://opencv.org/">OpenCV</a> for efficient image loading and augmentation operations.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopencv-dev
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Download MXNet sources and build MXNet core shared library. You can clone the repository as described in the following code block, or you may try the <a href="download.html">download links</a> for your desired MXNet version.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ git clone --recursive https://github.com/dmlc/mxnet
+$ <span class="nb">cd</span> mxnet
+$ make -j <span class="k">$(</span>nproc<span class="k">)</span> <span class="nv">USE_OPENCV</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_BLAS</span><span class="o">=</span>openblas <span class="nv">USE_CUDA</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_CUDA_PATH</span><span class="o">=</span>/usr/local/cuda <span class="nv">USE_CUDNN</span><span class="o">=</span><span class="m">1</span>
+</pre></div>
+</div>
+<p><em>Note</em> - USE_OPENCV, USE_BLAS, USE_CUDA, USE_CUDA_PATH AND USE_CUDNN are make file flags to set compilation options to use OpenCV, OpenBLAS, CUDA and cuDNN libraries. You can explore and use more compilation options in <code class="docutils literal"><span class="pre">make/config.mk</span></code>. Make sure to set USE_CUDA_PATH to right CUDA installation path. In most cases it is - <em>/usr/local/cuda</em>.</p>
+<p><br/></p>
+<p><strong>Build and install the MXNet R binding</strong></p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ make rpkg
+$ R CMD INSTALL mxnet_current_r.tar.gz
+</pre></div>
+</div>
+</div></div>
+</div><!-- START - Windows R CPU Installation Instructions --><div class="windows">
+ <div class="r">
+ <div class="cpu"><p>The CPU version of MXNet R package can be installed in R like other packages</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span>cran <span class="o"><-</span> <span class="kp">getOption</span><span class="p">(</span><span class="s">"repos"</span><span class="p">)</span>
+cran<span class="p">[</span><span class="s">"dmlc"</span><span class="p">]</span> <span class="o"><-</span> <span class="s">"https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/R/CRAN/"</span>
+<span class="kp">options</span><span class="p">(</span>repos <span class="o">=</span> cran<span class="p">)</span>
+install.packages<span class="p">(</span><span class="s">"mxnet"</span><span class="p">)</span>
+</pre></div>
+</div>
+</div><!-- END - Windows R CPU Installation Instructions --><div class="gpu"><p>The GPU version of MXNet R package can be installed in R like other packages</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span>cran <span class="o"><-</span> <span class="kp">getOption</span><span class="p">(</span><span class="s">"repos"</span><span class="p">)</span>
+cran<span class="p">[</span><span class="s">"dmlc"</span><span class="p">]</span> <span class="o"><-</span> <span class="s">"https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/R/CRAN/GPU"</span>
+<span class="kp">options</span><span class="p">(</span>repos <span class="o">=</span> cran<span class="p">)</span>
+install.packages<span class="p">(</span><span class="s">"mxnet"</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>Alternatively, You can also follow the installation instructions <a class="reference internal" href="windows_setup.html"><span class="doc">in this guide</span></a> to build MXNet from source.</p>
+</div>
+</div>
+</div><!-- END - Windows R GPU Installation Instructions --><div class="linux">
+ <div class="scala julia perl">
+ <div class="cpu gpu"><p>Follow the installation instructions <a class="reference internal" href="ubuntu_setup.html"><span class="doc">in this guide</span></a> to set up MXNet.</p>
+</div>
+</div>
+</div><div class="macos">
+ <div class="scala julia perl">
+ <div class="cpu gpu"><p>Follow the installation instructions <a class="reference internal" href="osx_setup.html"><span class="doc">in this guide</span></a> to set up MXNet.</p>
+</div>
+</div>
+</div><div class="windows">
+ <div class="python scala julia perl">
+ <div class="gpu"><p>Follow the installation instructions <a class="reference internal" href="windows_setup.html"><span class="doc">in this guide</span></a> to set up MXNet.</p>
+</div>
+</div>
+</div><div class="devices">
+ <div class="raspberry-pi"><p>MXNet supports the Debian based Raspbian ARM based operating system so you can run MXNet on Raspberry Pi Devices.</p>
+<p>These instructions will walk through how to build MXNet for the Raspberry Pi and install the Python bindings for the library.</p>
+<p>The complete MXNet library and its requirements can take almost 200MB of RAM, and loading large models with the library can take over 1GB of RAM. Because of this, we recommend running MXNet on the Raspberry Pi 3 or an equivalent device that has more than 1 GB of RAM and a Secure Digital (SD) card that has at least 4 GB of free memory.</p>
+<p><strong>Install MXNet</strong></p>
+<p>Installing MXNet is a two-step process:</p>
+<ol class="simple">
+<li>Build the shared library from the MXNet C++ source code.</li>
+<li>Install the supported language-specific packages for MXNet.</li>
+</ol>
+<p><strong>Step 1</strong> Build the Shared Library</p>
+<p>On Raspbian versions Wheezy and later, you need the following dependencies:</p>
+<ul class="simple">
+<li>Git (to pull code from GitHub)</li>
+<li>libblas (for linear algebraic operations)</li>
+<li>libopencv (for computer vision operations. This is optional if you want to save RAM and Disk Space)</li>
+<li>A C++ compiler that supports C++ 11. The C++ compiler compiles and builds MXNet source code. Supported compilers include the following:</li>
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">G++ (4.8 or later)</a></li>
+</ul>
+<p>Install these dependencies using the following commands in any directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> sudo apt-get update
+ sudo apt-get -y install git cmake build-essential g++-4.8 c++-4.8 liblapack* libblas* libopencv*
+</pre></div>
+</div>
+<p>Clone the MXNet source code repository using the following <code class="docutils literal"><span class="pre">git</span></code> command in your home directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> git clone https://github.com/dmlc/mxnet.git --recursive
+ <span class="nb">cd</span> mxnet
+</pre></div>
+</div>
+<p>If you aren’t processing images with MXNet on the Raspberry Pi, you can minimize the size of the compiled library by building MXNet without the Open Source Computer Vision (OpenCV) library with the following commands:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> <span class="nb">export</span> <span class="nv">USE_OPENCV</span> <span class="o">=</span> <span class="m">0</span>
+ make
+</pre></div>
+</div>
+<p>Otherwise, you can build the complete MXNet library with the following command:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> make
+</pre></div>
+</div>
+<p>Executing either of these commands start the build process, which can take up to a couple hours, and creates a file called <code class="docutils literal"><span class="pre">libmxnet.so</span></code> in the mxnet/lib directory.</p>
+<p>If you are getting build errors in which the compiler is being killed, it is likely that the compiler is running out of memory (especially if you are on Raspberry Pi 1, 2 or Zero, which have less than 1GB of RAM), this can often be rectified by increasing the swapfile size on the Pi by editing the file /etc/dphys-swapfile and changing the line CONF_SWAPSIZE=100 to CONF_SWAPSIZE=1024, then running:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> sudo /etc/init.d/dphys-swapfile stop
+ sudo /etc/init.d/dphys-swapfile start
+ free -m <span class="c1"># to verify the swapfile size has been increased</span>
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install MXNet Python Bindings</p>
+<p>To install python bindings run the following commands in the MXNet directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> <span class="nb">cd</span> python
+ pip install --upgrade pip
+ pip install -e .
+</pre></div>
+</div>
+<p>Note that the <code class="docutils literal"><span class="pre">-e</span></code> flag is optional. It is equivalent to <code class="docutils literal"><span class="pre">--editable</span></code> and means that if you edit the source files, these changes will be reflected in the package installed.</p>
+<p>You are now ready to run MXNet on your Raspberry Pi device. You can get started by following the tutorial on <a class="reference external" href="http://mxnet.io/tutorials/embedded/wine_detector.html">Real-time Object Detection with MXNet On The Raspberry Pi</a>.</p>
+<p><em>Note - Because the complete MXNet library takes up a significant amount of the Raspberry Pi’s limited RAM, when loading training data or large models into memory, you might have to turn off the GUI and terminate running processes to free RAM.</em></p>
+</div><div class="nvidia-jetson-tx2"><p>MXNet supports the Ubuntu Arch64 based operating system so you can run MXNet on NVIDIA Jetson Devices.</p>
+<p>These instructions will walk through how to build MXNet for the Pascal based <a class="reference external" href="http://www.nvidia.com/object/embedded-systems-dev-kits-modules.html">NVIDIA Jetson TX2</a> and install the corresponding python language bindings.</p>
+<p>For the purposes of this install guide we will assume that CUDA is already installed on your Jetson device.</p>
+<p><strong>Install MXNet</strong></p>
+<p>Installing MXNet is a two-step process:</p>
+<ol class="simple">
+<li>Build the shared library from the MXNet C++ source code.</li>
+<li>Install the supported language-specific packages for MXNet.</li>
+</ol>
+<p><strong>Step 1</strong> Build the Shared Library</p>
+<p>You need the following additional dependencies:</p>
+<ul class="simple">
+<li>Git (to pull code from GitHub)</li>
+<li>libatlas (for linear algebraic operations)</li>
+<li>libopencv (for computer vision operations)</li>
+<li>python pip (to load relevant python packages for our language bindings)</li>
+</ul>
+<p>Install these dependencies using the following commands in any directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> sudo apt-get update
+ sudo apt-get -y install git build-essential libatlas-base-dev libopencv-dev graphviz python-pip
+ sudo pip install pip --upgrade
+ sudo pip install setuptools numpy --upgrade
+ sudo pip install graphviz jupyter
+</pre></div>
+</div>
+<p>Clone the MXNet source code repository using the following <code class="docutils literal"><span class="pre">git</span></code> command in your home directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> git clone https://github.com/dmlc/mxnet.git --recursive
+ <span class="nb">cd</span> mxnet
+</pre></div>
+</div>
+<p>Edit the Makefile to install the MXNet with CUDA bindings to leverage the GPU on the Jetson:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> cp make/config.mk .
+ <span class="nb">echo</span> <span class="s2">"USE_CUDA=1"</span> >> config.mk
+ <span class="nb">echo</span> <span class="s2">"USE_CUDA_PATH=/usr/local/cuda"</span> >> config.mk
+ <span class="nb">echo</span> <span class="s2">"USE_CUDNN=1"</span> >> config.mk
+</pre></div>
+</div>
+<p>Edit the Mshadow Makefile to ensure MXNet builds with Pascal’s hardware level low precision acceleration by editing mshadow/make/mshadow.mk and adding the following after line 122:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="nv">MSHADOW_CFLAGS</span> <span class="o">+=</span> -DMSHADOW_USE_PASCAL<span class="o">=</span><span class="m">1</span>
+</pre></div>
+</div>
+<p>Now you can build the complete MXNet library with the following command:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> make -j <span class="k">$(</span>nproc<span class="k">)</span>
+</pre></div>
+</div>
+<p>Executing this command creates a file called <code class="docutils literal"><span class="pre">libmxnet.so</span></code> in the mxnet/lib directory.</p>
+<p><strong>Step 2</strong> Install MXNet Python Bindings</p>
+<p>To install python bindings run the following commands in the MXNet directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> <span class="nb">cd</span> python
+ pip install --upgrade pip
+ pip install -e .
+</pre></div>
+</div>
+<p>Note that the <code class="docutils literal"><span class="pre">-e</span></code> flag is optional. It is equivalent to <code class="docutils literal"><span class="pre">--editable</span></code> and means that if you edit the source files, these changes will be reflected in the package installed.</p>
+<p>Add the mxnet folder to the path:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> <span class="nb">cd</span> ..
+ <span class="nb">export</span> <span class="nv">MXNET_HOME</span><span class="o">=</span><span class="k">$(</span><span class="nb">pwd</span><span class="k">)</span>
+ <span class="nb">echo</span> <span class="s2">"export PYTHONPATH=</span><span class="nv">$MXNET_HOME</span><span class="s2">/python:</span><span class="nv">$PYTHONPATH</span><span class="s2">"</span> >> ~/.bashrc
+ <span class="nb">source</span> ~/.bashrc
+</pre></div>
+</div>
+<p>You are now ready to run MXNet on your NVIDIA Jetson TX2 device.</p>
+</div>
+</div><p><br/></p>
+</div>
+<div class="section" id="validate-mxnet-installation">
+<span id="validate-mxnet-installation"></span><h1>Validate MXNet Installation<a class="headerlink" href="#validate-mxnet-installation" title="Permalink to this headline">¶</a></h1>
+<div class="linux macos">
+ <div class="python">
+ <div class="cpu"><div class="pip build-from-source"><p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+</div><div class="docker"><p>Launch a Docker container with <code class="docutils literal"><span class="pre">mxnet/python</span></code> image and run example <em>MXNet</em> python program on the terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker run -it mxnet/python bash <span class="c1"># Use sudo if you skip Step 2 in the installation instruction</span>
+
+<span class="c1"># Start a python terminal</span>
+root@4919c4f58cac:/# python
+</pre></div>
+</div>
+</div><div class="virtualenv"><p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+</div><p>Run a short <em>MXNet</em> python program to create a 2X3 matrix of ones, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">mxnet</span> <span class="kn">as</span> <span class="nn">mx</span>
+<span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">))</span>
+<span class="gp">>>> </span><span class="n">b</span> <span class="o">=</span> <span class="n">a</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+<span class="gp">>>> </span><span class="n">b</span><span class="o">.</span><span class="n">asnumpy</span><span class="p">()</span>
+<span class="go">array([[ 3., 3., 3.],</span>
+<span class="go"> [ 3., 3., 3.]], dtype=float32)</span>
+</pre></div>
+</div>
+</div>
+</div>
+</div><!-- Validation for GPU machines --><div class="linux">
+ <div class="python">
+ <div class="gpu"><div class="pip build-from-source"><p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+</div><div class="docker"><p>Launch a NVIDIA Docker container with <code class="docutils literal"><span class="pre">mxnet/python:gpu</span></code> image and run example <em>MXNet</em> python program on the terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ nvidia-docker run -it mxnet/python:gpu bash <span class="c1"># Use sudo if you skip Step 2 in the installation instruction</span>
+
+<span class="c1"># Start a python terminal</span>
+root@4919c4f58cac:/# python
+</pre></div>
+</div>
+</div><div class="virtualenv"><p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+</div><p>Run a short <em>MXNet</em> python program to create a 2X3 matrix of ones <em>a</em> on a <em>GPU</em>, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3. We use <em>mx.gpu()</em>, to set <em>MXNet</em> context to be GPUs.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">mxnet</span> <span class="kn">as</span> <span class="nn">mx</span>
+<span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">mx</span><span class="o">.</span><span class="n">gpu</span><span class="p">())</span>
+<span class="gp">>>> </span><span class="n">b</span> <span class="o">=</span> <span class="n">a</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+<span class="gp">>>> </span><span class="n">b</span><span class="o">.</span><span class="n">asnumpy</span><span class="p">()</span>
+<span class="go">array([[ 3., 3., 3.],</span>
+<span class="go"> [ 3., 3., 3.]], dtype=float32)</span>
+</pre></div>
+</div>
+</div>
+</div>
+</div><div class="macos">
+ <div class="python">
+ <div class="gpu"><p>More details and verified validation instructions for macOS, with GPUs, coming soon.</p>
+</div>
+</div>
+</div><!-- Linux Clean up -->
+<div class="linux">
+ <div class="python">
+ <div class="cpu"><div class="pip build-from-source"><p>Exit the Python terminal.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><div class="virtualenv"><p>Exit the Python terminal and Deactivate the virtualenv <em>MXNet</em> environment.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">(mxnet)$ deactivate</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><div class="docker"><p>Exit the Python terminal and mxnet/python docker container.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">root@4919c4f58cac:/# exit</span>
+</pre></div>
+</div>
+</div></div>
+</div>
+</div><!-- MacOS Clean up -->
+<div class="macos">
+ <div class="python">
+ <div class="cpu"><div class="pip build-from-source"><p>Exit the Python terminal.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><div class="virtualenv"><p>Exit the Python terminal and Deactivate the virtualenv <em>MXNet</em> environment.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">(mxnet)$ deactivate</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><div class="docker"><p>Exit the Python terminal and then the docker container.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">root@4919c4f58cac:/# exit</span>
+</pre></div>
+</div>
+</div></div>
+</div>
+</div><!-- Validation for cloud installation --><div class="cloud"><p>Login to the cloud instance you launched, with pre-installed <em>MXNet</em>, following the guide by corresponding cloud provider.</p>
+<p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+<!-- Example Python code for CPU --><div class="cpu"><p>Run a short <em>MXNet</em> python program to create a 2X3 matrix of ones, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">mxnet</span> <span class="kn">as</span> <span class="nn">mx</span>
+<span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">))</span>
+<span class="gp">>>> </span><span class="n">b</span> <span class="o">=</span> <span class="n">a</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+<span class="gp">>>> </span><span class="n">b</span><span class="o">.</span><span class="n">asnumpy</span><span class="p">()</span>
+<span class="go">array([[ 3., 3., 3.],</span>
+<span class="go"> [ 3., 3., 3.]], dtype=float32)</span>
+</pre></div>
+</div>
+<p>Exit the Python terminal.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><!-- Example Python code for CPU --><div class="gpu"><p>Run a short <em>MXNet</em> python program to create a 2X3 matrix of ones <em>a</em> on a <em>GPU</em>, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3. We use <em>mx.gpu()</em>, to set <em>MXNet</em> context to be GPUs.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">mxnet</span> <span class="kn">as</span> <span class="nn">mx</span>
+<span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">mx</span><span class="o">.</span><span class="n">gpu</span><span class="p">())</span>
+<span class="gp">>>> </span><span class="n">b</span> <span class="o">=</span> <span class="n">a</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+<span class="gp">>>> </span><span class="n">b</span><span class="o">.</span><span class="n">asnumpy</span><span class="p">()</span>
+<span class="go">array([[ 3., 3., 3.],</span>
+<span class="go"> [ 3., 3., 3.]], dtype=float32)</span>
+</pre></div>
+</div>
+</div></div><!-- Example R code for CPU --><div class="linux macos windows">
+ <div class="r">
+ <div class="cpu"><p>Run a short <em>MXNet</em> R program to create a 2X3 matrix of ones, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3.</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span><span class="kn">library</span><span class="p">(</span>mxnet<span class="p">)</span>
+a <span class="o"><-</span> mx.nd.ones<span class="p">(</span><span class="kt">c</span><span class="p">(</span><span class="m">2</span><span class="p">,</span><span class="m">3</span><span class="p">),</span> ctx <span class="o">=</span> mx.cpu<span class="p">())</span>
+b <span class="o"><-</span> a <span class="o">*</span> <span class="m">2</span> <span class="o">+</span> <span class="m">1</span>
+b
+</pre></div>
+</div>
+</div>
+</div>
+</div><!-- Example R code for GPU --><div class="linux macos windows">
+ <div class="r">
+ <div class="gpu"><p>Run a short <em>MXNet</em> R program to create a 2X3 matrix of ones <em>a</em> on a <em>GPU</em>, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3. We use <em>mx.gpu()</em>, to set <em>MXNet</em> context to be GPUs.</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span><span class="kn">library</span><span class="p">(</span>mxnet<span class="p">)</span>
+a <span class="o"><-</span> mx.nd.ones<span class="p">(</span><span class="kt">c</span><span class="p">(</span><span class="m">2</span><span class="p">,</span><span class="m">3</span><span class="p">),</span> ctx <span class="o">=</span> mx.gpu<span class="p">())</span>
+b <span class="o"><-</span> a <span class="o">*</span> <span class="m">2</span> <span class="o">+</span> <span class="m">1</span>
+b
+</pre></div>
+</div>
+</div>
+</div>
+</div><div class="linux">
+ <div class="scala julia perl">
+ <div class="cpu gpu"><p>Will be available soon.</p>
+</div>
+</div>
+</div><div class="macos">
+ <div class="scala julia perl">
+ <div class="cpu gpu"><p>Will be available soon.</p>
+</div>
+</div>
+</div><div class="windows">
+ <div class="python">
+ <div class="cpu">
+ </div>
+ <div class="gpu"><p>The following installation instructions have been tested on Ubuntu 14.04 and 16.04.</p>
+<p><strong>Prerequisites</strong></p>
+<p>Install the following NVIDIA libraries to setup <em>MXNet</em> with GPU support:</p>
+<ol class="simple">
+<li>Install CUDA 8.0 following the NVIDIA’s <a class="reference external" href="http://docs.nvidia.com/cuda/cuda-installation-guide-microsoft-windows">installation guide</a>.</li>
+<li>Install cuDNN 7 for CUDA 8.0 following the NVIDIA’s <a class="reference external" href="https://developer.nvidia.com/cudnn">installation guide</a>. You may need to register with NVIDIA for downloading the cuDNN library.</li>
+</ol>
+<p><strong>Note:</strong> Make sure to add CUDA install path to <code class="docutils literal"><span class="pre">PATH</span></code>.</p>
+<div class="pip">
+<br/><p><strong>Step 1</strong> Install python.</p>
+<p>Recommend install <code class="docutils literal"><span class="pre">Anaconda3</span></code> <a class="reference external" href="https://www.anaconda.com/download/">here</a></p>
+<p><strong>Step 2</strong> Install <em>MXNet</em> with GPU support using CUDA 8.0</p>
+<p>To install MXNet 0.12.1 use the following: </p><div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet-cu80==0.12.1</pre></div> </div>
+<p>To install MXNet 0.12.0 use the following: </p><div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet-cu80==0.12.0</pre></div> </div>
+</div>
+<div class="build-from-source">
+<br/>
+To build and install MXNet yourself, you need the following dependencies. Install the required dependencies:<ol class="simple">
+<li>If <a class="reference external" href="https://www.visualstudio.com/downloads/">Microsoft Visual Studio 2015</a> is not already installed, download and install it. You can download and install the free community edition.</li>
+<li>Download and Install <a class="reference external" href="https://cmake.org/">CMake</a> if it is not already installed.</li>
+<li>Download and install <a class="reference external" href="http://sourceforge.net/projects/opencvlibrary/files/opencv-win/3.0.0/opencv-3.0.0.exe/download">OpenCV</a>.</li>
+<li>Unzip the OpenCV package.</li>
+<li>Set the environment variable <code class="docutils literal"><span class="pre">OpenCV_DIR</span></code> to point to the <code class="docutils literal"><span class="pre">OpenCV</span> <span class="pre">build</span> <span class="pre">directory</span></code>.</li>
+<li>If you don’t have the Intel Math Kernel Library (MKL) installed, download and install <a class="reference external" href="http://sourceforge.net/projects/openblas/files/v0.2.14/">OpenBlas</a>.</li>
+<li>Set the environment variable <code class="docutils literal"><span class="pre">OpenBLAS_HOME</span></code> to point to the <code class="docutils literal"><span class="pre">OpenBLAS</span></code> directory that contains the <code class="docutils literal"><span class="pre">include</span></code> and <code class="docutils literal"><span class="pre">lib</span></code> directories. Typically, you can find the directory in <code class="docutils literal"><span class="pre">C:\Program</span> <span class="pre">files</span> <span class="pre">(x86)\OpenBLAS\</span></code>.</li>
+<li>Download and install <a class="reference external" href="https://developer.nvidia.com/cudnn">CuDNN</a>. To get access to the download link, register as an NVIDIA community user.</li>
+</ol>
+<p>After you have installed all of the required dependencies, build the MXNet source code:</p>
+<ol class="simple">
+<li>Download the MXNet source code from <a class="reference external" href="https://github.com/dmlc/mxnet">GitHub</a>.</li>
+<li>Use <a class="reference external" href="https://cmake.org/">CMake</a> to create a Visual Studio solution in <code class="docutils literal"><span class="pre">./build</span></code>.</li>
+<li>In Visual Studio, open the solution file,<code class="docutils literal"><span class="pre">.sln</span></code>, and compile it.
+These commands produce a library called <code class="docutils literal"><span class="pre">mxnet.dll</span></code> in the <code class="docutils literal"><span class="pre">./build/Release/</span></code> or <code class="docutils literal"><span class="pre">./build/Debug</span></code> folder.</li>
+</ol>
+<p><p> </p>
+
+Next, we install <code class="docutils literal"><span class="pre">graphviz</span></code> library that we use for visualizing network graphs you build on MXNet. We will also install <a class="reference external" href="http://jupyter.readthedocs.io/">Jupyter Notebook</a> used for running MXNet tutorials and examples.</p>
+<ul class="simple">
+<li>Install <code class="docutils literal"><span class="pre">graphviz</span></code> by downloading MSI installer from <a class="reference external" href="http://www.graphviz.org/Download_windows.php">Graphviz Download Page</a>.
+<strong>Note</strong> Make sure to add graphviz executable path to PATH environment variable. Refer <a class="reference external" href="http://stackoverflow.com/questions/35064304/runtimeerror-make-sure-the-graphviz-executables-are-on-your-systems-path-aft">here for more details</a></li>
+</ul>
+<p><p> </p>
+</p>
+</div> </div>
+ </div>
+ <div class="scala julia perl">
+ <div class="cpu gpu"><p>Will be available soon.</p>
+ </div>
+ </div>
+</div><div class="devices">
+ <div class="raspberry-pi"><p>Will be available soon.</p>
+</div>
+<div class="nvidia-jetson-tx2"><p>Will be available soon.</p>
+</div>
+</div></div>
+<div class="section" id="download-source-package">
+<span id="download-source-package"></span><h1>Download Source Package<a class="headerlink" href="#download-source-package" title="Permalink to this headline">¶</a></h1>
+</div>
+
+
+ </div>
+ </div>
+
+ <div class="sphinxsidebar rightsidebar" role="navigation" aria-label="main navigation">
+ <div class="sphinxsidebarwrapper">
+ <h3><a href="../index.html">Table Of Contents</a></h3>
+ <ul>
+<li><a class="reference internal" href="#">Installing MXNet</a></li>
+<li><a class="reference internal" href="#validate-mxnet-installation">Validate MXNet Installation</a></li>
+<li><a class="reference internal" href="#download-source-package">Download Source Package</a></li>
+</ul>
+
+ </div>
+ </div>
+ </div><div class="footer">
+<div class="section-disclaimer">
+<div class="container">
+ <div>
+ <img src="https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/apache_incubator_logo.png" height=60>
+ <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 src="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.6/js/bootstrap.min.js" integrity="sha384-0mSbJDEHialfmuBBQP6A4Qrprq5OVfW37PRR3j5ELqxss1yVqOtnepnHVP9aJ7xS" crossorigin="anonymous"></script>
+ <script type="text/javascript" src="../_static/js/sidebar.js"></script>
+ <script type="text/javascript" src="../_static/js/search.js"></script>
+ <script type="text/javascript" src="../_static/js/navbar.js"></script>
+ <script type="text/javascript" src="../_static/js/clipboard.min.js"></script>
+ <script type="text/javascript" src="../_static/js/copycode.js"></script>
+ <script type="text/javascript" src="../_static/js/page.js"></script>
+ <script type="text/javascript">
+ $('body').ready(function () {
+ $('body').css('visibility', 'visible');
+ });
+ </script>
+ </body>
+</html>
diff --git a/docs/build_version_doc/artifacts/1.0.0.index.html b/docs/build_version_doc/artifacts/1.0.0.index.html
new file mode 100644
index 00000000000..b7c7a514e4e
--- /dev/null
+++ b/docs/build_version_doc/artifacts/1.0.0.index.html
@@ -0,0 +1,1302 @@
+<!--
+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.
+--><!DOCTYPE html>
+
+<html lang="en">
+ <head>
+ <meta charset="utf-8">
+ <meta http-equiv="X-UA-Compatible" content="IE=edge">
+ <meta name="viewport" content="width=device-width, initial-scale=1">
+
+
+
+ <title>Installing MXNet — mxnet documentation</title>
+
+ <link rel="stylesheet" href="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.6/css/bootstrap.min.css" integrity="sha384-1q8mTJOASx8j1Au+a5WDVnPi2lkFfwwEAa8hDDdjZlpLegxhjVME1fgjWPGmkzs7" crossorigin="anonymous">
+ <link rel="stylesheet" href="https://maxcdn.bootstrapcdn.com/font-awesome/4.5.0/css/font-awesome.min.css">
+
+
+ <link rel="stylesheet" href="../_static/basic.css" type="text/css" />
+ <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+
+ <link rel="stylesheet" href="../_static/mxnet.css" type="text/css">
+
+ <script type="text/javascript">
+ var DOCUMENTATION_OPTIONS = {
+ URL_ROOT: '../',
+ VERSION: '',
+ COLLAPSE_INDEX: false,
+ FILE_SUFFIX: '.html',
+ HAS_SOURCE: true,
+ SOURCELINK_SUFFIX: '.txt'
+ };
+ </script>
+
+ <script type="text/javascript" src="../_static/jquery-1.11.1.js"></script>
+
+ <script type="text/javascript" src="../_static/underscore.js"></script>
+
+ <script type="text/javascript" src="../_static/searchtools_custom.js"></script>
+
+ <script type="text/javascript" src="../_static/doctools.js"></script>
+
+ <script type="text/javascript" src="../_static/selectlang.js"></script>
+
+
+ <script type="text/javascript" src="https://cdn.mathjax.org/mathjax/latest/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></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://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.0/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script> -->
+ <!-- -->
+ <link rel="index" title="Index" href="../genindex.html" />
+ <link rel="search" title="Search" href="../search.html" />
+
+ <link rel="icon" type="image/png" href="https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/mxnet-icon.png">
+ </head>
+ <body role="document" background="https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/mxnet-background.png">
+ <div class='content-block'><!--
+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="navbar navbar-fixed-top">
+ <div class="container" id="navContainer">
+ <div id="header-inner" class="innder">
+ <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 id="main-nav" class='nav-bar'>
+ <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 href="#" class="main-nav-link dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">Gluon <span class="caret"></span></a>
+ <ul id="package-dropdown-menu" class="dropdown-menu navbar-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 href="#" class="main-nav-link dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">API <span class="caret"></span></a>
+ <ul id="package-dropdown-menu" class="dropdown-menu navbar-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 href="#" class="main-nav-link dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">Docs <span class="caret"></span></a>
+ <ul id="package-dropdown-menu-docs" class="dropdown-menu navbar-menu">
+ <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/master/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 href="#" class="main-nav-link dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">Community <span class="caret"></span></a>
+ <ul id="package-dropdown-menu-community" class="dropdown-menu navbar-menu">
+ <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>
+ </nav>
+ <script> function getRootPath(){ return "../" } </script>
+ <div class="burgerIcon dropdown">
+ <a href="#" class="dropdown-toggle" data-toggle="dropdown" role="button">☰</a>
+ <ul id="burgerMenu" class="dropdown-menu">
+ <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 tabindex="-1" href="../community/index.html">Community</a></li>
+ <li><a tabindex="-1" href="../community/contribute.html">Contribute</a></li>
+ <li><a tabindex="-1" href="../community/powered_by.html">Powered By</a></li>
+ </ul>
+ </li>
+
+ <li class="dropdown-submenu">
+ <a href="#" tabindex="-1">API</a>
+ <ul class="dropdown-menu">
+
+ <li><a tabindex="-1" href="../api/python/index.html">Python</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/scala/index.html">Scala</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/r/index.html">R</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/julia/index.html">Julia</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/c++/index.html">C++</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/perl/index.html">Perl</a>
+ </li>
+
+ </ul>
+ </li>
+
+ <li class="dropdown-submenu">
+ <a href="#" tabindex="-1">Docs</a>
+ <ul class="dropdown-menu">
+ <li><a tabindex="-1" href="../tutorials/index.html">Tutorials</a></li>
+ <li><a tabindex="-1" href="../faq/index.html">FAQ</a></li>
+ <li><a tabindex="-1" href="../architecture/index.html">Architecture</a></li>
+ <li><a tabindex="-1" href="https://github.com/apache/incubator-mxnet/tree/master/example">Examples</a></li>
+ <li><a tabindex="-1" href="../model_zoo/index.html">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>
+ </ul>
+ </div>
+ <div class="plusIcon dropdown">
+ <a href="#" class="dropdown-toggle" data-toggle="dropdown" role="button"><span class="glyphicon glyphicon-plus" aria-hidden="true"></span></a>
+ <ul id="plusMenu" class="dropdown-menu dropdown-menu-right"></ul>
+ </div>
+ <div id="search-input-wrap">
+
+<form class="" role="search" action="../search.html" method="get" autocomplete="off">
+ <div class="form-group inner-addon left-addon">
+ <i class="glyphicon glyphicon-search"></i>
+ <input type="text" name="q" class="form-control" placeholder="Search">
+ </div>
+ <input type="hidden" name="check_keywords" value="yes" />
+ <input type="hidden" name="area" value="default" />
+
+</form>
+ <div id="search-preview"></div>
+ </div>
+ <div id='searchIcon'>
+ <span class="glyphicon glyphicon-search" aria-hidden="true"></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 class="sphinxsidebar leftsidebar" role="navigation" aria-label="main 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>
+
+ <div class="section" id="installing-mxnet">
+<span id="installing-mxnet"></span><h1>Installing MXNet<a class="headerlink" href="#installing-mxnet" title="Permalink to this headline">¶</a></h1>
+<p>Indicate your preferred configuration. Then, follow the customized commands to install <em>MXNet</em>.</p>
+<script type="text/javascript" src='../_static/js/options.js'></script><!-- START - OS Menu --><div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">Linux</button>
+ <button type="button" class="btn btn-default opt">MacOS</button>
+ <button type="button" class="btn btn-default opt">Windows</button>
+ <button type="button" class="btn btn-default opt">Cloud</button>
+ <button type="button" class="btn btn-default opt">Devices</button>
+</div><!-- START - Language Menu --><div class="linux macos windows">
+<div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">Python</button>
+ <button type="button" class="btn btn-default opt">Scala</button>
+ <button type="button" class="btn btn-default opt">R</button>
+ <button type="button" class="btn btn-default opt">Julia</button>
+ <button type="button" class="btn btn-default opt">Perl</button>
+</div>
+</div><!-- No CPU GPU for other Devices -->
+<div class="linux macos windows cloud">
+<div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">CPU</button>
+ <button type="button" class="btn btn-default opt">GPU</button>
+</div>
+</div><!-- other devices -->
+<div class="devices">
+<div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">Raspberry Pi</button>
+ <button type="button" class="btn btn-default opt">NVIDIA Jetson TX2</button>
+</div>
+</div><!-- Linux Python GPU Options --><div class="linux macos">
+<div class="python">
+<div class="cpu gpu">
+<div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">Pip</button>
+ <button type="button" class="btn btn-default opt">Virtualenv</button>
+ <button type="button" class="btn btn-default opt">Docker</button>
+ <button type="button" class="btn btn-default opt">Build from Source</button>
+</div>
+</div>
+</div>
+</div><!-- END - Main Menu --><!-- START - Linux Python CPU Installation Instructions --><div class="linux">
+ <div class="python">
+ <div class="cpu"><p>The following installation instructions have been tested on Ubuntu 14.04 and 16.04.</p>
+<div class="virtualenv">
+<br/><p><strong>Step 1</strong> Install virtualenv for Ubuntu.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y python-dev python-virtualenv
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Create and activate virtualenv environment for MXNet.</p>
+<p>Following command creates a virtualenv environment at <code class="docutils literal"><span class="pre">~/mxnet</span></code> directory. However, you can choose any directory by replacing <code class="docutils literal"><span class="pre">~/mxnet</span></code> with a directory of your choice.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ virtualenv --system-site-packages ~/mxnet
+</pre></div>
+</div>
+<p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install MXNet in the active virtualenv environment.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install --upgrade pip
+</pre></div>
+</div>
+<p>Install <em>MXNet</em> with OpenBLAS acceleration.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet==1.0.0
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 5</strong> Validate the installation by running simple <em>MXNet</em> code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Note</strong> You can read more about virtualenv <a class="reference external" href="https://virtualenv.pypa.io/en/stable/userguide/">here</a>.</p>
+</div><div class="pip">
+<br/><p><strong>Step 1</strong> Install prerequisites - wget and latest pip.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command in the terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y wget python gcc
+$ wget https://bootstrap.pypa.io/get-pip.py <span class="o">&&</span> sudo python get-pip.py
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install MXNet with OpenBLAS acceleration.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet==1.0.0
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Experimental Choice</strong> If You would like to install mxnet with Intel MKL, try the experimental pip package with MKL:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet-mkl==1.0.0
+</pre></div>
+</div>
+</div><div class="docker">
+<br/><p>Docker images with <em>MXNet</em> are available at <a class="reference external" href="https://hub.docker.com/r/mxnet/">Docker Hub</a>.</p>
+<p><strong>Step 1</strong> Install Docker on your machine by following the <a class="reference external" href="https://docs.docker.com/engine/installation/linux/ubuntu/#install-using-the-repository">docker installation instructions</a>.</p>
+<p><em>Note</em> - You can install Community Edition (CE) to get started with <em>MXNet</em>.</p>
+<p><strong>Step 2</strong> [Optional] Post installation steps to manage Docker as a non-root user.</p>
+<p>Follow the four steps in this <a class="reference external" href="https://docs.docker.com/engine/installation/linux/linux-postinstall/#manage-docker-as-a-non-root-user">docker documentation</a> to allow managing docker containers without <em>sudo</em>.</p>
+<p>If you skip this step, you need to use <em>sudo</em> each time you invoke Docker.</p>
+<p><strong>Step 2</strong> Pull the MXNet docker image.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker pull mxnet/python <span class="c1"># Use sudo if you skip Step 2</span>
+</pre></div>
+</div>
+<p>You can list docker images to see if mxnet/python docker image pull was successful.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker images <span class="c1"># Use sudo if you skip Step 2</span>
+
+REPOSITORY TAG IMAGE ID CREATED SIZE
+mxnet/python latest 00d026968b3c <span class="m">3</span> weeks ago <span class="m">1</span>.41 GB
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div><div class="build-from-source">
+<br/><p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings. Example - Python bindings, Scala bindings.</li>
+</ol>
+<p><strong>Minimum Requirements</strong></p>
+<ol class="simple">
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">GCC 4.8</a> or later to compile C++ 11.</li>
+<li><a class="reference external" href="https://www.gnu.org/software/make/">GNU Make</a></li>
+</ol>
+<p><br/></p>
+<p><strong>Build the MXNet core shared library</strong></p>
+<p><strong>Step 1</strong> Install build tools and git.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y build-essential git
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install OpenBLAS.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="https://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprograms">BLAS</a> and <a class="reference external" href="https://en.wikipedia.org/wiki/LAPACK">LAPACK</a> libraries for accelerated numerical computations on CPU machine. There are several flavors of BLAS/LAPACK libraries - <a class="reference external" href="http://www.openblas.net/">OpenBLAS</a>, <a class="reference external" href="http://math-atlas.sourceforge.net/">ATLAS</a> and <a class="reference external" href="https://software.intel.com/en-us/intel-mkl">MKL</a>. In this step we install OpenBLAS. You can choose to install ATLAS or MKL.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopenblas-dev liblapack-dev
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install OpenCV.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="http://opencv.org/">OpenCV</a> for efficient image loading and augmentation operations.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopencv-dev
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Download MXNet sources and build MXNet core shared library. You can clone the repository as described in the following code block, or you may try the <a href="download.html">download links</a> for your desired MXNet version.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ git clone --recursive https://github.com/dmlc/mxnet
+$ <span class="nb">cd</span> mxnet
+$ make -j <span class="k">$(</span>nproc<span class="k">)</span> <span class="nv">USE_OPENCV</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_BLAS</span><span class="o">=</span>openblas
+</pre></div>
+</div>
+<p><em>Note</em> - USE_OPENCV and USE_BLAS are make file flags to set compilation options to use OpenCV and BLAS library. You can explore and use more compilation options in <code class="docutils literal"><span class="pre">make/config.mk</span></code>.</p>
+<p><br/></p>
+<p><strong>Build the MXNet Python binding</strong></p>
+<p><strong>Step 1</strong> Install prerequisites - python, setup-tools, python-pip and numpy.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y python-dev python-setuptools python-numpy python-pip
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install the MXNet Python binding.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">cd</span> python
+$ pip install --upgrade pip
+$ pip install -e .
+</pre></div>
+</div>
+<p>Note that the <code class="docutils literal"><span class="pre">-e</span></code> flag is optional. It is equivalent to <code class="docutils literal"><span class="pre">--editable</span></code> and means that if you edit the source files, these changes will be reflected in the package installed.</p>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div></div>
+</div>
+</div>
+<!-- END - Linux Python CPU Installation Instructions --><!-- START - Linux Python GPU Installation Instructions --><div class="linux">
+<div class="python">
+<div class="gpu"><p>The following installation instructions have been tested on Ubuntu 14.04 and 16.04.</p>
+<p><strong>Prerequisites</strong></p>
+<p>Install the following NVIDIA libraries to setup <em>MXNet</em> with GPU support:</p>
+<ol class="simple">
+<li>Install CUDA 8.0 following the NVIDIA’s <a class="reference external" href="http://docs.nvidia.com/cuda/cuda-installation-guide-linux/">installation guide</a>.</li>
+<li>Install cuDNN 5 for CUDA 8.0 following the NVIDIA’s <a class="reference external" href="https://developer.nvidia.com/cudnn">installation guide</a>. You may need to register with NVIDIA for downloading the cuDNN library.</li>
+</ol>
+<p><strong>Note:</strong> Make sure to add CUDA install path to <code class="docutils literal"><span class="pre">LD_LIBRARY_PATH</span></code>.</p>
+<p>Example - <em>export LD_LIBRARY_PATH=/usr/local/cuda/lib64/:$LD_LIBRARY_PATH</em></p>
+<div class="pip">
+<br/><p><strong>Step 1</strong> Install prerequisites - wget and latest pip.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command in the terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y wget python
+$ wget https://bootstrap.pypa.io/get-pip.py <span class="o">&&</span> sudo python get-pip.py
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install <em>MXNet</em> with GPU support using CUDA 8.0</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet-cu80==1.0.0
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Experimental Choice</strong> If You would like to install mxnet with Intel MKL, try the experimental pip package with MKL:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet-cu80mkl==1.0.0
+</pre></div>
+</div>
+</div><div class="virtualenv"><p><br/></p>
+<p><strong>Step 1</strong> Install virtualenv for Ubuntu.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y python-dev python-virtualenv
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Create and activate virtualenv environment for MXNet.</p>
+<p>Following command creates a virtualenv environment at <code class="docutils literal"><span class="pre">~/mxnet</span></code> directory. However, you can choose any directory by replacing <code class="docutils literal"><span class="pre">~/mxnet</span></code> with a directory of your choice.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ virtualenv --system-site-packages ~/mxnet
+</pre></div>
+</div>
+<p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install MXNet in the active virtualenv environment.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install --upgrade pip
+</pre></div>
+</div>
+<p>Install <em>MXNet</em> with GPU support using CUDA 8.0.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet-cu80==1.0.0
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 5</strong> Validate the installation by running simple <em>MXNet</em> code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Note</strong> You can read more about virtualenv <a class="reference external" href="https://virtualenv.pypa.io/en/stable/userguide/">here</a>.</p>
+</div><div class="docker"><p><br/></p>
+<p>Docker images with <em>MXNet</em> are available at <a class="reference external" href="https://hub.docker.com/r/mxnet/">Docker Hub</a>.</p>
+<p><strong>Step 1</strong> Install Docker on your machine by following the <a class="reference external" href="https://docs.docker.com/engine/installation/linux/ubuntu/#install-using-the-repository">docker installation instructions</a>.</p>
+<p><em>Note</em> - You can install Community Edition (CE) to get started with <em>MXNet</em>.</p>
+<p><strong>Step 2</strong> [Optional] Post installation steps to manage Docker as a non-root user.</p>
+<p>Follow the four steps in this <a class="reference external" href="https://docs.docker.com/engine/installation/linux/linux-postinstall/#manage-docker-as-a-non-root-user">docker documentation</a> to allow managing docker containers without <em>sudo</em>.</p>
+<p>If you skip this step, you need to use <em>sudo</em> each time you invoke Docker.</p>
+<p><strong>Step 3</strong> Install <em>nvidia-docker-plugin</em> following the <a class="reference external" href="https://github.com/NVIDIA/nvidia-docker/wiki/Installation">installation instructions</a>. <em>nvidia-docker-plugin</em> is required to enable the usage of GPUs from the docker containers.</p>
+<p><strong>Step 4</strong> Pull the MXNet docker image.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker pull mxnet/python:gpu <span class="c1"># Use sudo if you skip Step 2</span>
+</pre></div>
+</div>
+<p>You can list docker images to see if mxnet/python docker image pull was successful.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker images <span class="c1"># Use sudo if you skip Step 2</span>
+
+REPOSITORY TAG IMAGE ID CREATED SIZE
+mxnet/python gpu 493b2683c269 <span class="m">3</span> weeks ago <span class="m">4</span>.77 GB
+</pre></div>
+</div>
+<p><strong>Step 5</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div><div class="build-from-source"><p><br/></p>
+<p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings. Example - Python bindings, Scala bindings.</li>
+</ol>
+<p><strong>Minimum Requirements</strong></p>
+<ol class="simple">
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">GCC 4.8</a> or later to compile C++ 11.</li>
+<li><a class="reference external" href="https://www.gnu.org/software/make/">GNU Make</a></li>
+</ol>
+<p><br/></p>
+<p><strong>Build the MXNet core shared library</strong></p>
+<p><strong>Step 1</strong> Install build tools and git.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y build-essential git
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install OpenBLAS.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="https://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprograms">BLAS</a> and <a class="reference external" href="https://en.wikipedia.org/wiki/LAPACK">LAPACK</a> libraries for accelerated numerical computations on CPU machine. There are several flavors of BLAS/LAPACK libraries - <a class="reference external" href="http://www.openblas.net/">OpenBLAS</a>, <a class="reference external" href="http://math-atlas.sourceforge.net/">ATLAS</a> and <a class="reference external" href="https://software.intel.com/en-us/intel-mkl">MKL</a>. In this step we install OpenBLAS. You can choose to install ATLAS or MKL.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopenblas-dev liblapack-dev
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install OpenCV.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="http://opencv.org/">OpenCV</a> for efficient image loading and augmentation operations.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopencv-dev
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Download MXNet sources and build MXNet core shared library. You can clone the repository as described in the following code block, or you may try the <a href="download.html">download links</a> for your desired MXNet version.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ git clone --recursive https://github.com/dmlc/mxnet
+$ <span class="nb">cd</span> mxnet
+$ make -j <span class="k">$(</span>nproc<span class="k">)</span> <span class="nv">USE_OPENCV</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_BLAS</span><span class="o">=</span>openblas <span class="nv">USE_CUDA</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_CUDA_PATH</span><span class="o">=</span>/usr/local/cuda <span class="nv">USE_CUDNN</span><span class="o">=</span><span class="m">1</span>
+</pre></div>
+</div>
+<p><em>Note</em> - USE_OPENCV, USE_BLAS, USE_CUDA, USE_CUDA_PATH AND USE_CUDNN are make file flags to set compilation options to use OpenCV, OpenBLAS, CUDA and cuDNN libraries. You can explore and use more compilation options in <code class="docutils literal"><span class="pre">make/config.mk</span></code>. Make sure to set USE_CUDA_PATH to right CUDA installation path. In most cases it is - <em>/usr/local/cuda</em>.</p>
+<p><br/></p>
+<p><strong>Install the MXNet Python binding</strong></p>
+<p><strong>Step 1</strong> Install prerequisites - python, setup-tools, python-pip and numpy.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y python-dev python-setuptools python-numpy python-pip
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install the MXNet Python binding.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">cd</span> python
+$ pip install --upgrade pip
+$ pip install -e .
+</pre></div>
+</div>
+<p>Note that the <code class="docutils literal"><span class="pre">-e</span></code> flag is optional. It is equivalent to <code class="docutils literal"><span class="pre">--editable</span></code> and means that if you edit the source files, these changes will be reflected in the package installed.</p>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div></div>
+</div>
+</div>
+<!-- END - Linux Python GPU Installation Instructions --><!-- START - MacOS Python CPU Installation Instructions --><div class="macos">
+ <div class="python">
+ <div class="cpu"><p>The following installation instructions have been tested on OSX Sierra and El Capitan.</p>
+<div class="virtualenv">
+<br/><p><strong>Step 1</strong> Install prerequisites - Homebrew, python development tools.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="c1"># Install Homebrew</span>
+$ /usr/bin/ruby -e <span class="s2">"</span><span class="k">$(</span>curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install<span class="k">)</span><span class="s2">"</span>
+$ <span class="nb">export</span> <span class="nv">PATH</span><span class="o">=</span>/usr/local/bin:/usr/local/sbin:<span class="nv">$PATH</span>
+
+<span class="c1"># Install python development tools - python2.7, pip, python-setuptools</span>
+$ brew install python
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install virtualenv for macOS.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install virtualenv
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Create and activate virtualenv environment for MXNet.</p>
+<p>Following command creates a virtualenv environment at <code class="docutils literal"><span class="pre">~/mxnet</span></code> directory. However, you can choose any directory by replacing <code class="docutils literal"><span class="pre">~/mxnet</span></code> with a directory of your choice.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ virtualenv --system-site-packages ~/mxnet
+</pre></div>
+</div>
+<p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Install MXNet in the active virtualenv environment.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install --upgrade pip
+<span class="o">(</span>mxnet<span class="o">)</span>$ pip install --upgrade setuptools
+</pre></div>
+</div>
+<p>Install <em>MXNet</em> with OpenBLAS acceleration.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet==1.0.0
+</pre></div>
+</div>
+<p><strong>Step 5</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ brew install graphviz
+<span class="o">(</span>mxnet<span class="o">)</span>$ pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 6</strong> Validate the installation by running simple <em>MXNet</em> code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Note</strong> You can read more about virtualenv <a class="reference external" href="https://virtualenv.pypa.io/en/stable/userguide/">here</a>.</p>
+</div><div class="pip">
+<br/><p><strong>Step 1</strong> Install prerequisites - Homebrew, python development tools.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="c1"># Install Homebrew</span>
+$ /usr/bin/ruby -e <span class="s2">"</span><span class="k">$(</span>curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install<span class="k">)</span><span class="s2">"</span>
+$ <span class="nb">export</span> <span class="nv">PATH</span><span class="o">=</span>/usr/local/bin:/usr/local/sbin:<span class="nv">$PATH</span>
+
+<span class="c1"># Install python development tools - python2.7, pip, python-setuptools</span>
+$ brew install python
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install MXNet with OpenBLAS acceleration.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install --upgrade pip
+$ pip install --upgrade setuptools
+</pre></div>
+</div>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet==1.0.0
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ brew install graphviz
+$ pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div><div class="docker">
+<br/><p>Docker images with <em>MXNet</em> are available at <a class="reference external" href="https://hub.docker.com/r/mxnet/">Docker Hub</a>.</p>
+<p><strong>Step 1</strong> Install Docker on your machine by following the <a class="reference external" href="https://docs.docker.com/docker-for-mac/install/#install-and-run-docker-for-mac">docker installation instructions</a>.</p>
+<p><em>Note</em> - You can install Community Edition (CE) to get started with <em>MXNet</em>.</p>
+<p><strong>Step 2</strong> Pull the MXNet docker image.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker pull mxnet/python
+</pre></div>
+</div>
+<p>You can list docker images to see if mxnet/python docker image pull was successful.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker images
+
+REPOSITORY TAG IMAGE ID CREATED SIZE
+mxnet/python latest 00d026968b3c <span class="m">3</span> weeks ago <span class="m">1</span>.41 GB
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div><div class="build-from-source">
+<br/><p><strong>Prerequisites</strong></p>
+<p>If not already installed, <a class="reference external" href="https://developer.apple.com/xcode/">download and install Xcode</a> (or <a class="reference external" href="https://itunes.apple.com/us/app/xcode/id497799835">insall it from the App Store</a>) for macOS. <a class="reference external" href="https://en.wikipedia.org/wiki/Xcode">Xcode</a> is an integrated development environment for macOS containing a suite of software development tools like C/C++ compilers, BLAS library and more.</p>
+<p><br/></p>
+<p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings. Example - Python bindings, Scala bindings.</li>
+</ol>
+<p>Make sure you have installed Xcode before proceeding further.</p>
+<p><br/></p>
+<p>All the instructions to build <em>MXNet</em> core shared library and <em>MXNet</em> Python bindings are compiled as one helper <em>bash</em> script. You can use <a class="reference external" href="https://raw.githubusercontent.com/dmlc/mxnet/master/setup-utils/install-mxnet-osx-python.sh">this bash script</a> to build <em>MXNet</em> for Python, from source, on macOS.</p>
+<p><strong>Step 1</strong> Download the bash script for building MXNet from source.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ curl -O https://raw.githubusercontent.com/dmlc/mxnet/master/setup-utils/install-mxnet-osx-python.sh
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Run the script to get latest MXNet source and build.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="c1"># Make the script executable</span>
+$ chmod <span class="m">744</span> install-mxnet-osx-python.sh
+
+<span class="c1"># Run the script. It takes around 5 mins.</span>
+$ bash install-mxnet-osx-python.sh
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div> <!-- End of source build --></div>
+</div>
+</div><!-- END - Mac OS Python CPU Installation Instructions --><!-- START - Mac OS Python GPU Installation Instructions --><div class="macos">
+ <div class="python">
+ <div class="gpu"><p>More details and verified installation instructions for macOS, with GPUs, coming soon.</p>
+<p><em>MXNet</em> is expected to be compatible on macOS with NVIDIA GPUs. Please install CUDA 8.0 and cuDNN 5.0, prior to installing GPU version of <em>MXNet</em>.</p>
+</div>
+</div>
+</div><!-- END - Mac OS Python GPU Installation Instructions --><!-- START - Cloud Python Installation Instructions --><div class="cloud"><p>AWS Marketplace distributes AMIs (Amazon Machine Image) with MXNet pre-installed. You can launch an Amazon EC2 instance with one of the below AMIs:</p>
+<ol class="simple">
+<li>Deep Learning AMI (Amazon Machine Image) for <a class="reference external" href="https://aws.amazon.com/marketplace/pp/B06VSPXKDX">Ubuntu</a></li>
+<li>Deep Learning AMI for <a class="reference external" href="https://aws.amazon.com/marketplace/pp/B01M0AXXQB">Amazon Linux</a></li>
+</ol>
+<p>You could also run distributed deeplearning with <em>MXNet</em> on AWS using <a class="reference external" href="https://github.com/awslabs/deeplearning-cfn/blob/master/README.md">Cloudformation Template</a>.</p>
+</div><!-- END - Cloud Python Installation Instructions --><!-- START - MacOS R CPU Installation Instructions --><div class="macos">
+ <div class="r">
+ <div class="cpu"><p>The CPU version of MXNet R package can be installed in R like other packages</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span>cran <span class="o"><-</span> <span class="kp">getOption</span><span class="p">(</span><span class="s">"repos"</span><span class="p">)</span>
+cran<span class="p">[</span><span class="s">"dmlc"</span><span class="p">]</span> <span class="o"><-</span> <span class="s">"https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/R/CRAN/"</span>
+<span class="kp">options</span><span class="p">(</span>repos <span class="o">=</span> cran<span class="p">)</span>
+install.packages<span class="p">(</span><span class="s">"mxnet"</span><span class="p">)</span>
+</pre></div>
+</div>
+</div><div class="gpu"><p>Will be available soon.</p>
+</div></div>
+</div>
+<!-- END - MacOS R CPU Installation Instructions --><div class="linux">
+ <div class="r">
+ <div class="cpu">
+<br/><p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings.</li>
+</ol>
+<p><strong>Minimum Requirements</strong></p>
+<ol class="simple">
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">GCC 4.8</a> or later to compile C++ 11.</li>
+<li><a class="reference external" href="https://www.gnu.org/software/make/">GNU Make</a></li>
+</ol>
+<p><br/></p>
+<p><strong>Build the MXNet core shared library</strong></p>
+<p><strong>Step 1</strong> Install build tools and git.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y build-essential git
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install OpenBLAS.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="https://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprograms">BLAS</a> and <a class="reference external" href="https://en.wikipedia.org/wiki/LAPACK">LAPACK</a> libraries for accelerated numerical computations on CPU machine. There are several flavors of BLAS/LAPACK libraries - <a class="reference external" href="http://www.openblas.net/">OpenBLAS</a>, <a class="reference external" href="http://math-atlas.sourceforge.net/">ATLAS</a> and <a class="reference external" href="https://software.intel.com/en-us/intel-mkl">MKL</a>. In this step we install OpenBLAS. You can choose to install ATLAS or MKL.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopenblas-dev liblapack-dev
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install OpenCV.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="http://opencv.org/">OpenCV</a> for efficient image loading and augmentation operations.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopencv-dev
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Download MXNet sources and build MXNet core shared library. You can clone the repository as described in the following code block, or you may try the <a href="download.html">download links</a> for your desired MXNet version.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ git clone --recursive https://github.com/dmlc/mxnet
+$ <span class="nb">cd</span> mxnet
+$ make -j <span class="k">$(</span>nproc<span class="k">)</span> <span class="nv">USE_OPENCV</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_BLAS</span><span class="o">=</span>openblas
+</pre></div>
+</div>
+<p><em>Note</em> - USE_OPENCV and USE_BLAS are make file flags to set compilation options to use OpenCV and BLAS library. You can explore and use more compilation options in <code class="docutils literal"><span class="pre">make/config.mk</span></code>.</p>
+<p><br/></p>
+<p><strong>Build and install the MXNet R binding</strong></p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ make rpkg
+$ R CMD INSTALL mxnet_current_r.tar.gz
+</pre></div>
+</div>
+</div><div class="gpu"><p>The following installation instructions have been tested on Ubuntu 14.04 and 16.04.</p>
+<p><strong>Prerequisites</strong></p>
+<p>Install the following NVIDIA libraries to setup <em>MXNet</em> with GPU support:</p>
+<ol class="simple">
+<li>Install CUDA 8.0 following the NVIDIA’s <a class="reference external" href="http://docs.nvidia.com/cuda/cuda-installation-guide-linux/">installation guide</a>.</li>
+<li>Install cuDNN 5 for CUDA 8.0 following the NVIDIA’s <a class="reference external" href="https://developer.nvidia.com/cudnn">installation guide</a>. You may need to register with NVIDIA for downloading the cuDNN library.</li>
+</ol>
+<p><strong>Note:</strong> Make sure to add CUDA install path to <code class="docutils literal"><span class="pre">LD_LIBRARY_PATH</span></code>.</p>
+<p>Example - <em>export LD_LIBRARY_PATH=/usr/local/cuda/lib64/:$LD_LIBRARY_PATH</em></p>
+<p><br/></p>
+<p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings.</li>
+</ol>
+<p><strong>Minimum Requirements</strong></p>
+<ol class="simple">
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">GCC 4.8</a> or later to compile C++ 11.</li>
+<li><a class="reference external" href="https://www.gnu.org/software/make/">GNU Make</a></li>
+</ol>
+<p><br/></p>
+<p><strong>Build the MXNet core shared library</strong></p>
+<p><strong>Step 1</strong> Install build tools and git.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y build-essential git
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install OpenBLAS.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="https://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprograms">BLAS</a> and <a class="reference external" href="https://en.wikipedia.org/wiki/LAPACK">LAPACK</a> libraries for accelerated numerical computations on CPU machine. There are several flavors of BLAS/LAPACK libraries - <a class="reference external" href="http://www.openblas.net/">OpenBLAS</a>, <a class="reference external" href="http://math-atlas.sourceforge.net/">ATLAS</a> and <a class="reference external" href="https://software.intel.com/en-us/intel-mkl">MKL</a>. In this step we install OpenBLAS. You can choose to install ATLAS or MKL.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopenblas-dev liblapack-dev
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install OpenCV.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="http://opencv.org/">OpenCV</a> for efficient image loading and augmentation operations.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopencv-dev
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Download MXNet sources and build MXNet core shared library. You can clone the repository as described in the following code block, or you may try the <a href="download.html">download links</a> for your desired MXNet version.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ git clone --recursive https://github.com/dmlc/mxnet
+$ <span class="nb">cd</span> mxnet
+$ make -j <span class="k">$(</span>nproc<span class="k">)</span> <span class="nv">USE_OPENCV</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_BLAS</span><span class="o">=</span>openblas <span class="nv">USE_CUDA</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_CUDA_PATH</span><span class="o">=</span>/usr/local/cuda <span class="nv">USE_CUDNN</span><span class="o">=</span><span class="m">1</span>
+</pre></div>
+</div>
+<p><em>Note</em> - USE_OPENCV, USE_BLAS, USE_CUDA, USE_CUDA_PATH AND USE_CUDNN are make file flags to set compilation options to use OpenCV, OpenBLAS, CUDA and cuDNN libraries. You can explore and use more compilation options in <code class="docutils literal"><span class="pre">make/config.mk</span></code>. Make sure to set USE_CUDA_PATH to right CUDA installation path. In most cases it is - <em>/usr/local/cuda</em>.</p>
+<p><br/></p>
+<p><strong>Build and install the MXNet R binding</strong></p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ make rpkg
+$ R CMD INSTALL mxnet_current_r.tar.gz
+</pre></div>
+</div>
+</div></div>
+</div><!-- START - Windows R CPU Installation Instructions --><div class="windows">
+ <div class="r">
+ <div class="cpu"><p>The CPU version of MXNet R package can be installed in R like other packages</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span>cran <span class="o"><-</span> <span class="kp">getOption</span><span class="p">(</span><span class="s">"repos"</span><span class="p">)</span>
+cran<span class="p">[</span><span class="s">"dmlc"</span><span class="p">]</span> <span class="o"><-</span> <span class="s">"https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/R/CRAN/"</span>
+<span class="kp">options</span><span class="p">(</span>repos <span class="o">=</span> cran<span class="p">)</span>
+install.packages<span class="p">(</span><span class="s">"mxnet"</span><span class="p">)</span>
+</pre></div>
+</div>
+</div><!-- END - Windows R CPU Installation Instructions --><div class="gpu"><p>The GPU version of MXNet R package can be installed in R like other packages</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span>cran <span class="o"><-</span> <span class="kp">getOption</span><span class="p">(</span><span class="s">"repos"</span><span class="p">)</span>
+cran<span class="p">[</span><span class="s">"dmlc"</span><span class="p">]</span> <span class="o"><-</span> <span class="s">"https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/R/CRAN/GPU"</span>
+<span class="kp">options</span><span class="p">(</span>repos <span class="o">=</span> cran<span class="p">)</span>
+install.packages<span class="p">(</span><span class="s">"mxnet"</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>Alternatively, You can also follow the installation instructions <a class="reference internal" href="windows_setup.html"><span class="doc">in this guide</span></a> to build MXNet from source.</p>
+</div>
+</div>
+</div><!-- END - Windows R GPU Installation Instructions --><div class="linux">
+ <div class="scala julia perl">
+ <div class="cpu gpu"><p>Follow the installation instructions <a class="reference internal" href="ubuntu_setup.html"><span class="doc">in this guide</span></a> to set up MXNet.</p>
+</div>
+</div>
+</div><div class="macos">
+ <div class="scala julia perl">
+ <div class="cpu gpu"><p>Follow the installation instructions <a class="reference internal" href="osx_setup.html"><span class="doc">in this guide</span></a> to set up MXNet.</p>
+</div>
+</div>
+</div><div class="windows">
+ <div class="python scala julia perl">
+ <div class="gpu"><p>Follow the installation instructions <a class="reference internal" href="windows_setup.html"><span class="doc">in this guide</span></a> to set up MXNet.</p>
+</div>
+</div>
+</div><div class="devices">
+ <div class="raspberry-pi"><p>MXNet supports the Debian based Raspbian ARM based operating system so you can run MXNet on Raspberry Pi Devices.</p>
+<p>These instructions will walk through how to build MXNet for the Raspberry Pi and install the Python bindings for the library.</p>
+<p>The complete MXNet library and its requirements can take almost 200MB of RAM, and loading large models with the library can take over 1GB of RAM. Because of this, we recommend running MXNet on the Raspberry Pi 3 or an equivalent device that has more than 1 GB of RAM and a Secure Digital (SD) card that has at least 4 GB of free memory.</p>
+<p><strong>Install MXNet</strong></p>
+<p>Installing MXNet is a two-step process:</p>
+<ol class="simple">
+<li>Build the shared library from the MXNet C++ source code.</li>
+<li>Install the supported language-specific packages for MXNet.</li>
+</ol>
+<p><strong>Step 1</strong> Build the Shared Library</p>
+<p>On Raspbian versions Wheezy and later, you need the following dependencies:</p>
+<ul class="simple">
+<li>Git (to pull code from GitHub)</li>
+<li>libblas (for linear algebraic operations)</li>
+<li>libopencv (for computer vision operations. This is optional if you want to save RAM and Disk Space)</li>
+<li>A C++ compiler that supports C++ 11. The C++ compiler compiles and builds MXNet source code. Supported compilers include the following:</li>
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">G++ (4.8 or later)</a></li>
+</ul>
+<p>Install these dependencies using the following commands in any directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> sudo apt-get update
+ sudo apt-get -y install git cmake build-essential g++-4.8 c++-4.8 liblapack* libblas* libopencv*
+</pre></div>
+</div>
+<p>Clone the MXNet source code repository using the following <code class="docutils literal"><span class="pre">git</span></code> command in your home directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> git clone https://github.com/dmlc/mxnet.git --recursive
+ <span class="nb">cd</span> mxnet
+</pre></div>
+</div>
+<p>If you aren’t processing images with MXNet on the Raspberry Pi, you can minimize the size of the compiled library by building MXNet without the Open Source Computer Vision (OpenCV) library with the following commands:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> <span class="nb">export</span> <span class="nv">USE_OPENCV</span> <span class="o">=</span> <span class="m">0</span>
+ make
+</pre></div>
+</div>
+<p>Otherwise, you can build the complete MXNet library with the following command:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> make
+</pre></div>
+</div>
+<p>Executing either of these commands start the build process, which can take up to a couple hours, and creates a file called <code class="docutils literal"><span class="pre">libmxnet.so</span></code> in the mxnet/lib directory.</p>
+<p>If you are getting build errors in which the compiler is being killed, it is likely that the compiler is running out of memory (especially if you are on Raspberry Pi 1, 2 or Zero, which have less than 1GB of RAM), this can often be rectified by increasing the swapfile size on the Pi by editing the file /etc/dphys-swapfile and changing the line CONF_SWAPSIZE=100 to CONF_SWAPSIZE=1024, then running:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> sudo /etc/init.d/dphys-swapfile stop
+ sudo /etc/init.d/dphys-swapfile start
+ free -m <span class="c1"># to verify the swapfile size has been increased</span>
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install MXNet Python Bindings</p>
+<p>To install python bindings run the following commands in the MXNet directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> <span class="nb">cd</span> python
+ pip install --upgrade pip
+ pip install -e .
+</pre></div>
+</div>
+<p>Note that the <code class="docutils literal"><span class="pre">-e</span></code> flag is optional. It is equivalent to <code class="docutils literal"><span class="pre">--editable</span></code> and means that if you edit the source files, these changes will be reflected in the package installed.</p>
+<p>You are now ready to run MXNet on your Raspberry Pi device. You can get started by following the tutorial on <a class="reference external" href="http://mxnet.io/tutorials/embedded/wine_detector.html">Real-time Object Detection with MXNet On The Raspberry Pi</a>.</p>
+<p><em>Note - Because the complete MXNet library takes up a significant amount of the Raspberry Pi’s limited RAM, when loading training data or large models into memory, you might have to turn off the GUI and terminate running processes to free RAM.</em></p>
+</div><div class="nvidia-jetson-tx2"><p>MXNet supports the Ubuntu Arch64 based operating system so you can run MXNet on NVIDIA Jetson Devices.</p>
+<p>These instructions will walk through how to build MXNet for the Pascal based <a class="reference external" href="http://www.nvidia.com/object/embedded-systems-dev-kits-modules.html">NVIDIA Jetson TX2</a> and install the corresponding python language bindings.</p>
+<p>For the purposes of this install guide we will assume that CUDA is already installed on your Jetson device.</p>
+<p><strong>Install MXNet</strong></p>
+<p>Installing MXNet is a two-step process:</p>
+<ol class="simple">
+<li>Build the shared library from the MXNet C++ source code.</li>
+<li>Install the supported language-specific packages for MXNet.</li>
+</ol>
+<p><strong>Step 1</strong> Build the Shared Library</p>
+<p>You need the following additional dependencies:</p>
+<ul class="simple">
+<li>Git (to pull code from GitHub)</li>
+<li>libatlas (for linear algebraic operations)</li>
+<li>libopencv (for computer vision operations)</li>
+<li>python pip (to load relevant python packages for our language bindings)</li>
+</ul>
+<p>Install these dependencies using the following commands in any directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> sudo apt-get update
+ sudo apt-get -y install git build-essential libatlas-base-dev libopencv-dev graphviz python-pip
+ sudo pip install pip --upgrade
+ sudo pip install setuptools numpy --upgrade
+ sudo pip install graphviz jupyter
+</pre></div>
+</div>
+<p>Clone the MXNet source code repository using the following <code class="docutils literal"><span class="pre">git</span></code> command in your home directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> git clone https://github.com/dmlc/mxnet.git --recursive
+ <span class="nb">cd</span> mxnet
+</pre></div>
+</div>
+<p>Edit the Makefile to install the MXNet with CUDA bindings to leverage the GPU on the Jetson:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> cp make/config.mk .
+ <span class="nb">echo</span> <span class="s2">"USE_CUDA=1"</span> >> config.mk
+ <span class="nb">echo</span> <span class="s2">"USE_CUDA_PATH=/usr/local/cuda"</span> >> config.mk
+ <span class="nb">echo</span> <span class="s2">"USE_CUDNN=1"</span> >> config.mk
+</pre></div>
+</div>
+<p>Edit the Mshadow Makefile to ensure MXNet builds with Pascal’s hardware level low precision acceleration by editing mshadow/make/mshadow.mk and adding the following after line 122:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="nv">MSHADOW_CFLAGS</span> <span class="o">+=</span> -DMSHADOW_USE_PASCAL<span class="o">=</span><span class="m">1</span>
+</pre></div>
+</div>
+<p>Now you can build the complete MXNet library with the following command:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> make -j <span class="k">$(</span>nproc<span class="k">)</span>
+</pre></div>
+</div>
+<p>Executing this command creates a file called <code class="docutils literal"><span class="pre">libmxnet.so</span></code> in the mxnet/lib directory.</p>
+<p><strong>Step 2</strong> Install MXNet Python Bindings</p>
+<p>To install python bindings run the following commands in the MXNet directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> <span class="nb">cd</span> python
+ pip install --upgrade pip
+ pip install -e .
+</pre></div>
+</div>
+<p>Note that the <code class="docutils literal"><span class="pre">-e</span></code> flag is optional. It is equivalent to <code class="docutils literal"><span class="pre">--editable</span></code> and means that if you edit the source files, these changes will be reflected in the package installed.</p>
+<p>Add the mxnet folder to the path:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> <span class="nb">cd</span> ..
+ <span class="nb">export</span> <span class="nv">MXNET_HOME</span><span class="o">=</span><span class="k">$(</span><span class="nb">pwd</span><span class="k">)</span>
+ <span class="nb">echo</span> <span class="s2">"export PYTHONPATH=</span><span class="nv">$MXNET_HOME</span><span class="s2">/python:</span><span class="nv">$PYTHONPATH</span><span class="s2">"</span> >> ~/.bashrc
+ <span class="nb">source</span> ~/.bashrc
+</pre></div>
+</div>
+<p>You are now ready to run MXNet on your NVIDIA Jetson TX2 device.</p>
+</div>
+</div><p><br/></p>
+</div>
+<div class="section" id="validate-mxnet-installation">
+<span id="validate-mxnet-installation"></span><h1>Validate MXNet Installation<a class="headerlink" href="#validate-mxnet-installation" title="Permalink to this headline">¶</a></h1>
+<div class="linux macos">
+ <div class="python">
+ <div class="cpu"><div class="pip build-from-source"><p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+</div><div class="docker"><p>Launch a Docker container with <code class="docutils literal"><span class="pre">mxnet/python</span></code> image and run example <em>MXNet</em> python program on the terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker run -it mxnet/python bash <span class="c1"># Use sudo if you skip Step 2 in the installation instruction</span>
+
+<span class="c1"># Start a python terminal</span>
+root@4919c4f58cac:/# python
+</pre></div>
+</div>
+</div><div class="virtualenv"><p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+</div><p>Run a short <em>MXNet</em> python program to create a 2X3 matrix of ones, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">mxnet</span> <span class="kn">as</span> <span class="nn">mx</span>
+<span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">))</span>
+<span class="gp">>>> </span><span class="n">b</span> <span class="o">=</span> <span class="n">a</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+<span class="gp">>>> </span><span class="n">b</span><span class="o">.</span><span class="n">asnumpy</span><span class="p">()</span>
+<span class="go">array([[ 3., 3., 3.],</span>
+<span class="go"> [ 3., 3., 3.]], dtype=float32)</span>
+</pre></div>
+</div>
+</div>
+</div>
+</div><!-- Validation for GPU machines --><div class="linux">
+ <div class="python">
+ <div class="gpu"><div class="pip build-from-source"><p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+</div><div class="docker"><p>Launch a NVIDIA Docker container with <code class="docutils literal"><span class="pre">mxnet/python:gpu</span></code> image and run example <em>MXNet</em> python program on the terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ nvidia-docker run -it mxnet/python:gpu bash <span class="c1"># Use sudo if you skip Step 2 in the installation instruction</span>
+
+<span class="c1"># Start a python terminal</span>
+root@4919c4f58cac:/# python
+</pre></div>
+</div>
+</div><div class="virtualenv"><p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+</div><p>Run a short <em>MXNet</em> python program to create a 2X3 matrix of ones <em>a</em> on a <em>GPU</em>, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3. We use <em>mx.gpu()</em>, to set <em>MXNet</em> context to be GPUs.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">mxnet</span> <span class="kn">as</span> <span class="nn">mx</span>
+<span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">mx</span><span class="o">.</span><span class="n">gpu</span><span class="p">())</span>
+<span class="gp">>>> </span><span class="n">b</span> <span class="o">=</span> <span class="n">a</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+<span class="gp">>>> </span><span class="n">b</span><span class="o">.</span><span class="n">asnumpy</span><span class="p">()</span>
+<span class="go">array([[ 3., 3., 3.],</span>
+<span class="go"> [ 3., 3., 3.]], dtype=float32)</span>
+</pre></div>
+</div>
+</div>
+</div>
+</div><div class="macos">
+ <div class="python">
+ <div class="gpu"><p>More details and verified validation instructions for macOS, with GPUs, coming soon.</p>
+</div>
+</div>
+</div><!-- Linux Clean up -->
+<div class="linux">
+ <div class="python">
+ <div class="cpu"><div class="pip build-from-source"><p>Exit the Python terminal.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><div class="virtualenv"><p>Exit the Python terminal and Deactivate the virtualenv <em>MXNet</em> environment.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">(mxnet)$ deactivate</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><div class="docker"><p>Exit the Python terminal and mxnet/python docker container.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">root@4919c4f58cac:/# exit</span>
+</pre></div>
+</div>
+</div></div>
+</div>
+</div><!-- MacOS Clean up -->
+<div class="macos">
+ <div class="python">
+ <div class="cpu"><div class="pip build-from-source"><p>Exit the Python terminal.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><div class="virtualenv"><p>Exit the Python terminal and Deactivate the virtualenv <em>MXNet</em> environment.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">(mxnet)$ deactivate</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><div class="docker"><p>Exit the Python terminal and then the docker container.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">root@4919c4f58cac:/# exit</span>
+</pre></div>
+</div>
+</div></div>
+</div>
+</div><!-- Validation for cloud installation --><div class="cloud"><p>Login to the cloud instance you launched, with pre-installed <em>MXNet</em>, following the guide by corresponding cloud provider.</p>
+<p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+<!-- Example Python code for CPU --><div class="cpu"><p>Run a short <em>MXNet</em> python program to create a 2X3 matrix of ones, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">mxnet</span> <span class="kn">as</span> <span class="nn">mx</span>
+<span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">))</span>
+<span class="gp">>>> </span><span class="n">b</span> <span class="o">=</span> <span class="n">a</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+<span class="gp">>>> </span><span class="n">b</span><span class="o">.</span><span class="n">asnumpy</span><span class="p">()</span>
+<span class="go">array([[ 3., 3., 3.],</span>
+<span class="go"> [ 3., 3., 3.]], dtype=float32)</span>
+</pre></div>
+</div>
+<p>Exit the Python terminal.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><!-- Example Python code for CPU --><div class="gpu"><p>Run a short <em>MXNet</em> python program to create a 2X3 matrix of ones <em>a</em> on a <em>GPU</em>, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3. We use <em>mx.gpu()</em>, to set <em>MXNet</em> context to be GPUs.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">mxnet</span> <span class="kn">as</span> <span class="nn">mx</span>
+<span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">mx</span><span class="o">.</span><span class="n">gpu</span><span class="p">())</span>
+<span class="gp">>>> </span><span class="n">b</span> <span class="o">=</span> <span class="n">a</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+<span class="gp">>>> </span><span class="n">b</span><span class="o">.</span><span class="n">asnumpy</span><span class="p">()</span>
+<span class="go">array([[ 3., 3., 3.],</span>
+<span class="go"> [ 3., 3., 3.]], dtype=float32)</span>
+</pre></div>
+</div>
+</div></div><!-- Example R code for CPU --><div class="linux macos windows">
+ <div class="r">
+ <div class="cpu"><p>Run a short <em>MXNet</em> R program to create a 2X3 matrix of ones, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3.</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span><span class="kn">library</span><span class="p">(</span>mxnet<span class="p">)</span>
+a <span class="o"><-</span> mx.nd.ones<span class="p">(</span><span class="kt">c</span><span class="p">(</span><span class="m">2</span><span class="p">,</span><span class="m">3</span><span class="p">),</span> ctx <span class="o">=</span> mx.cpu<span class="p">())</span>
+b <span class="o"><-</span> a <span class="o">*</span> <span class="m">2</span> <span class="o">+</span> <span class="m">1</span>
+b
+</pre></div>
+</div>
+</div>
+</div>
+</div><!-- Example R code for GPU --><div class="linux macos windows">
+ <div class="r">
+ <div class="gpu"><p>Run a short <em>MXNet</em> R program to create a 2X3 matrix of ones <em>a</em> on a <em>GPU</em>, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3. We use <em>mx.gpu()</em>, to set <em>MXNet</em> context to be GPUs.</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span><span class="kn">library</span><span class="p">(</span>mxnet<span class="p">)</span>
+a <span class="o"><-</span> mx.nd.ones<span class="p">(</span><span class="kt">c</span><span class="p">(</span><span class="m">2</span><span class="p">,</span><span class="m">3</span><span class="p">),</span> ctx <span class="o">=</span> mx.gpu<span class="p">())</span>
+b <span class="o"><-</span> a <span class="o">*</span> <span class="m">2</span> <span class="o">+</span> <span class="m">1</span>
+b
+</pre></div>
+</div>
+</div>
+</div>
+</div><div class="linux">
+ <div class="scala julia perl">
+ <div class="cpu gpu"><p>Will be available soon.</p>
+</div>
+</div>
+</div><div class="macos">
+ <div class="scala julia perl">
+ <div class="cpu gpu"><p>Will be available soon.</p>
+</div>
+</div>
+</div><div class="windows">
+ <div class="python">
+ <div class="cpu">
+ </div>
+ <div class="gpu"><p>The following installation instructions have been tested on Ubuntu 14.04 and 16.04.</p>
+<p><strong>Prerequisites</strong></p>
+<p>Install the following NVIDIA libraries to setup <em>MXNet</em> with GPU support:</p>
+<ol class="simple">
+<li>Install CUDA 8.0 following the NVIDIA’s <a class="reference external" href="http://docs.nvidia.com/cuda/cuda-installation-guide-microsoft-windows">installation guide</a>.</li>
+<li>Install cuDNN 7 for CUDA 8.0 following the NVIDIA’s <a class="reference external" href="https://developer.nvidia.com/cudnn">installation guide</a>. You may need to register with NVIDIA for downloading the cuDNN library.</li>
+</ol>
+<p><strong>Note:</strong> Make sure to add CUDA install path to <code class="docutils literal"><span class="pre">PATH</span></code>.</p>
+<div class="pip">
+<br/><p><strong>Step 1</strong> Install python.</p>
+<p>Recommend install <code class="docutils literal"><span class="pre">Anaconda3</span></code> <a class="reference external" href="https://www.anaconda.com/download/">here</a></p>
+<p><strong>Step 2</strong> Install <em>MXNet</em> with GPU support using CUDA 8.0</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet-cu80==1.0.0
+</pre></div>
+</div>
+</div>
+<div class="build-from-source">
+<br/>
+To build and install MXNet yourself, you need the following dependencies. Install the required dependencies:<ol class="simple">
+<li>If <a class="reference external" href="https://www.visualstudio.com/downloads/">Microsoft Visual Studio 2015</a> is not already installed, download and install it. You can download and install the free community edition.</li>
+<li>Download and Install <a class="reference external" href="https://cmake.org/">CMake</a> if it is not already installed.</li>
+<li>Download and install <a class="reference external" href="http://sourceforge.net/projects/opencvlibrary/files/opencv-win/3.0.0/opencv-3.0.0.exe/download">OpenCV</a>.</li>
+<li>Unzip the OpenCV package.</li>
+<li>Set the environment variable <code class="docutils literal"><span class="pre">OpenCV_DIR</span></code> to point to the <code class="docutils literal"><span class="pre">OpenCV</span> <span class="pre">build</span> <span class="pre">directory</span></code>.</li>
+<li>If you don’t have the Intel Math Kernel Library (MKL) installed, download and install <a class="reference external" href="http://sourceforge.net/projects/openblas/files/v0.2.14/">OpenBlas</a>.</li>
+<li>Set the environment variable <code class="docutils literal"><span class="pre">OpenBLAS_HOME</span></code> to point to the <code class="docutils literal"><span class="pre">OpenBLAS</span></code> directory that contains the <code class="docutils literal"><span class="pre">include</span></code> and <code class="docutils literal"><span class="pre">lib</span></code> directories. Typically, you can find the directory in <code class="docutils literal"><span class="pre">C:\Program</span> <span class="pre">files</span> <span class="pre">(x86)\OpenBLAS\</span></code>.</li>
+<li>Download and install <a class="reference external" href="https://developer.nvidia.com/cudnn">CuDNN</a>. To get access to the download link, register as an NVIDIA community user.</li>
+</ol>
+<p>After you have installed all of the required dependencies, build the MXNet source code:</p>
+<ol class="simple">
+<li>Download the MXNet source code from <a class="reference external" href="https://github.com/dmlc/mxnet">GitHub</a>.</li>
+<li>Use <a class="reference external" href="https://cmake.org/">CMake</a> to create a Visual Studio solution in <code class="docutils literal"><span class="pre">./build</span></code>.</li>
+<li>In Visual Studio, open the solution file,<code class="docutils literal"><span class="pre">.sln</span></code>, and compile it.
+These commands produce a library called <code class="docutils literal"><span class="pre">mxnet.dll</span></code> in the <code class="docutils literal"><span class="pre">./build/Release/</span></code> or <code class="docutils literal"><span class="pre">./build/Debug</span></code> folder.</li>
+</ol>
+<p><p> </p>
+
+Next, we install <code class="docutils literal"><span class="pre">graphviz</span></code> library that we use for visualizing network graphs you build on MXNet. We will also install <a class="reference external" href="http://jupyter.readthedocs.io/">Jupyter Notebook</a> used for running MXNet tutorials and examples.</p>
+<ul class="simple">
+<li>Install <code class="docutils literal"><span class="pre">graphviz</span></code> by downloading MSI installer from <a class="reference external" href="http://www.graphviz.org/Download_windows.php">Graphviz Download Page</a>.
+<strong>Note</strong> Make sure to add graphviz executable path to PATH environment variable. Refer <a class="reference external" href="http://stackoverflow.com/questions/35064304/runtimeerror-make-sure-the-graphviz-executables-are-on-your-systems-path-aft">here for more details</a></li>
+</ul>
+<p><p> </p>
+</p>
+</div> </div>
+ </div>
+ <div class="scala julia perl">
+ <div class="cpu gpu"><p>Will be available soon.</p>
+ </div>
+ </div>
+</div><div class="devices">
+ <div class="raspberry-pi"><p>Will be available soon.</p>
+</div>
+<div class="nvidia-jetson-tx2"><p>Will be available soon.</p>
+</div>
+</div></div>
+<div class="section" id="download-source-package">
+<span id="download-source-package"></span><h1>Download Source Package<a class="headerlink" href="#download-source-package" title="Permalink to this headline">¶</a></h1>
+</div>
+
+
+ </div>
+ </div>
+
+ <div class="sphinxsidebar rightsidebar" role="navigation" aria-label="main navigation">
+ <div class="sphinxsidebarwrapper">
+ <h3><a href="../index.html">Table Of Contents</a></h3>
+ <ul>
+<li><a class="reference internal" href="#">Installing MXNet</a></li>
+<li><a class="reference internal" href="#validate-mxnet-installation">Validate MXNet Installation</a></li>
+<li><a class="reference internal" href="#download-source-package">Download Source Package</a></li>
+</ul>
+
+ </div>
+ </div>
+ </div><!--
+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="footer">
+<div class="section-disclaimer">
+<div class="container">
+ <div>
+ <img src="https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/apache_incubator_logo.png" height=60>
+ <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 src="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.6/js/bootstrap.min.js" integrity="sha384-0mSbJDEHialfmuBBQP6A4Qrprq5OVfW37PRR3j5ELqxss1yVqOtnepnHVP9aJ7xS" crossorigin="anonymous"></script>
+ <script type="text/javascript" src="../_static/js/sidebar.js"></script>
+ <script type="text/javascript" src="../_static/js/search.js"></script>
+ <script type="text/javascript" src="../_static/js/navbar.js"></script>
+ <script type="text/javascript" src="../_static/js/clipboard.min.js"></script>
+ <script type="text/javascript" src="../_static/js/copycode.js"></script>
+ <script type="text/javascript" src="../_static/js/page.js"></script>
+ <script type="text/javascript">
+ $('body').ready(function () {
+ $('body').css('visibility', 'visible');
+ });
+ </script>
+ </body>
+</html>
diff --git a/docs/build_version_doc/artifacts/1.1.0.index.html b/docs/build_version_doc/artifacts/1.1.0.index.html
new file mode 100644
index 00000000000..2de028e3f75
--- /dev/null
+++ b/docs/build_version_doc/artifacts/1.1.0.index.html
@@ -0,0 +1,1338 @@
+<!DOCTYPE html>
+
+<html lang="en">
+ <head>
+ <meta charset="utf-8">
+ <meta http-equiv="X-UA-Compatible" content="IE=edge">
+ <meta name="viewport" content="width=device-width, initial-scale=1">
+
+
+
+ <title>Installing MXNet — mxnet documentation</title>
+
+ <link rel="stylesheet" href="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.6/css/bootstrap.min.css" integrity="sha384-1q8mTJOASx8j1Au+a5WDVnPi2lkFfwwEAa8hDDdjZlpLegxhjVME1fgjWPGmkzs7" crossorigin="anonymous">
+ <link rel="stylesheet" href="https://maxcdn.bootstrapcdn.com/font-awesome/4.5.0/css/font-awesome.min.css">
+
+
+ <link rel="stylesheet" href="../_static/basic.css" type="text/css" />
+ <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+
+ <link rel="stylesheet" href="../_static/mxnet.css" type="text/css">
+
+ <script type="text/javascript">
+ var DOCUMENTATION_OPTIONS = {
+ URL_ROOT: '../',
+ VERSION: '',
+ COLLAPSE_INDEX: false,
+ FILE_SUFFIX: '.html',
+ HAS_SOURCE: true,
+ SOURCELINK_SUFFIX: '.txt'
+ };
+ </script>
+
+ <script type="text/javascript" src="https://code.jquery.com/jquery-1.11.1.min.js"></script>
+
+ <script type="text/javascript" src="../_static/underscore.js"></script>
+
+ <script type="text/javascript" src="../_static/searchtools_custom.js"></script>
+
+ <script type="text/javascript" src="../_static/doctools.js"></script>
+
+ <script type="text/javascript" src="../_static/selectlang.js"></script>
+
+
+ <script type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.1/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></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://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.0/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script> -->
+ <!-- -->
+ <link rel="index" title="Index" href="../genindex.html" />
+ <link rel="search" title="Search" href="../search.html" />
+
+ <link rel="icon" type="image/png" href="https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/mxnet-icon.png">
+ </head>
+ <body role="document" background="https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/mxnet-background-compressed.jpeg">
+ <div class='content-block'><div class="navbar navbar-fixed-top">
+ <div class="container" id="navContainer">
+ <div id="header-inner" class="innder">
+ <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 id="main-nav" class='nav-bar'>
+ <a class="main-nav-link" href="../install/index.html">Install</a>
+
+ <span id="dropdown-menu-position-anchor">
+ <a href="#" class="main-nav-link dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">Gluon <span class="caret"></span></a>
+ <ul id="package-dropdown-menu" class="dropdown-menu navbar-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 href="#" class="main-nav-link dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">API <span class="caret"></span></a>
+ <ul id="package-dropdown-menu" class="dropdown-menu navbar-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 href="#" class="main-nav-link dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">Docs <span class="caret"></span></a>
+ <ul id="package-dropdown-menu-docs" class="dropdown-menu navbar-menu">
+ <li><a class="main-nav-link" href="../tutorials/index.html">Tutorials</a>
+ <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/master/example">Examples</a></li>
+ <li><a class="main-nav-link" href="../api/python/gluon/model_zoo.html">Gluon 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 href="#" class="main-nav-link dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">Community <span class="caret"></span></a>
+ <ul id="package-dropdown-menu-community" class="dropdown-menu navbar-menu">
+ <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>
+ <li><a class="main-nav-link" href="http://discuss.mxnet.io">Discuss</a></li>
+ </ul>
+ </span>
+ </nav>
+ <script> function getRootPath(){ return "../" } </script>
+ <div class="burgerIcon dropdown">
+ <a href="#" class="dropdown-toggle" data-toggle="dropdown" role="button">☰</a>
+ <ul id="burgerMenu" class="dropdown-menu">
+ <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 tabindex="-1" href="../community/index.html">Community</a></li>
+ <li><a tabindex="-1" href="../community/contribute.html">Contribute</a></li>
+ <li><a tabindex="-1" href="../community/powered_by.html">Powered By</a></li>
+ </ul>
+ </li>
+
+ <li class="dropdown-submenu">
+ <a href="#" tabindex="-1">API</a>
+ <ul class="dropdown-menu">
+
+ <li><a tabindex="-1" href="../api/python/index.html">Python</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/scala/index.html">Scala</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/r/index.html">R</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/julia/index.html">Julia</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/c++/index.html">C++</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/perl/index.html">Perl</a>
+ </li>
+
+ </ul>
+ </li>
+
+ <li class="dropdown-submenu">
+ <a href="#" tabindex="-1">Docs</a>
+ <ul class="dropdown-menu">
+ <li><a tabindex="-1" href="../tutorials/index.html">Tutorials</a></li>
+ <li><a tabindex="-1" href="../faq/index.html">FAQ</a></li>
+ <li><a tabindex="-1" href="../architecture/index.html">Architecture</a></li>
+ <li><a tabindex="-1" href="https://github.com/apache/incubator-mxnet/tree/master/example">Examples</a></li>
+ <li><a tabindex="-1" href="../api/python/gluon/model_zoo.html">Gluon 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>
+ </ul>
+ </div>
+ <div class="plusIcon dropdown">
+ <a href="#" class="dropdown-toggle" data-toggle="dropdown" role="button"><span class="glyphicon glyphicon-plus" aria-hidden="true"></span></a>
+ <ul id="plusMenu" class="dropdown-menu dropdown-menu-right"></ul>
+ </div>
+ <div id="search-input-wrap">
+
+<form class="" role="search" action="../search.html" method="get" autocomplete="off">
+ <div class="form-group inner-addon left-addon">
+ <i class="glyphicon glyphicon-search"></i>
+ <input type="text" name="q" class="form-control" placeholder="Search">
+ </div>
+ <input type="hidden" name="check_keywords" value="yes" />
+ <input type="hidden" name="area" value="default" />
+
+</form>
+ <div id="search-preview"></div>
+ </div>
+ <div id='searchIcon'>
+ <span class="glyphicon glyphicon-search" aria-hidden="true"></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 class="sphinxsidebar leftsidebar" role="navigation" aria-label="main 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>
+
+ <div class="section" id="installing-mxnet">
+<span id="installing-mxnet"></span><h1>Installing MXNet<a class="headerlink" href="#installing-mxnet" title="Permalink to this headline">¶</a></h1>
+<p>Indicate your preferred configuration. Then, follow the customized commands to install <em>MXNet</em>.</p>
+<script type="text/javascript" src='../_static/js/options.js'></script><!-- START - OS Menu --><div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">Linux</button>
+ <button type="button" class="btn btn-default opt">MacOS</button>
+ <button type="button" class="btn btn-default opt">Windows</button>
+ <button type="button" class="btn btn-default opt">Cloud</button>
+ <button type="button" class="btn btn-default opt">Devices</button>
+</div><!-- START - Language Menu --><div class="linux macos windows">
+<div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">Python</button>
+ <button type="button" class="btn btn-default opt">Scala</button>
+ <button type="button" class="btn btn-default opt">R</button>
+ <button type="button" class="btn btn-default opt">Julia</button>
+ <button type="button" class="btn btn-default opt">Perl</button>
+</div>
+</div><!-- No CPU GPU for other Devices -->
+<div class="linux macos windows cloud">
+<div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">CPU</button>
+ <button type="button" class="btn btn-default opt">GPU</button>
+</div>
+</div><!-- other devices -->
+<div class="devices">
+<div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">Raspberry Pi</button>
+ <button type="button" class="btn btn-default opt">NVIDIA Jetson TX2</button>
+</div>
+</div><!-- Linux Python GPU Options --><div class="linux macos windows">
+<div class="python">
+<div class="cpu gpu">
+<div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">Pip</button>
+ <button type="button" class="btn btn-default opt">Virtualenv</button>
+ <button type="button" class="btn btn-default opt">Docker</button>
+ <button type="button" class="btn btn-default opt">Build from Source</button>
+</div>
+</div>
+</div>
+</div><!-- END - Main Menu --><!-- START - Linux Python CPU Installation Instructions --><div class="linux">
+ <div class="python">
+ <div class="cpu"><p>The following installation instructions have been tested on Ubuntu 14.04 and 16.04.</p>
+<div class="virtualenv">
+<br/><p><strong>Step 1</strong> Install virtualenv for Ubuntu.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y python-dev python-virtualenv
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Create and activate virtualenv environment for MXNet.</p>
+<p>Following command creates a virtualenv environment at <code class="docutils literal"><span class="pre">~/mxnet</span></code> directory. However, you can choose any directory by replacing <code class="docutils literal"><span class="pre">~/mxnet</span></code> with a directory of your choice.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ virtualenv --system-site-packages ~/mxnet
+</pre></div>
+</div>
+<p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install MXNet in the active virtualenv environment.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install --upgrade pip
+</pre></div>
+</div>
+<p>Install <em>MXNet</em> with OpenBLAS acceleration.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet==1.1.0
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 5</strong> Validate the installation by running simple <em>MXNet</em> code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Note</strong> You can read more about virtualenv <a class="reference external" href="https://virtualenv.pypa.io/en/stable/userguide/">here</a>.</p>
+</div><div class="pip">
+<br/><p><strong>Step 1</strong> Install prerequisites - wget and latest pip.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command in the terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y wget python gcc
+$ wget https://bootstrap.pypa.io/get-pip.py <span class="o">&&</span> sudo python get-pip.py
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install MXNet with OpenBLAS acceleration.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet==1.1.0
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Experimental Choice</strong> If You would like to install mxnet with Intel MKL, try the experimental pip package with MKL:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet-mkl==1.1.0
+</pre></div>
+</div>
+</div><div class="docker">
+<br/><p>Docker images with <em>MXNet</em> are available at <a class="reference external" href="https://hub.docker.com/r/mxnet/">Docker Hub</a>.</p>
+<p><strong>Step 1</strong> Install Docker on your machine by following the <a class="reference external" href="https://docs.docker.com/engine/installation/linux/ubuntu/#install-using-the-repository">docker installation instructions</a>.</p>
+<p><em>Note</em> - You can install Community Edition (CE) to get started with <em>MXNet</em>.</p>
+<p><strong>Step 2</strong> [Optional] Post installation steps to manage Docker as a non-root user.</p>
+<p>Follow the four steps in this <a class="reference external" href="https://docs.docker.com/engine/installation/linux/linux-postinstall/#manage-docker-as-a-non-root-user">docker documentation</a> to allow managing docker containers without <em>sudo</em>.</p>
+<p>If you skip this step, you need to use <em>sudo</em> each time you invoke Docker.</p>
+<p><strong>Step 2</strong> Pull the MXNet docker image.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker pull mxnet/python <span class="c1"># Use sudo if you skip Step 2</span>
+</pre></div>
+</div>
+<p>You can list docker images to see if mxnet/python docker image pull was successful.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker images <span class="c1"># Use sudo if you skip Step 2</span>
+
+REPOSITORY TAG IMAGE ID CREATED SIZE
+mxnet/python latest 00d026968b3c <span class="m">3</span> weeks ago <span class="m">1</span>.41 GB
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div><div class="build-from-source">
+<br/><p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings. Example - Python bindings, Scala bindings.</li>
+</ol>
+<p><strong>Minimum Requirements</strong></p>
+<ol class="simple">
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">GCC 4.8</a> or later to compile C++ 11.</li>
+<li><a class="reference external" href="https://www.gnu.org/software/make/">GNU Make</a></li>
+</ol>
+<p><br/></p>
+<p><strong>Build the MXNet core shared library</strong></p>
+<p><strong>Step 1</strong> Install build tools and git.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y build-essential git
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install OpenBLAS.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="https://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprograms">BLAS</a> and <a class="reference external" href="https://en.wikipedia.org/wiki/LAPACK">LAPACK</a> libraries for accelerated numerical computations on CPU machine. There are several flavors of BLAS/LAPACK libraries - <a class="reference external" href="http://www.openblas.net/">OpenBLAS</a>, <a class="reference external" href="http://math-atlas.sourceforge.net/">ATLAS</a> and <a class="reference external" href="https://software.intel.com/en-us/intel-mkl">MKL</a>. In this step we install OpenBLAS. You can choose to install ATLAS or MKL.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopenblas-dev liblapack-dev
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install OpenCV.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="http://opencv.org/">OpenCV</a> for efficient image loading and augmentation operations.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopencv-dev
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Download MXNet sources and build MXNet core shared library. You can clone the repository as described in the following code block, or you may try the <a href="download.html">download links</a> for your desired MXNet version.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ git clone --recursive https://github.com/apache/incubator-mxnet
+$ <span class="nb">cd</span> incubator-mxnet
+$ make -j <span class="k">$(</span>nproc<span class="k">)</span> <span class="nv">USE_OPENCV</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_BLAS</span><span class="o">=</span>openblas
+</pre></div>
+</div>
+<p><em>Note</em> - USE_OPENCV and USE_BLAS are make file flags to set compilation options to use OpenCV and BLAS library. You can explore and use more compilation options in <code class="docutils literal"><span class="pre">make/config.mk</span></code>.</p>
+<p><br/></p>
+<p><strong>Build the MXNet Python binding</strong></p>
+<p><strong>Step 1</strong> Install prerequisites - python, setup-tools, python-pip and libfortran (required for Numpy).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y python-dev python-setuptools python-pip libgfortran3
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install the MXNet Python binding.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">cd</span> python
+$ pip install --upgrade pip
+$ pip install -e .
+</pre></div>
+</div>
+<p>Note that the <code class="docutils literal"><span class="pre">-e</span></code> flag is optional. It is equivalent to <code class="docutils literal"><span class="pre">--editable</span></code> and means that if you edit the source files, these changes will be reflected in the package installed.</p>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div></div>
+</div>
+</div>
+<!-- END - Linux Python CPU Installation Instructions --><!-- START - Linux Python GPU Installation Instructions --><div class="linux">
+<div class="python">
+<div class="gpu"><p>The following installation instructions have been tested on Ubuntu 14.04 and 16.04.</p>
+<p><strong>Prerequisites</strong></p>
+<p>Install the following NVIDIA libraries to setup <em>MXNet</em> with GPU support:</p>
+<ol class="simple">
+<li>Install CUDA 9.0 following the NVIDIA’s <a class="reference external" href="http://docs.nvidia.com/cuda/cuda-installation-guide-linux/">installation guide</a>.</li>
+<li>Install cuDNN 7 for CUDA 9.0 following the NVIDIA’s <a class="reference external" href="https://developer.nvidia.com/cudnn">installation guide</a>. You may need to register with NVIDIA for downloading the cuDNN library.</li>
+</ol>
+<p><strong>Note:</strong> Make sure to add CUDA install path to <code class="docutils literal"><span class="pre">LD_LIBRARY_PATH</span></code>.</p>
+<p>Example - <em>export LD_LIBRARY_PATH=/usr/local/cuda/lib64/:$LD_LIBRARY_PATH</em></p>
+<div class="pip">
+<br/><p><strong>Step 1</strong> Install prerequisites - wget and latest pip.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command in the terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y wget python
+$ wget https://bootstrap.pypa.io/get-pip.py <span class="o">&&</span> sudo python get-pip.py
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install <em>MXNet</em> with GPU support using CUDA 9.0</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet-cu90==1.1.0
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Experimental Choice</strong> If You would like to install mxnet with Intel MKL, try the experimental pip package with MKL:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet-cu90mkl==1.1.0
+</pre></div>
+</div>
+</div><div class="virtualenv"><p><br/></p>
+<p><strong>Step 1</strong> Install virtualenv for Ubuntu.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y python-dev python-virtualenv
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Create and activate virtualenv environment for MXNet.</p>
+<p>Following command creates a virtualenv environment at <code class="docutils literal"><span class="pre">~/mxnet</span></code> directory. However, you can choose any directory by replacing <code class="docutils literal"><span class="pre">~/mxnet</span></code> with a directory of your choice.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ virtualenv --system-site-packages ~/mxnet
+</pre></div>
+</div>
+<p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install MXNet in the active virtualenv environment.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install --upgrade pip
+</pre></div>
+</div>
+<p>Install <em>MXNet</em> with GPU support using CUDA 9.0.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet-cu90==1.1.0
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 5</strong> Validate the installation by running simple <em>MXNet</em> code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Note</strong> You can read more about virtualenv <a class="reference external" href="https://virtualenv.pypa.io/en/stable/userguide/">here</a>.</p>
+</div><div class="docker"><p><br/></p>
+<p>Docker images with <em>MXNet</em> are available at <a class="reference external" href="https://hub.docker.com/r/mxnet/">Docker Hub</a>.</p>
+<p><strong>Step 1</strong> Install Docker on your machine by following the <a class="reference external" href="https://docs.docker.com/engine/installation/linux/ubuntu/#install-using-the-repository">docker installation instructions</a>.</p>
+<p><em>Note</em> - You can install Community Edition (CE) to get started with <em>MXNet</em>.</p>
+<p><strong>Step 2</strong> [Optional] Post installation steps to manage Docker as a non-root user.</p>
+<p>Follow the four steps in this <a class="reference external" href="https://docs.docker.com/engine/installation/linux/linux-postinstall/#manage-docker-as-a-non-root-user">docker documentation</a> to allow managing docker containers without <em>sudo</em>.</p>
+<p>If you skip this step, you need to use <em>sudo</em> each time you invoke Docker.</p>
+<p><strong>Step 3</strong> Install <em>nvidia-docker-plugin</em> following the <a class="reference external" href="https://github.com/NVIDIA/nvidia-docker/wiki/Installation">installation instructions</a>. <em>nvidia-docker-plugin</em> is required to enable the usage of GPUs from the docker containers.</p>
+<p><strong>Step 4</strong> Pull the MXNet docker image.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker pull mxnet/python:gpu <span class="c1"># Use sudo if you skip Step 2</span>
+</pre></div>
+</div>
+<p>You can list docker images to see if mxnet/python docker image pull was successful.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker images <span class="c1"># Use sudo if you skip Step 2</span>
+
+REPOSITORY TAG IMAGE ID CREATED SIZE
+mxnet/python gpu 493b2683c269 <span class="m">3</span> weeks ago <span class="m">4</span>.77 GB
+</pre></div>
+</div>
+<p><strong>Step 5</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div><div class="build-from-source"><p><br/></p>
+<p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings. Example - Python bindings, Scala bindings.</li>
+</ol>
+<p><strong>Minimum Requirements</strong></p>
+<ol class="simple">
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">GCC 4.8</a> or later to compile C++ 11.</li>
+<li><a class="reference external" href="https://www.gnu.org/software/make/">GNU Make</a></li>
+</ol>
+<p><br/></p>
+<p><strong>Build the MXNet core shared library</strong></p>
+<p><strong>Step 1</strong> Install build tools and git.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y build-essential git
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install OpenBLAS.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="https://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprograms">BLAS</a> and <a class="reference external" href="https://en.wikipedia.org/wiki/LAPACK">LAPACK</a> libraries for accelerated numerical computations on CPU machine. There are several flavors of BLAS/LAPACK libraries - <a class="reference external" href="http://www.openblas.net/">OpenBLAS</a>, <a class="reference external" href="http://math-atlas.sourceforge.net/">ATLAS</a> and <a class="reference external" href="https://software.intel.com/en-us/intel-mkl">MKL</a>. In this step we install OpenBLAS. You can choose to install ATLAS or MKL.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopenblas-dev liblapack-dev
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install OpenCV.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="http://opencv.org/">OpenCV</a> for efficient image loading and augmentation operations.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopencv-dev
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Download MXNet sources and build MXNet core shared library. You can clone the repository as described in the following code block, or you may try the <a href="download.html">download links</a> for your desired MXNet version.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ git clone --recursive https://github.com/apache/incubator-mxnet
+$ <span class="nb">cd</span> incubator-mxnet
+$ make -j <span class="k">$(</span>nproc<span class="k">)</span> <span class="nv">USE_OPENCV</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_BLAS</span><span class="o">=</span>openblas <span class="nv">USE_CUDA</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_CUDA_PATH</span><span class="o">=</span>/usr/local/cuda <span class="nv">USE_CUDNN</span><span class="o">=</span><span class="m">1</span>
+</pre></div>
+</div>
+<p><em>Note</em> - USE_OPENCV, USE_BLAS, USE_CUDA, USE_CUDA_PATH AND USE_CUDNN are make file flags to set compilation options to use OpenCV, OpenBLAS, CUDA and cuDNN libraries. You can explore and use more compilation options in <code class="docutils literal"><span class="pre">make/config.mk</span></code>. Make sure to set USE_CUDA_PATH to right CUDA installation path. In most cases it is - <em>/usr/local/cuda</em>.</p>
+<p><br/></p>
+<p><strong>Install the MXNet Python binding</strong></p>
+<p><strong>Step 1</strong> Install prerequisites - python, setup-tools, python-pip and libfortran (required for Numpy)..</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y python-dev python-setuptools python-pip libgfortran3
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install the MXNet Python binding.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">cd</span> python
+$ pip install --upgrade pip
+$ pip install -e .
+</pre></div>
+</div>
+<p>Note that the <code class="docutils literal"><span class="pre">-e</span></code> flag is optional. It is equivalent to <code class="docutils literal"><span class="pre">--editable</span></code> and means that if you edit the source files, these changes will be reflected in the package installed.</p>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div></div>
+</div>
+</div>
+<!-- END - Linux Python GPU Installation Instructions --><!-- START - MacOS Python CPU Installation Instructions --><div class="macos">
+ <div class="python">
+ <div class="cpu"><p>The following installation instructions have been tested on OSX Sierra and El Capitan.</p>
+<div class="virtualenv">
+<br/><p><strong>Step 1</strong> Install prerequisites - Homebrew, python development tools.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="c1"># Install Homebrew</span>
+$ /usr/bin/ruby -e <span class="s2">"</span><span class="k">$(</span>curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install<span class="k">)</span><span class="s2">"</span>
+$ <span class="nb">export</span> <span class="nv">PATH</span><span class="o">=</span>/usr/local/bin:/usr/local/sbin:<span class="nv">$PATH</span>
+
+<span class="c1"># Install python development tools - python2.7, pip, python-setuptools</span>
+$ brew install python
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install virtualenv for macOS.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install virtualenv
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Create and activate virtualenv environment for MXNet.</p>
+<p>Following command creates a virtualenv environment at <code class="docutils literal"><span class="pre">~/mxnet</span></code> directory. However, you can choose any directory by replacing <code class="docutils literal"><span class="pre">~/mxnet</span></code> with a directory of your choice.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ virtualenv --system-site-packages ~/mxnet
+</pre></div>
+</div>
+<p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Install MXNet in the active virtualenv environment.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install --upgrade pip
+<span class="o">(</span>mxnet<span class="o">)</span>$ pip install --upgrade setuptools
+</pre></div>
+</div>
+<p>Install <em>MXNet</em> with OpenBLAS acceleration.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet==1.1.0
+</pre></div>
+</div>
+<p><strong>Step 5</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ brew install graphviz
+<span class="o">(</span>mxnet<span class="o">)</span>$ pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 6</strong> Validate the installation by running simple <em>MXNet</em> code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Note</strong> You can read more about virtualenv <a class="reference external" href="https://virtualenv.pypa.io/en/stable/userguide/">here</a>.</p>
+</div><div class="pip">
+<br/><p><strong>Step 1</strong> Install prerequisites - Homebrew, python development tools.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="c1"># Install Homebrew</span>
+$ /usr/bin/ruby -e <span class="s2">"</span><span class="k">$(</span>curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install<span class="k">)</span><span class="s2">"</span>
+$ <span class="nb">export</span> <span class="nv">PATH</span><span class="o">=</span>/usr/local/bin:/usr/local/sbin:<span class="nv">$PATH</span>
+
+<span class="c1"># Install python development tools - python2.7, pip, python-setuptools</span>
+$ brew install python
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install MXNet with OpenBLAS acceleration.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install --upgrade pip
+$ pip install --upgrade setuptools
+</pre></div>
+</div>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet==1.1.0
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ brew install graphviz
+$ pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div><div class="docker">
+<br/><p>Docker images with <em>MXNet</em> are available at <a class="reference external" href="https://hub.docker.com/r/mxnet/">Docker Hub</a>.</p>
+<p><strong>Step 1</strong> Install Docker on your machine by following the <a class="reference external" href="https://docs.docker.com/docker-for-mac/install/#install-and-run-docker-for-mac">docker installation instructions</a>.</p>
+<p><em>Note</em> - You can install Community Edition (CE) to get started with <em>MXNet</em>.</p>
+<p><strong>Step 2</strong> Pull the MXNet docker image.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker pull mxnet/python
+</pre></div>
+</div>
+<p>You can list docker images to see if mxnet/python docker image pull was successful.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker images
+
+REPOSITORY TAG IMAGE ID CREATED SIZE
+mxnet/python latest 00d026968b3c <span class="m">3</span> weeks ago <span class="m">1</span>.41 GB
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div><div class="build-from-source">
+<br/><p><strong>Prerequisites</strong></p>
+<p>If not already installed, <a class="reference external" href="https://developer.apple.com/xcode/">download and install Xcode</a> (or <a class="reference external" href="https://itunes.apple.com/us/app/xcode/id497799835">insall it from the App Store</a>) for macOS. <a class="reference external" href="https://en.wikipedia.org/wiki/Xcode">Xcode</a> is an integrated development environment for macOS containing a suite of software development tools like C/C++ compilers, BLAS library and more.</p>
+<p><br/></p>
+<p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings. Example - Python bindings, Scala bindings.</li>
+</ol>
+<p>Make sure you have installed Xcode before proceeding further.</p>
+<p><br/></p>
+<p>All the instructions to build <em>MXNet</em> core shared library and <em>MXNet</em> Python bindings are compiled as one helper <em>bash</em> script. You can use <a class="reference external" href="https://raw.githubusercontent.com/dmlc/mxnet/master/setup-utils/install-mxnet-osx-python.sh">this bash script</a> to build <em>MXNet</em> for Python, from source, on macOS.</p>
+<p><strong>Step 1</strong> Download the bash script for building MXNet from source.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ curl -O https://raw.githubusercontent.com/dmlc/mxnet/master/setup-utils/install-mxnet-osx-python.sh
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Run the script to get latest MXNet source and build.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="c1"># Make the script executable</span>
+$ chmod <span class="m">744</span> install-mxnet-osx-python.sh
+
+<span class="c1"># Run the script. It takes around 5 mins.</span>
+$ bash install-mxnet-osx-python.sh
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div> <!-- End of source build --></div>
+</div>
+</div><!-- END - Mac OS Python CPU Installation Instructions --><!-- START - Mac OS Python GPU Installation Instructions --><div class="macos">
+ <div class="python">
+ <div class="gpu"><div class="pip virtualenv docker">
+</br><p>Try the <strong>Build from Source</strong> option for now.</p>
+</div><div class="build-from-source"><p><strong>Step 1</strong> Install prerequisites - Homebrew, python development tools.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="c1"># Install Homebrew</span>
+$ /usr/bin/ruby -e <span class="s2">"</span><span class="k">$(</span>curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install<span class="k">)</span><span class="s2">"</span>
+$ <span class="nb">export</span> <span class="nv">PATH</span><span class="o">=</span>/usr/local/bin:/usr/local/sbin:<span class="nv">$PATH</span>
+
+<span class="c1"># Install python development tools - python2.7, pip, python-setuptools</span>
+$ brew install python pkg-config graphviz
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install optional components - OpenCV</p>
+<p>If you want to use OpenCV you should install it first, then build MXNet with the <code class="docutils literal"><span class="pre">USE_OPENCV=1</span></code> option in the later steps.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>brew tap homebrew/science
+brew install opencv
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install CUDA and cuDNN</p>
+<p>The following instructions are for CUDA 9.1 and cuDNN 7 for macOS 10.12+ and a CUDA-capable GPU. They summarize confirmed successful builds in <a class="reference external" href="https://github.com/apache/incubator-mxnet/issues/9217">#9217</a>.
+Alternatively, you may follow the <a class="reference external" href="https://docs.nvidia.com/cuda/cuda-installation-guide-mac-os-x/index.html">CUDA installation instructions for macOS</a>.</p>
+<ol class="simple">
+<li><a class="reference external" href="https://developer.apple.com/download/more/">Download Xcode 8.3.3 from Apple</a>. This is the version <a class="reference external" href="https://docs.nvidia.com/cuda/cuda-installation-guide-mac-os-x/index.html">NVIDIA specifies in its instructions for macOS</a>. Unzip and rename to <code class="docutils literal"><span class="pre">Xcode8.3.3.app</span></code>.</li>
+<li>Run <code class="docutils literal"><span class="pre">sudo</span> <span class="pre">xcode-select</span> <span class="pre">-s</span> <span class="pre">/Applications/Xcode8.3.3.app</span></code> or to wherever you have placed Xcode.</li>
+<li>Run <code class="docutils literal"><span class="pre">xcode-select</span> <span class="pre">--install</span></code> to install all command line tools, compilers, etc.</li>
+<li>Run <code class="docutils literal"><span class="pre">sudo</span> <span class="pre">xcodebuild</span> <span class="pre">-license</span> <span class="pre">accept</span></code> to accept Xcode’s licensing terms.</li>
+<li>Install CUDA for macOS. Specific steps are provided in NVIDIA’s <a class="reference external" href="https://docs.nvidia.com/cuda/cuda-installation-guide-mac-os-x/index.html#installation">CUDA installation instructions</a>.</li>
+<li><a class="reference external" href="http://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html#download-mac">Download</a> and <a class="reference external" href="http://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html#installmac">install</a> cuDNN for macOS. You will need to <a class="reference external" href="https://developer.nvidia.com/accelerated-computing-developer">create a free developer account</a> with NVIDIA prior to getting the download link.</li>
+</ol>
+<p><strong>Step 4</strong> Build MXNet</p>
+<ol>
+<li><p class="first">Run <code class="docutils literal"><span class="pre">git</span> <span class="pre">clone</span> <span class="pre">--recursive</span> <span class="pre">https://github.com/apache/incubator-mxnet.git</span> <span class="pre">mxnet</span></code> to get the latest version.</p>
+</li>
+<li><p class="first">Run <code class="docutils literal"><span class="pre">cd</span> <span class="pre">mxnet</span></code>.</p>
+</li>
+<li><p class="first">Edit the <code class="docutils literal"><span class="pre">make/osx.mk</span></code> file to set the following parameters:</p>
+<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">USE_CUDA</span> <span class="o">=</span> <span class="mi">1</span>
+<span class="n">USE_CUDA_PATH</span> <span class="o">=</span> <span class="o">/</span><span class="n">usr</span><span class="o">/</span><span class="n">local</span><span class="o">/</span><span class="n">cuda</span>
+<span class="n">USE_CUDNN</span> <span class="o">=</span> <span class="mi">1</span>
+<span class="n">USE_OPENCV</span> <span class="o">=</span> <span class="mi">0</span> <span class="c1"># set to 1 if you want to build with OpenCV</span>
+</pre></div>
+</div>
+</li>
+<li><p class="first">Copy the <code class="docutils literal"><span class="pre">make/osx.mk</span></code> to <code class="docutils literal"><span class="pre">config.mk</span></code></p>
+</li>
+<li><p class="first">Run <code class="docutils literal"><span class="pre">make</span></code>. If you previously attempted to compile you might want to do <code class="docutils literal"><span class="pre">make</span> <span class="pre">clean_all</span></code> first. You can also run <code class="docutils literal"><span class="pre">make</span> <span class="pre">-j</span></code> with the number of processors you have to compile with multithreading. There’ll be plenty of warnings, but there should be no errors.</p>
+</li>
+<li><p class="first">Once finished, you should have a file called <code class="docutils literal"><span class="pre">libmxnet.so</span></code> in <code class="docutils literal"><span class="pre">lib/</span></code>.</p>
+</li>
+<li><p class="first">Do <code class="docutils literal"><span class="pre">cd</span> <span class="pre">python</span></code>.</p>
+</li>
+<li><p class="first">Run <code class="docutils literal"><span class="pre">sudo</span> <span class="pre">pip</span> <span class="pre">install</span> <span class="pre">-e</span> <span class="pre">.</span></code> <strong>Note</strong>: the <code class="docutils literal"><span class="pre">.</span></code> is part of the command.</p>
+</li>
+</ol>
+</div>
+</div>
+</div>
+</div><!-- END - Mac OS Python GPU Installation Instructions --><!-- START - Cloud Python Installation Instructions --><div class="cloud"><p>AWS Marketplace distributes Deep Learning AMIs (Amazon Machine Image) with MXNet pre-installed. You can launch one of these Deep Learning AMIs by following instructions in the <a class="reference external" href="http://docs.aws.amazon.com/dlami/latest/devguide/what-is-dlami.html">AWS Deep Learning AMI Developer Guide</a>.</p>
+<p>You can also run distributed deep learning with <em>MXNet</em> on AWS using <a class="reference external" href="https://github.com/awslabs/deeplearning-cfn/blob/master/README.md">Cloudformation Template</a>.</p>
+</div><!-- END - Cloud Python Installation Instructions --><!-- START - MacOS R CPU Installation Instructions --><div class="macos">
+ <div class="r">
+ <div class="cpu"><p>The CPU version of MXNet R package can be installed in R like other packages</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span>cran <span class="o"><-</span> <span class="kp">getOption</span><span class="p">(</span><span class="s">"repos"</span><span class="p">)</span>
+cran<span class="p">[</span><span class="s">"dmlc"</span><span class="p">]</span> <span class="o"><-</span> <span class="s">"https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/R/CRAN/"</span>
+<span class="kp">options</span><span class="p">(</span>repos <span class="o">=</span> cran<span class="p">)</span>
+install.packages<span class="p">(</span><span class="s">"mxnet"</span><span class="p">)</span>
+</pre></div>
+</div>
+</div><div class="gpu"><p>Will be available soon.</p>
+</div></div>
+</div>
+<!-- END - MacOS R CPU Installation Instructions --><div class="linux">
+ <div class="r">
+ <div class="cpu">
+<br/><p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings.</li>
+</ol>
+<p><strong>Minimum Requirements</strong></p>
+<ol class="simple">
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">GCC 4.8</a> or later to compile C++ 11.</li>
+<li><a class="reference external" href="https://www.gnu.org/software/make/">GNU Make</a></li>
+</ol>
+<p><br/></p>
+<p><strong>Build the MXNet core shared library</strong></p>
+<p><strong>Step 1</strong> Install build tools and git.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y build-essential git
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install OpenBLAS.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="https://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprograms">BLAS</a> and <a class="reference external" href="https://en.wikipedia.org/wiki/LAPACK">LAPACK</a> libraries for accelerated numerical computations on CPU machine. There are several flavors of BLAS/LAPACK libraries - <a class="reference external" href="http://www.openblas.net/">OpenBLAS</a>, <a class="reference external" href="http://math-atlas.sourceforge.net/">ATLAS</a> and <a class="reference external" href="https://software.intel.com/en-us/intel-mkl">MKL</a>. In this step we install OpenBLAS. You can choose to install ATLAS or MKL.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopenblas-dev liblapack-dev
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install OpenCV.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="http://opencv.org/">OpenCV</a> for efficient image loading and augmentation operations.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopencv-dev
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Download MXNet sources and build MXNet core shared library. You can clone the repository as described in the following code block, or you may try the <a href="download.html">download links</a> for your desired MXNet version.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ git clone --recursive https://github.com/apache/incubator-mxnet
+$ <span class="nb">cd</span> incubator-mxnet
+$ make -j <span class="k">$(</span>nproc<span class="k">)</span> <span class="nv">USE_OPENCV</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_BLAS</span><span class="o">=</span>openblas
+</pre></div>
+</div>
+<p><em>Note</em> - USE_OPENCV and USE_BLAS are make file flags to set compilation options to use OpenCV and BLAS library. You can explore and use more compilation options in <code class="docutils literal"><span class="pre">make/config.mk</span></code>.</p>
+<p><br/></p>
+<p><strong>Build and install the MXNet R binding</strong></p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ make rpkg
+$ R CMD INSTALL mxnet_current_r.tar.gz
+</pre></div>
+</div>
+</div><div class="gpu"><p>The following installation instructions have been tested on Ubuntu 14.04 and 16.04.</p>
+<p><strong>Prerequisites</strong></p>
+<p>Install the following NVIDIA libraries to setup <em>MXNet</em> with GPU support:</p>
+<ol class="simple">
+<li>Install CUDA 9.0 following the NVIDIA’s <a class="reference external" href="http://docs.nvidia.com/cuda/cuda-installation-guide-linux/">installation guide</a>.</li>
+<li>Install cuDNN 7 for CUDA 9.0 following the NVIDIA’s <a class="reference external" href="https://developer.nvidia.com/cudnn">installation guide</a>. You may need to register with NVIDIA for downloading the cuDNN library.</li>
+</ol>
+<p><strong>Note:</strong> Make sure to add CUDA install path to <code class="docutils literal"><span class="pre">LD_LIBRARY_PATH</span></code>.</p>
+<p>Example - <em>export LD_LIBRARY_PATH=/usr/local/cuda/lib64/:$LD_LIBRARY_PATH</em></p>
+<p><br/></p>
+<p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings.</li>
+</ol>
+<p><strong>Minimum Requirements</strong></p>
+<ol class="simple">
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">GCC 4.8</a> or later to compile C++ 11.</li>
+<li><a class="reference external" href="https://www.gnu.org/software/make/">GNU Make</a></li>
+</ol>
+<p><br/></p>
+<p><strong>Build the MXNet core shared library</strong></p>
+<p><strong>Step 1</strong> Install build tools and git.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y build-essential git
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install OpenBLAS.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="https://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprograms">BLAS</a> and <a class="reference external" href="https://en.wikipedia.org/wiki/LAPACK">LAPACK</a> libraries for accelerated numerical computations on CPU machine. There are several flavors of BLAS/LAPACK libraries - <a class="reference external" href="http://www.openblas.net/">OpenBLAS</a>, <a class="reference external" href="http://math-atlas.sourceforge.net/">ATLAS</a> and <a class="reference external" href="https://software.intel.com/en-us/intel-mkl">MKL</a>. In this step we install OpenBLAS. You can choose to install ATLAS or MKL.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopenblas-dev liblapack-dev
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install OpenCV.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="http://opencv.org/">OpenCV</a> for efficient image loading and augmentation operations.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopencv-dev
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Download MXNet sources and build MXNet core shared library. You can clone the repository as described in the following code block, or you may try the <a href="download.html">download links</a> for your desired MXNet version.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ git clone --recursive https://github.com/apache/incubator-mxnet
+$ <span class="nb">cd</span> incubator-mxnet
+$ make -j <span class="k">$(</span>nproc<span class="k">)</span> <span class="nv">USE_OPENCV</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_BLAS</span><span class="o">=</span>openblas <span class="nv">USE_CUDA</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_CUDA_PATH</span><span class="o">=</span>/usr/local/cuda <span class="nv">USE_CUDNN</span><span class="o">=</span><span class="m">1</span>
+</pre></div>
+</div>
+<p><em>Note</em> - USE_OPENCV, USE_BLAS, USE_CUDA, USE_CUDA_PATH AND USE_CUDNN are make file flags to set compilation options to use OpenCV, OpenBLAS, CUDA and cuDNN libraries. You can explore and use more compilation options in <code class="docutils literal"><span class="pre">make/config.mk</span></code>. Make sure to set USE_CUDA_PATH to right CUDA installation path. In most cases it is - <em>/usr/local/cuda</em>.</p>
+<p><br/></p>
+<p><strong>Build and install the MXNet R binding</strong></p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ make rpkg
+$ R CMD INSTALL mxnet_current_r.tar.gz
+</pre></div>
+</div>
+</div></div>
+</div><!-- START - Windows R CPU Installation Instructions --><div class="windows">
+ <div class="r">
+ <div class="cpu"><p>The CPU version of MXNet R package can be installed in R like other packages</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span>cran <span class="o"><-</span> <span class="kp">getOption</span><span class="p">(</span><span class="s">"repos"</span><span class="p">)</span>
+cran<span class="p">[</span><span class="s">"dmlc"</span><span class="p">]</span> <span class="o"><-</span> <span class="s">"https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/R/CRAN/"</span>
+<span class="kp">options</span><span class="p">(</span>repos <span class="o">=</span> cran<span class="p">)</span>
+install.packages<span class="p">(</span><span class="s">"mxnet"</span><span class="p">)</span>
+</pre></div>
+</div>
+</div><!-- END - Windows R CPU Installation Instructions --><div class="gpu"><p>The GPU version of MXNet R package can be installed in R like other packages</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span>cran <span class="o"><-</span> <span class="kp">getOption</span><span class="p">(</span><span class="s">"repos"</span><span class="p">)</span>
+cran<span class="p">[</span><span class="s">"dmlc"</span><span class="p">]</span> <span class="o"><-</span> <span class="s">"https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/R/CRAN/GPU"</span>
+<span class="kp">options</span><span class="p">(</span>repos <span class="o">=</span> cran<span class="p">)</span>
+install.packages<span class="p">(</span><span class="s">"mxnet"</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>Alternatively, You can also follow the installation instructions <a class="reference internal" href="windows_setup.html"><span class="doc">in this guide</span></a> to build MXNet from source.</p>
+</div>
+</div>
+</div><!-- END - Windows R GPU Installation Instructions --><div class="linux">
+ <div class="scala julia perl">
+ <div class="cpu gpu"><p>Follow the installation instructions <a class="reference internal" href="ubuntu_setup.html"><span class="doc">in this guide</span></a> to set up MXNet.</p>
+</div>
+</div>
+</div><div class="macos">
+ <div class="scala julia perl">
+ <div class="cpu gpu"><p>Follow the installation instructions <a class="reference internal" href="osx_setup.html"><span class="doc">in this guide</span></a> to set up MXNet.</p>
+</div>
+</div>
+</div><div class="windows">
+ <div class="python">
+ <div class="cpu">
+ <div class="pip">
+<br/><p><strong>Step 1</strong> Install Python.</p>
+<p><a class="reference external" href="https://www.anaconda.com/download/">Anaconda</a> is recommended.</p>
+<p><strong>Step 2</strong> Install <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet==1.1.0
+</pre></div>
+</div>
+</div>
+</div> <div class="gpu"><div class="pip">
+<br/><p><strong>Step 1</strong> Install Python.</p>
+<p><a class="reference external" href="https://www.anaconda.com/download/">Anaconda</a> is recommended.</p>
+<p><strong>Step 2</strong> Install <em>MXNet</em> with GPU support using CUDA 9.0.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet-cu90==1.1.0
+</pre></div>
+</div>
+<p>Refer to <a class="reference external" href="https://github.com/apache/incubator-mxnet/issues/8671">#8671</a> for status on CUDA 9.1 support.</p>
+</div>
+<div class="build-from-source">
+<br/>
+To build and install MXNet yourself, you need the following dependencies. Install the required dependencies:<ol class="simple">
+<li>If <a class="reference external" href="https://www.visualstudio.com/downloads/">Microsoft Visual Studio 2015</a> is not already installed, download and install it. You can download and install the free community edition.</li>
+<li>Download and install <a class="reference external" href="https://cmake.org/">CMake</a> if it is not already installed.</li>
+<li>Download and install <a class="reference external" href="http://sourceforge.net/projects/opencvlibrary/files/opencv-win/3.0.0/opencv-3.0.0.exe/download">OpenCV</a>.</li>
+<li>Unzip the OpenCV package.</li>
+<li>Set the environment variable <code class="docutils literal"><span class="pre">OpenCV_DIR</span></code> to point to the <code class="docutils literal"><span class="pre">OpenCV</span> <span class="pre">build</span> <span class="pre">directory</span></code>.</li>
+<li>If you don’t have the Intel Math Kernel Library (MKL) installed, download and install <a class="reference external" href="http://sourceforge.net/projects/openblas/files/v0.2.14/">OpenBlas</a>.</li>
+<li>Set the environment variable <code class="docutils literal"><span class="pre">OpenBLAS_HOME</span></code> to point to the <code class="docutils literal"><span class="pre">OpenBLAS</span></code> directory that contains the <code class="docutils literal"><span class="pre">include</span></code> and <code class="docutils literal"><span class="pre">lib</span></code> directories. Typically, you can find the directory in <code class="docutils literal"><span class="pre">C:\Program</span> <span class="pre">files</span> <span class="pre">(x86)\OpenBLAS\</span></code>.</li>
+<li>Download and install <a class="reference external" href="https://developer.nvidia.com/cuda-downloads?target_os=Windows&target_arch=x86_64">CUDA</a> and <a class="reference external" href="https://developer.nvidia.com/cudnn">cuDNN</a>. To get access to the download link, register as an NVIDIA community user.</li>
+</ol>
+<p>After you have installed all of the required dependencies, build the MXNet source code:</p>
+<ol class="simple">
+<li>Download the MXNet source code from <a class="reference external" href="https://github.com/apache/incubator-mxnet">GitHub</a>.</li>
+<li>Use <a class="reference external" href="https://cmake.org/">CMake</a> to create a Visual Studio solution in <code class="docutils literal"><span class="pre">./build</span></code>.</li>
+<li>In Visual Studio, open the solution file,<code class="docutils literal"><span class="pre">.sln</span></code>, and compile it.
+These commands produce a library called <code class="docutils literal"><span class="pre">mxnet.dll</span></code> in the <code class="docutils literal"><span class="pre">./build/Release/</span></code> or <code class="docutils literal"><span class="pre">./build/Debug</span></code> folder.</li>
+</ol>
+<p><p> </p>
+
+Next, we install the <code class="docutils literal"><span class="pre">graphviz</span></code> library that we use for visualizing network graphs that you build on MXNet. We will also install <a class="reference external" href="http://jupyter.readthedocs.io/">Jupyter Notebook</a> which is used for running MXNet tutorials and examples.</p>
+<ul class="simple">
+<li>Install the <code class="docutils literal"><span class="pre">graphviz</span></code> by downloading the installer from the <a class="reference external" href="https://graphviz.gitlab.io/_pages/Download/Download_windows.html">Graphviz Download Page</a>.
+<strong>Note</strong> Make sure to add the <code class="docutils literal"><span class="pre">graphviz</span></code> executable path to the PATH environment variable. Refer <a class="reference external" href="http://stackoverflow.com/questions/35064304/runtimeerror-make-sure-the-graphviz-executables-are-on-your-systems-path-aft">here for more details</a></li>
+</ul>
+<p><p> </p>
+</p>
+</div>
+</div>
+</div><div class="windows">
+ <div class="scala julia perl">
+ <div class="cpu gpu"><p>Follow the installation instructions <a class="reference internal" href="windows_setup.html"><span class="doc">in this guide</span></a> to set up MXNet.</p>
+</div>
+</div>
+</div><div class="windows">
+ <div class="python">
+ <div class="cpu">
+ <div class="virtualenv docker build-from-source"><p>Follow the installation instructions <a class="reference internal" href="windows_setup.html"><span class="doc">in this guide</span></a> to set up MXNet.</p>
+</div>
+</div>
+</div>
+</div><div class="devices">
+ <div class="raspberry-pi"><p>MXNet supports the Debian based Raspbian ARM based operating system so you can run MXNet on Raspberry Pi Devices.</p>
+<p>These instructions will walk through how to build MXNet for the Raspberry Pi and install the Python bindings for the library.</p>
+<p>The complete MXNet library and its requirements can take almost 200MB of RAM, and loading large models with the library can take over 1GB of RAM. Because of this, we recommend running MXNet on the Raspberry Pi 3 or an equivalent device that has more than 1 GB of RAM and a Secure Digital (SD) card that has at least 4 GB of free memory.</p>
+<p><strong>Install MXNet</strong></p>
+<p>Installing MXNet is a two-step process:</p>
+<ol class="simple">
+<li>Build the shared library from the MXNet C++ source code.</li>
+<li>Install the supported language-specific packages for MXNet.</li>
+</ol>
+<p><strong>Step 1</strong> Build the Shared Library</p>
+<p>On Raspbian versions Wheezy and later, you need the following dependencies:</p>
+<ul class="simple">
+<li>Git (to pull code from GitHub)</li>
+<li>libblas (for linear algebraic operations)</li>
+<li>libopencv (for computer vision operations. This is optional if you want to save RAM and Disk Space)</li>
+<li>A C++ compiler that supports C++ 11. The C++ compiler compiles and builds MXNet source code. Supported compilers include the following:</li>
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">G++ (4.8 or later)</a></li>
+</ul>
+<p>Install these dependencies using the following commands in any directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> sudo apt-get update
+ sudo apt-get -y install git cmake build-essential g++-4.8 c++-4.8 liblapack* libblas* libopencv*
+</pre></div>
+</div>
+<p>Clone the MXNet source code repository using the following <code class="docutils literal"><span class="pre">git</span></code> command in your home directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> git clone https://github.com/apache/incubator-mxnet.git --recursive
+ <span class="nb">cd</span> incubator-mxnet
+</pre></div>
+</div>
+<p>If you aren’t processing images with MXNet on the Raspberry Pi, you can minimize the size of the compiled library by building MXNet without the Open Source Computer Vision (OpenCV) library with the following commands:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> <span class="nb">export</span> <span class="nv">USE_OPENCV</span> <span class="o">=</span> <span class="m">0</span>
+ make
+</pre></div>
+</div>
+<p>Otherwise, you can build the complete MXNet library with the following command:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> make
+</pre></div>
+</div>
+<p>Executing either of these commands start the build process, which can take up to a couple hours, and creates a file called <code class="docutils literal"><span class="pre">libmxnet.so</span></code> in the mxnet/lib directory.</p>
+<p>If you are getting build errors in which the compiler is being killed, it is likely that the compiler is running out of memory (especially if you are on Raspberry Pi 1, 2 or Zero, which have less than 1GB of RAM), this can often be rectified by increasing the swapfile size on the Pi by editing the file /etc/dphys-swapfile and changing the line CONF_SWAPSIZE=100 to CONF_SWAPSIZE=1024, then running:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> sudo /etc/init.d/dphys-swapfile stop
+ sudo /etc/init.d/dphys-swapfile start
+ free -m <span class="c1"># to verify the swapfile size has been increased</span>
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install MXNet Python Bindings</p>
+<p>To install Python bindings run the following commands in the MXNet directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> <span class="nb">cd</span> python
+ pip install --upgrade pip
+ pip install -e .
+</pre></div>
+</div>
+<p>Note that the <code class="docutils literal"><span class="pre">-e</span></code> flag is optional. It is equivalent to <code class="docutils literal"><span class="pre">--editable</span></code> and means that if you edit the source files, these changes will be reflected in the package installed.</p>
+<p>You are now ready to run MXNet on your Raspberry Pi device. You can get started by following the tutorial on <a class="reference external" href="http://mxnet.io/tutorials/embedded/wine_detector.html">Real-time Object Detection with MXNet On The Raspberry Pi</a>.</p>
+<p><em>Note - Because the complete MXNet library takes up a significant amount of the Raspberry Pi’s limited RAM, when loading training data or large models into memory, you might have to turn off the GUI and terminate running processes to free RAM.</em></p>
+</div><div class="nvidia-jetson-tx2"><p>MXNet supports the Ubuntu Arch64 based operating system so you can run MXNet on NVIDIA Jetson Devices.</p>
+<p>These instructions will walk through how to build MXNet for the Pascal based <a class="reference external" href="http://www.nvidia.com/object/embedded-systems-dev-kits-modules.html">NVIDIA Jetson TX2</a> and install the corresponding python language bindings.</p>
+<p>For the purposes of this install guide we will assume that CUDA is already installed on your Jetson device.</p>
+<p><strong>Install MXNet</strong></p>
+<p>Installing MXNet is a two-step process:</p>
+<ol class="simple">
+<li>Build the shared library from the MXNet C++ source code.</li>
+<li>Install the supported language-specific packages for MXNet.</li>
+</ol>
+<p><strong>Step 1</strong> Build the Shared Library</p>
+<p>You need the following additional dependencies:</p>
+<ul class="simple">
+<li>Git (to pull code from GitHub)</li>
+<li>libatlas (for linear algebraic operations)</li>
+<li>libopencv (for computer vision operations)</li>
+<li>python pip (to load relevant python packages for our language bindings)</li>
+</ul>
+<p>Install these dependencies using the following commands in any directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> sudo apt-get update
+ sudo apt-get -y install git build-essential libatlas-base-dev libopencv-dev graphviz python-pip
+ sudo pip install pip --upgrade
+ sudo pip install setuptools numpy --upgrade
+ sudo pip install graphviz jupyter
+</pre></div>
+</div>
+<p>Clone the MXNet source code repository using the following <code class="docutils literal"><span class="pre">git</span></code> command in your home directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> git clone https://github.com/apache/incubator-mxnet.git --recursive
+ <span class="nb">cd</span> incubator-mxnet
+</pre></div>
+</div>
+<p>Edit the Makefile to install the MXNet with CUDA bindings to leverage the GPU on the Jetson:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> cp make/config.mk .
+ <span class="nb">echo</span> <span class="s2">"USE_CUDA=1"</span> >> config.mk
+ <span class="nb">echo</span> <span class="s2">"USE_CUDA_PATH=/usr/local/cuda"</span> >> config.mk
+ <span class="nb">echo</span> <span class="s2">"USE_CUDNN=1"</span> >> config.mk
+</pre></div>
+</div>
+<p>Edit the Mshadow Makefile to ensure MXNet builds with Pascal’s hardware level low precision acceleration by editing mshadow/make/mshadow.mk and adding the following after line 122:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="nv">MSHADOW_CFLAGS</span> <span class="o">+=</span> -DMSHADOW_USE_PASCAL<span class="o">=</span><span class="m">1</span>
+</pre></div>
+</div>
+<p>Now you can build the complete MXNet library with the following command:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> make -j <span class="k">$(</span>nproc<span class="k">)</span>
+</pre></div>
+</div>
+<p>Executing this command creates a file called <code class="docutils literal"><span class="pre">libmxnet.so</span></code> in the mxnet/lib directory.</p>
+<p><strong>Step 2</strong> Install MXNet Python Bindings</p>
+<p>To install Python bindings run the following commands in the MXNet directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> <span class="nb">cd</span> python
+ pip install --upgrade pip
+ pip install -e .
+</pre></div>
+</div>
+<p>Note that the <code class="docutils literal"><span class="pre">-e</span></code> flag is optional. It is equivalent to <code class="docutils literal"><span class="pre">--editable</span></code> and means that if you edit the source files, these changes will be reflected in the package installed.</p>
+<p>Add the mxnet folder to the path:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> <span class="nb">cd</span> ..
+ <span class="nb">export</span> <span class="nv">MXNET_HOME</span><span class="o">=</span><span class="k">$(</span><span class="nb">pwd</span><span class="k">)</span>
+ <span class="nb">echo</span> <span class="s2">"export PYTHONPATH=</span><span class="nv">$MXNET_HOME</span><span class="s2">/python:</span><span class="nv">$PYTHONPATH</span><span class="s2">"</span> >> ~/.bashrc
+ <span class="nb">source</span> ~/.bashrc
+</pre></div>
+</div>
+<p>You are now ready to run MXNet on your NVIDIA Jetson TX2 device.</p>
+</div>
+</div><p><br/></p>
+<!-- This # tag restarts the page and allows reuse
+ of the div classes for validation sections, etc --></div>
+<div class="section" id="validate-mxnet-installation">
+<span id="validate-mxnet-installation"></span><h1>Validate MXNet Installation<a class="headerlink" href="#validate-mxnet-installation" title="Permalink to this headline">¶</a></h1>
+<div class="linux macos">
+ <div class="python">
+ <div class="cpu"><div class="pip build-from-source"><p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+</div><div class="docker"><p>Launch a Docker container with <code class="docutils literal"><span class="pre">mxnet/python</span></code> image and run example <em>MXNet</em> python program on the terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker run -it mxnet/python bash <span class="c1"># Use sudo if you skip Step 2 in the installation instruction</span>
+
+<span class="c1"># Start a python terminal</span>
+root@4919c4f58cac:/# python
+</pre></div>
+</div>
+</div><div class="virtualenv"><p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+</div><p>Run a short <em>MXNet</em> python program to create a 2X3 matrix of ones, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">mxnet</span> <span class="kn">as</span> <span class="nn">mx</span>
+<span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">))</span>
+<span class="gp">>>> </span><span class="n">b</span> <span class="o">=</span> <span class="n">a</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+<span class="gp">>>> </span><span class="n">b</span><span class="o">.</span><span class="n">asnumpy</span><span class="p">()</span>
+<span class="go">array([[ 3., 3., 3.],</span>
+<span class="go"> [ 3., 3., 3.]], dtype=float32)</span>
+</pre></div>
+</div>
+</div>
+</div>
+</div><!-- Validate Windows CPU pip install --><div class="windows">
+ <div class="python">
+ <div class="cpu">
+ <div class="pip"><p>Run a short <em>MXNet</em> python program to create a 2X3 matrix of ones, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">mxnet</span> <span class="kn">as</span> <span class="nn">mx</span>
+<span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">))</span>
+<span class="gp">>>> </span><span class="n">b</span> <span class="o">=</span> <span class="n">a</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+<span class="gp">>>> </span><span class="n">b</span><span class="o">.</span><span class="n">asnumpy</span><span class="p">()</span>
+<span class="go">array([[ 3., 3., 3.],</span>
+<span class="go"> [ 3., 3., 3.]], dtype=float32)</span>
+</pre></div>
+</div>
+</div>
+</div>
+</div>
+</div><!-- Mac OS GPU installation validation --><div class="macos">
+ <div class="python">
+ <div class="gpu"><div class="pip virtualenv docker">
+</br><p>Will be available soon.</p>
+</div><div class="build-from-source">
+</br><p>From the MXNet root directory run: <code class="docutils literal"><span class="pre">python</span> <span class="pre">example/image-classification/train_mnist.py</span> <span class="pre">--network</span> <span class="pre">lenet</span> <span class="pre">--gpus</span> <span class="pre">0</span></code> to test GPU training.</p>
+</div>
+</div>
+</div>
+</div><!-- Windows GPU installation validation --><div class="windows">
+ <div class="python">
+ <div class="gpu"><div class="virtualenv docker">
+</br><p>Will be available soon.</p>
+</div><div class="pip build-from-source">
+</br><p>From the MXNet root directory run: <code class="docutils literal"><span class="pre">python</span> <span class="pre">example/image-classification/train_mnist.py</span> <span class="pre">--network</span> <span class="pre">lenet</span> <span class="pre">--gpus</span> <span class="pre">0</span></code> to test GPU training.</p>
+</div>
+</div>
+</div>
+</div><!-- Validation for GPU machines --><div class="linux">
+ <div class="python">
+ <div class="gpu"><div class="pip build-from-source"><p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+</div><div class="docker"><p>Launch a NVIDIA Docker container with <code class="docutils literal"><span class="pre">mxnet/python:gpu</span></code> image and run example <em>MXNet</em> python program on the terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ nvidia-docker run -it mxnet/python:gpu bash <span class="c1"># Use sudo if you skip Step 2 in the installation instruction</span>
+
+<span class="c1"># Start a python terminal</span>
+root@4919c4f58cac:/# python
+</pre></div>
+</div>
+</div><div class="virtualenv"><p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+</div><p>Run a short <em>MXNet</em> python program to create a 2X3 matrix of ones <em>a</em> on a <em>GPU</em>, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3. We use <em>mx.gpu()</em>, to set <em>MXNet</em> context to be GPUs.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">mxnet</span> <span class="kn">as</span> <span class="nn">mx</span>
+<span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">mx</span><span class="o">.</span><span class="n">gpu</span><span class="p">())</span>
+<span class="gp">>>> </span><span class="n">b</span> <span class="o">=</span> <span class="n">a</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+<span class="gp">>>> </span><span class="n">b</span><span class="o">.</span><span class="n">asnumpy</span><span class="p">()</span>
+<span class="go">array([[ 3., 3., 3.],</span>
+<span class="go"> [ 3., 3., 3.]], dtype=float32)</span>
+</pre></div>
+</div>
+</div>
+</div>
+</div><!-- Linux Clean up -->
+<div class="linux">
+ <div class="python">
+ <div class="cpu"><div class="pip build-from-source"><p>Exit the Python terminal.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><div class="virtualenv"><p>Exit the Python terminal and Deactivate the virtualenv <em>MXNet</em> environment.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">(mxnet)$ deactivate</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><div class="docker"><p>Exit the Python terminal and mxnet/python docker container.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">root@4919c4f58cac:/# exit</span>
+</pre></div>
+</div>
+</div></div>
+</div>
+</div><!-- MacOS Clean up -->
+<div class="macos">
+ <div class="python">
+ <div class="cpu"><div class="pip build-from-source"><p>Exit the Python terminal.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><div class="virtualenv"><p>Exit the Python terminal and Deactivate the virtualenv <em>MXNet</em> environment.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">(mxnet)$ deactivate</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><div class="docker"><p>Exit the Python terminal and then the docker container.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">root@4919c4f58cac:/# exit</span>
+</pre></div>
+</div>
+</div></div>
+</div>
+</div><!-- Validation for cloud installation --><div class="cloud"><p>Login to the cloud instance you launched, with pre-installed <em>MXNet</em>, following the guide by corresponding cloud provider.</p>
+<p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+<!-- Example Python code for CPU --><div class="cpu"><p>Run a short <em>MXNet</em> python program to create a 2X3 matrix of ones, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">mxnet</span> <span class="kn">as</span> <span class="nn">mx</span>
+<span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">))</span>
+<span class="gp">>>> </span><span class="n">b</span> <span class="o">=</span> <span class="n">a</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+<span class="gp">>>> </span><span class="n">b</span><span class="o">.</span><span class="n">asnumpy</span><span class="p">()</span>
+<span class="go">array([[ 3., 3., 3.],</span>
+<span class="go"> [ 3., 3., 3.]], dtype=float32)</span>
+</pre></div>
+</div>
+<p>Exit the Python terminal.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><!-- Example Python code for CPU --><div class="gpu"><p>Run a short <em>MXNet</em> python program to create a 2X3 matrix of ones <em>a</em> on a <em>GPU</em>, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3. We use <em>mx.gpu()</em>, to set <em>MXNet</em> context to be GPUs.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">mxnet</span> <span class="kn">as</span> <span class="nn">mx</span>
+<span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">mx</span><span class="o">.</span><span class="n">gpu</span><span class="p">())</span>
+<span class="gp">>>> </span><span class="n">b</span> <span class="o">=</span> <span class="n">a</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+<span class="gp">>>> </span><span class="n">b</span><span class="o">.</span><span class="n">asnumpy</span><span class="p">()</span>
+<span class="go">array([[ 3., 3., 3.],</span>
+<span class="go"> [ 3., 3., 3.]], dtype=float32)</span>
+</pre></div>
+</div>
+</div></div><!-- Example R code for CPU --><div class="linux macos windows">
+ <div class="r">
+ <div class="cpu"><p>Run a short <em>MXNet</em> R program to create a 2X3 matrix of ones, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3.</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span><span class="kn">library</span><span class="p">(</span>mxnet<span class="p">)</span>
+a <span class="o"><-</span> mx.nd.ones<span class="p">(</span><span class="kt">c</span><span class="p">(</span><span class="m">2</span><span class="p">,</span><span class="m">3</span><span class="p">),</span> ctx <span class="o">=</span> mx.cpu<span class="p">())</span>
+b <span class="o"><-</span> a <span class="o">*</span> <span class="m">2</span> <span class="o">+</span> <span class="m">1</span>
+b
+</pre></div>
+</div>
+</div>
+</div>
+</div><!-- Example R code for GPU --><div class="linux macos windows">
+ <div class="r">
+ <div class="gpu"><p>Run a short <em>MXNet</em> R program to create a 2X3 matrix of ones <em>a</em> on a <em>GPU</em>, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3. We use <em>mx.gpu()</em>, to set <em>MXNet</em> context to be GPUs.</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span><span class="kn">library</span><span class="p">(</span>mxnet<span class="p">)</span>
+a <span class="o"><-</span> mx.nd.ones<span class="p">(</span><span class="kt">c</span><span class="p">(</span><span class="m">2</span><span class="p">,</span><span class="m">3</span><span class="p">),</span> ctx <span class="o">=</span> mx.gpu<span class="p">())</span>
+b <span class="o"><-</span> a <span class="o">*</span> <span class="m">2</span> <span class="o">+</span> <span class="m">1</span>
+b
+</pre></div>
+</div>
+</div>
+</div>
+</div><div class="linux">
+ <div class="scala julia perl">
+ <div class="cpu gpu"><p>Will be available soon.</p>
+</div>
+</div>
+</div><div class="macos">
+ <div class="scala julia perl">
+ <div class="cpu gpu"><p>Will be available soon.</p>
+</div>
+</div>
+</div><!-- Windows MXNet Installation validation -->
+<div class="windows">
+ <div class="python">
+ <div class="cpu"><div class="build-from-source virtualenv docker">
+<br/>
+Will be available soon.
+</div></div>
+</div><div class="scala julia perl">
+<div class="cpu gpu"><p>Will be available soon.</p>
+</div>
+</div>
+</div>
+<!-- End Windows Installation validation --><div class="devices">
+ <div class="raspberry-pi"><p>Will be available soon.</p>
+</div>
+<div class="nvidia-jetson-tx2"><p>Will be available soon.</p>
+</div>
+</div></div>
+<div class="section" id="download-source-package">
+<span id="download-source-package"></span><h1>Download Source Package<a class="headerlink" href="#download-source-package" title="Permalink to this headline">¶</a></h1>
+</div>
+
+
+ </div>
+ </div>
+
+ <div class="sphinxsidebar rightsidebar" role="navigation" aria-label="main navigation">
+ <div class="sphinxsidebarwrapper">
+ <h3><a href="../index.html">Table Of Contents</a></h3>
+ <ul>
+<li><a class="reference internal" href="#">Installing MXNet</a></li>
+<li><a class="reference internal" href="#validate-mxnet-installation">Validate MXNet Installation</a></li>
+<li><a class="reference internal" href="#download-source-package">Download Source Package</a></li>
+</ul>
+
+ </div>
+ </div>
+ </div><div class="footer">
+<div class="section-disclaimer">
+<div class="container">
+ <div>
+ <img src="https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/apache_incubator_logo.png" height=60>
+ <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 © 2017-2018, 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 src="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.6/js/bootstrap.min.js" integrity="sha384-0mSbJDEHialfmuBBQP6A4Qrprq5OVfW37PRR3j5ELqxss1yVqOtnepnHVP9aJ7xS" crossorigin="anonymous"></script>
+ <script type="text/javascript" src="../_static/js/sidebar.js"></script>
+ <script type="text/javascript" src="../_static/js/search.js"></script>
+ <script type="text/javascript" src="../_static/js/navbar.js"></script>
+ <script type="text/javascript" src="../_static/js/clipboard.min.js"></script>
+ <script type="text/javascript" src="../_static/js/copycode.js"></script>
+ <script type="text/javascript" src="../_static/js/page.js"></script>
+ <script type="text/javascript">
+ $('body').ready(function () {
+ $('body').css('visibility', 'visible');
+ });
+ </script>
+ </body>
+</html>
diff --git a/docs/build_version_doc/artifacts/1.2.0.index.html b/docs/build_version_doc/artifacts/1.2.0.index.html
new file mode 100644
index 00000000000..13792d7f1a4
--- /dev/null
+++ b/docs/build_version_doc/artifacts/1.2.0.index.html
@@ -0,0 +1,1338 @@
+<!DOCTYPE html>
+
+<html lang="en">
+ <head>
+ <meta charset="utf-8">
+ <meta http-equiv="X-UA-Compatible" content="IE=edge">
+ <meta name="viewport" content="width=device-width, initial-scale=1">
+
+
+
+ <title>Installing MXNet — mxnet documentation</title>
+
+ <link rel="stylesheet" href="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.6/css/bootstrap.min.css" integrity="sha384-1q8mTJOASx8j1Au+a5WDVnPi2lkFfwwEAa8hDDdjZlpLegxhjVME1fgjWPGmkzs7" crossorigin="anonymous">
+ <link rel="stylesheet" href="https://maxcdn.bootstrapcdn.com/font-awesome/4.5.0/css/font-awesome.min.css">
+
+
+ <link rel="stylesheet" href="../_static/basic.css" type="text/css" />
+ <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+
+ <link rel="stylesheet" href="../_static/mxnet.css" type="text/css">
+
+ <script type="text/javascript">
+ var DOCUMENTATION_OPTIONS = {
+ URL_ROOT: '../',
+ VERSION: '',
+ COLLAPSE_INDEX: false,
+ FILE_SUFFIX: '.html',
+ HAS_SOURCE: true,
+ SOURCELINK_SUFFIX: '.txt'
+ };
+ </script>
+
+ <script type="text/javascript" src="https://code.jquery.com/jquery-1.11.1.min.js"></script>
+
+ <script type="text/javascript" src="../_static/underscore.js"></script>
+
+ <script type="text/javascript" src="../_static/searchtools_custom.js"></script>
+
+ <script type="text/javascript" src="../_static/doctools.js"></script>
+
+ <script type="text/javascript" src="../_static/selectlang.js"></script>
+
+
+ <script type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.1/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></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://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.0/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script> -->
+ <!-- -->
+ <link rel="index" title="Index" href="../genindex.html" />
+ <link rel="search" title="Search" href="../search.html" />
+
+ <link rel="icon" type="image/png" href="https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/mxnet-icon.png">
+ </head>
+ <body role="document" background="https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/mxnet-background-compressed.jpeg">
+ <div class='content-block'><div class="navbar navbar-fixed-top">
+ <div class="container" id="navContainer">
+ <div id="header-inner" class="innder">
+ <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 id="main-nav" class='nav-bar'>
+ <a class="main-nav-link" href="../install/index.html">Install</a>
+
+ <span id="dropdown-menu-position-anchor">
+ <a href="#" class="main-nav-link dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">Gluon <span class="caret"></span></a>
+ <ul id="package-dropdown-menu" class="dropdown-menu navbar-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 href="#" class="main-nav-link dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">API <span class="caret"></span></a>
+ <ul id="package-dropdown-menu" class="dropdown-menu navbar-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 href="#" class="main-nav-link dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">Docs <span class="caret"></span></a>
+ <ul id="package-dropdown-menu-docs" class="dropdown-menu navbar-menu">
+ <li><a class="main-nav-link" href="../tutorials/index.html">Tutorials</a>
+ <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/master/example">Examples</a></li>
+ <li><a class="main-nav-link" href="../api/python/gluon/model_zoo.html">Gluon 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 href="#" class="main-nav-link dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">Community <span class="caret"></span></a>
+ <ul id="package-dropdown-menu-community" class="dropdown-menu navbar-menu">
+ <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>
+ <li><a class="main-nav-link" href="http://discuss.mxnet.io">Discuss</a></li>
+ </ul>
+ </span>
+ </nav>
+ <script> function getRootPath(){ return "../" } </script>
+ <div class="burgerIcon dropdown">
+ <a href="#" class="dropdown-toggle" data-toggle="dropdown" role="button">☰</a>
+ <ul id="burgerMenu" class="dropdown-menu">
+ <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 tabindex="-1" href="../community/index.html">Community</a></li>
+ <li><a tabindex="-1" href="../community/contribute.html">Contribute</a></li>
+ <li><a tabindex="-1" href="../community/powered_by.html">Powered By</a></li>
+ </ul>
+ </li>
+
+ <li class="dropdown-submenu">
+ <a href="#" tabindex="-1">API</a>
+ <ul class="dropdown-menu">
+
+ <li><a tabindex="-1" href="../api/python/index.html">Python</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/scala/index.html">Scala</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/r/index.html">R</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/julia/index.html">Julia</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/c++/index.html">C++</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/perl/index.html">Perl</a>
+ </li>
+
+ </ul>
+ </li>
+
+ <li class="dropdown-submenu">
+ <a href="#" tabindex="-1">Docs</a>
+ <ul class="dropdown-menu">
+ <li><a tabindex="-1" href="../tutorials/index.html">Tutorials</a></li>
+ <li><a tabindex="-1" href="../faq/index.html">FAQ</a></li>
+ <li><a tabindex="-1" href="../architecture/index.html">Architecture</a></li>
+ <li><a tabindex="-1" href="https://github.com/apache/incubator-mxnet/tree/master/example">Examples</a></li>
+ <li><a tabindex="-1" href="../api/python/gluon/model_zoo.html">Gluon 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>
+ </ul>
+ </div>
+ <div class="plusIcon dropdown">
+ <a href="#" class="dropdown-toggle" data-toggle="dropdown" role="button"><span class="glyphicon glyphicon-plus" aria-hidden="true"></span></a>
+ <ul id="plusMenu" class="dropdown-menu dropdown-menu-right"></ul>
+ </div>
+ <div id="search-input-wrap">
+
+<form class="" role="search" action="../search.html" method="get" autocomplete="off">
+ <div class="form-group inner-addon left-addon">
+ <i class="glyphicon glyphicon-search"></i>
+ <input type="text" name="q" class="form-control" placeholder="Search">
+ </div>
+ <input type="hidden" name="check_keywords" value="yes" />
+ <input type="hidden" name="area" value="default" />
+
+</form>
+ <div id="search-preview"></div>
+ </div>
+ <div id='searchIcon'>
+ <span class="glyphicon glyphicon-search" aria-hidden="true"></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 class="sphinxsidebar leftsidebar" role="navigation" aria-label="main 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>
+
+ <div class="section" id="installing-mxnet">
+<span id="installing-mxnet"></span><h1>Installing MXNet<a class="headerlink" href="#installing-mxnet" title="Permalink to this headline">¶</a></h1>
+<p>Indicate your preferred configuration. Then, follow the customized commands to install <em>MXNet</em>.</p>
+<script type="text/javascript" src='../_static/js/options.js'></script><!-- START - OS Menu --><div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">Linux</button>
+ <button type="button" class="btn btn-default opt">MacOS</button>
+ <button type="button" class="btn btn-default opt">Windows</button>
+ <button type="button" class="btn btn-default opt">Cloud</button>
+ <button type="button" class="btn btn-default opt">Devices</button>
+</div><!-- START - Language Menu --><div class="linux macos windows">
+<div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">Python</button>
+ <button type="button" class="btn btn-default opt">Scala</button>
+ <button type="button" class="btn btn-default opt">R</button>
+ <button type="button" class="btn btn-default opt">Julia</button>
+ <button type="button" class="btn btn-default opt">Perl</button>
+</div>
+</div><!-- No CPU GPU for other Devices -->
+<div class="linux macos windows cloud">
+<div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">CPU</button>
+ <button type="button" class="btn btn-default opt">GPU</button>
+</div>
+</div><!-- other devices -->
+<div class="devices">
+<div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">Raspberry Pi</button>
+ <button type="button" class="btn btn-default opt">NVIDIA Jetson TX2</button>
+</div>
+</div><!-- Linux Python GPU Options --><div class="linux macos windows">
+<div class="python">
+<div class="cpu gpu">
+<div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">Pip</button>
+ <button type="button" class="btn btn-default opt">Virtualenv</button>
+ <button type="button" class="btn btn-default opt">Docker</button>
+ <button type="button" class="btn btn-default opt">Build from Source</button>
+</div>
+</div>
+</div>
+</div><!-- END - Main Menu --><!-- START - Linux Python CPU Installation Instructions --><div class="linux">
+ <div class="python">
+ <div class="cpu"><p>The following installation instructions have been tested on Ubuntu 14.04 and 16.04.</p>
+<div class="virtualenv">
+<br/><p><strong>Step 1</strong> Install virtualenv for Ubuntu.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y python-dev python-virtualenv
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Create and activate virtualenv environment for MXNet.</p>
+<p>Following command creates a virtualenv environment at <code class="docutils literal"><span class="pre">~/mxnet</span></code> directory. However, you can choose any directory by replacing <code class="docutils literal"><span class="pre">~/mxnet</span></code> with a directory of your choice.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ virtualenv --system-site-packages ~/mxnet
+</pre></div>
+</div>
+<p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install MXNet in the active virtualenv environment.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install --upgrade pip
+</pre></div>
+</div>
+<p>Install <em>MXNet</em> with OpenBLAS acceleration.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet==1.2.0
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 5</strong> Validate the installation by running simple <em>MXNet</em> code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Note</strong> You can read more about virtualenv <a class="reference external" href="https://virtualenv.pypa.io/en/stable/userguide/">here</a>.</p>
+</div><div class="pip">
+<br/><p><strong>Step 1</strong> Install prerequisites - wget and latest pip.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command in the terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y wget python gcc
+$ wget https://bootstrap.pypa.io/get-pip.py <span class="o">&&</span> sudo python get-pip.py
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install MXNet with OpenBLAS acceleration.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet==1.2.0
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Experimental Choice</strong> If You would like to install mxnet with Intel MKL, try the experimental pip package with MKL:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet-mkl==1.2.0
+</pre></div>
+</div>
+</div><div class="docker">
+<br/><p>Docker images with <em>MXNet</em> are available at <a class="reference external" href="https://hub.docker.com/r/mxnet/">Docker Hub</a>.</p>
+<p><strong>Step 1</strong> Install Docker on your machine by following the <a class="reference external" href="https://docs.docker.com/engine/installation/linux/ubuntu/#install-using-the-repository">docker installation instructions</a>.</p>
+<p><em>Note</em> - You can install Community Edition (CE) to get started with <em>MXNet</em>.</p>
+<p><strong>Step 2</strong> [Optional] Post installation steps to manage Docker as a non-root user.</p>
+<p>Follow the four steps in this <a class="reference external" href="https://docs.docker.com/engine/installation/linux/linux-postinstall/#manage-docker-as-a-non-root-user">docker documentation</a> to allow managing docker containers without <em>sudo</em>.</p>
+<p>If you skip this step, you need to use <em>sudo</em> each time you invoke Docker.</p>
+<p><strong>Step 2</strong> Pull the MXNet docker image.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker pull mxnet/python <span class="c1"># Use sudo if you skip Step 2</span>
+</pre></div>
+</div>
+<p>You can list docker images to see if mxnet/python docker image pull was successful.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker images <span class="c1"># Use sudo if you skip Step 2</span>
+
+REPOSITORY TAG IMAGE ID CREATED SIZE
+mxnet/python latest 00d026968b3c <span class="m">3</span> weeks ago <span class="m">1</span>.41 GB
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div><div class="build-from-source">
+<br/><p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings. Example - Python bindings, Scala bindings.</li>
+</ol>
+<p><strong>Minimum Requirements</strong></p>
+<ol class="simple">
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">GCC 4.8</a> or later to compile C++ 11.</li>
+<li><a class="reference external" href="https://www.gnu.org/software/make/">GNU Make</a></li>
+</ol>
+<p><br/></p>
+<p><strong>Build the MXNet core shared library</strong></p>
+<p><strong>Step 1</strong> Install build tools and git.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y build-essential git
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install OpenBLAS.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="https://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprograms">BLAS</a> and <a class="reference external" href="https://en.wikipedia.org/wiki/LAPACK">LAPACK</a> libraries for accelerated numerical computations on CPU machine. There are several flavors of BLAS/LAPACK libraries - <a class="reference external" href="http://www.openblas.net/">OpenBLAS</a>, <a class="reference external" href="http://math-atlas.sourceforge.net/">ATLAS</a> and <a class="reference external" href="https://software.intel.com/en-us/intel-mkl">MKL</a>. In this step we install OpenBLAS. You can choose to install ATLAS or MKL.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopenblas-dev liblapack-dev
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install OpenCV.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="http://opencv.org/">OpenCV</a> for efficient image loading and augmentation operations.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopencv-dev
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Download MXNet sources and build MXNet core shared library. You can clone the repository as described in the following code block, or you may try the <a href="download.html">download links</a> for your desired MXNet version.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ git clone --recursive https://github.com/apache/incubator-mxnet
+$ <span class="nb">cd</span> incubator-mxnet
+$ make -j <span class="k">$(</span>nproc<span class="k">)</span> <span class="nv">USE_OPENCV</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_BLAS</span><span class="o">=</span>openblas
+</pre></div>
+</div>
+<p><em>Note</em> - USE_OPENCV and USE_BLAS are make file flags to set compilation options to use OpenCV and BLAS library. You can explore and use more compilation options in <code class="docutils literal"><span class="pre">make/config.mk</span></code>.</p>
+<p><br/></p>
+<p><strong>Build the MXNet Python binding</strong></p>
+<p><strong>Step 1</strong> Install prerequisites - python, setup-tools, python-pip and libfortran (required for Numpy).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y python-dev python-setuptools python-pip libgfortran3
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install the MXNet Python binding.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">cd</span> python
+$ pip install --upgrade pip
+$ pip install -e .
+</pre></div>
+</div>
+<p>Note that the <code class="docutils literal"><span class="pre">-e</span></code> flag is optional. It is equivalent to <code class="docutils literal"><span class="pre">--editable</span></code> and means that if you edit the source files, these changes will be reflected in the package installed.</p>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div></div>
+</div>
+</div>
+<!-- END - Linux Python CPU Installation Instructions --><!-- START - Linux Python GPU Installation Instructions --><div class="linux">
+<div class="python">
+<div class="gpu"><p>The following installation instructions have been tested on Ubuntu 14.04 and 16.04.</p>
+<p><strong>Prerequisites</strong></p>
+<p>Install the following NVIDIA libraries to setup <em>MXNet</em> with GPU support:</p>
+<ol class="simple">
+<li>Install CUDA 9.0 following the NVIDIA’s <a class="reference external" href="http://docs.nvidia.com/cuda/cuda-installation-guide-linux/">installation guide</a>.</li>
+<li>Install cuDNN 7 for CUDA 9.0 following the NVIDIA’s <a class="reference external" href="https://developer.nvidia.com/cudnn">installation guide</a>. You may need to register with NVIDIA for downloading the cuDNN library.</li>
+</ol>
+<p><strong>Note:</strong> Make sure to add CUDA install path to <code class="docutils literal"><span class="pre">LD_LIBRARY_PATH</span></code>.</p>
+<p>Example - <em>export LD_LIBRARY_PATH=/usr/local/cuda/lib64/:$LD_LIBRARY_PATH</em></p>
+<div class="pip">
+<br/><p><strong>Step 1</strong> Install prerequisites - wget and latest pip.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command in the terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y wget python
+$ wget https://bootstrap.pypa.io/get-pip.py <span class="o">&&</span> sudo python get-pip.py
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install <em>MXNet</em> with GPU support using CUDA 9.0</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet-cu90==1.2.0
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Experimental Choice</strong> If You would like to install mxnet with Intel MKL, try the experimental pip package with MKL:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet-cu90mkl==1.2.0
+</pre></div>
+</div>
+</div><div class="virtualenv"><p><br/></p>
+<p><strong>Step 1</strong> Install virtualenv for Ubuntu.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y python-dev python-virtualenv
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Create and activate virtualenv environment for MXNet.</p>
+<p>Following command creates a virtualenv environment at <code class="docutils literal"><span class="pre">~/mxnet</span></code> directory. However, you can choose any directory by replacing <code class="docutils literal"><span class="pre">~/mxnet</span></code> with a directory of your choice.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ virtualenv --system-site-packages ~/mxnet
+</pre></div>
+</div>
+<p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install MXNet in the active virtualenv environment.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install --upgrade pip
+</pre></div>
+</div>
+<p>Install <em>MXNet</em> with GPU support using CUDA 9.0.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet-cu90==1.2.0
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 5</strong> Validate the installation by running simple <em>MXNet</em> code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Note</strong> You can read more about virtualenv <a class="reference external" href="https://virtualenv.pypa.io/en/stable/userguide/">here</a>.</p>
+</div><div class="docker"><p><br/></p>
+<p>Docker images with <em>MXNet</em> are available at <a class="reference external" href="https://hub.docker.com/r/mxnet/">Docker Hub</a>.</p>
+<p><strong>Step 1</strong> Install Docker on your machine by following the <a class="reference external" href="https://docs.docker.com/engine/installation/linux/ubuntu/#install-using-the-repository">docker installation instructions</a>.</p>
+<p><em>Note</em> - You can install Community Edition (CE) to get started with <em>MXNet</em>.</p>
+<p><strong>Step 2</strong> [Optional] Post installation steps to manage Docker as a non-root user.</p>
+<p>Follow the four steps in this <a class="reference external" href="https://docs.docker.com/engine/installation/linux/linux-postinstall/#manage-docker-as-a-non-root-user">docker documentation</a> to allow managing docker containers without <em>sudo</em>.</p>
+<p>If you skip this step, you need to use <em>sudo</em> each time you invoke Docker.</p>
+<p><strong>Step 3</strong> Install <em>nvidia-docker-plugin</em> following the <a class="reference external" href="https://github.com/NVIDIA/nvidia-docker/wiki/Installation">installation instructions</a>. <em>nvidia-docker-plugin</em> is required to enable the usage of GPUs from the docker containers.</p>
+<p><strong>Step 4</strong> Pull the MXNet docker image.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker pull mxnet/python:gpu <span class="c1"># Use sudo if you skip Step 2</span>
+</pre></div>
+</div>
+<p>You can list docker images to see if mxnet/python docker image pull was successful.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker images <span class="c1"># Use sudo if you skip Step 2</span>
+
+REPOSITORY TAG IMAGE ID CREATED SIZE
+mxnet/python gpu 493b2683c269 <span class="m">3</span> weeks ago <span class="m">4</span>.77 GB
+</pre></div>
+</div>
+<p><strong>Step 5</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div><div class="build-from-source"><p><br/></p>
+<p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings. Example - Python bindings, Scala bindings.</li>
+</ol>
+<p><strong>Minimum Requirements</strong></p>
+<ol class="simple">
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">GCC 4.8</a> or later to compile C++ 11.</li>
+<li><a class="reference external" href="https://www.gnu.org/software/make/">GNU Make</a></li>
+</ol>
+<p><br/></p>
+<p><strong>Build the MXNet core shared library</strong></p>
+<p><strong>Step 1</strong> Install build tools and git.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y build-essential git
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install OpenBLAS.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="https://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprograms">BLAS</a> and <a class="reference external" href="https://en.wikipedia.org/wiki/LAPACK">LAPACK</a> libraries for accelerated numerical computations on CPU machine. There are several flavors of BLAS/LAPACK libraries - <a class="reference external" href="http://www.openblas.net/">OpenBLAS</a>, <a class="reference external" href="http://math-atlas.sourceforge.net/">ATLAS</a> and <a class="reference external" href="https://software.intel.com/en-us/intel-mkl">MKL</a>. In this step we install OpenBLAS. You can choose to install ATLAS or MKL.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopenblas-dev liblapack-dev
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install OpenCV.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="http://opencv.org/">OpenCV</a> for efficient image loading and augmentation operations.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopencv-dev
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Download MXNet sources and build MXNet core shared library. You can clone the repository as described in the following code block, or you may try the <a href="download.html">download links</a> for your desired MXNet version.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ git clone --recursive https://github.com/apache/incubator-mxnet
+$ <span class="nb">cd</span> incubator-mxnet
+$ make -j <span class="k">$(</span>nproc<span class="k">)</span> <span class="nv">USE_OPENCV</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_BLAS</span><span class="o">=</span>openblas <span class="nv">USE_CUDA</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_CUDA_PATH</span><span class="o">=</span>/usr/local/cuda <span class="nv">USE_CUDNN</span><span class="o">=</span><span class="m">1</span>
+</pre></div>
+</div>
+<p><em>Note</em> - USE_OPENCV, USE_BLAS, USE_CUDA, USE_CUDA_PATH AND USE_CUDNN are make file flags to set compilation options to use OpenCV, OpenBLAS, CUDA and cuDNN libraries. You can explore and use more compilation options in <code class="docutils literal"><span class="pre">make/config.mk</span></code>. Make sure to set USE_CUDA_PATH to right CUDA installation path. In most cases it is - <em>/usr/local/cuda</em>.</p>
+<p><br/></p>
+<p><strong>Install the MXNet Python binding</strong></p>
+<p><strong>Step 1</strong> Install prerequisites - python, setup-tools, python-pip and libfortran (required for Numpy)..</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y python-dev python-setuptools python-pip libgfortran3
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install the MXNet Python binding.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">cd</span> python
+$ pip install --upgrade pip
+$ pip install -e .
+</pre></div>
+</div>
+<p>Note that the <code class="docutils literal"><span class="pre">-e</span></code> flag is optional. It is equivalent to <code class="docutils literal"><span class="pre">--editable</span></code> and means that if you edit the source files, these changes will be reflected in the package installed.</p>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div></div>
+</div>
+</div>
+<!-- END - Linux Python GPU Installation Instructions --><!-- START - MacOS Python CPU Installation Instructions --><div class="macos">
+ <div class="python">
+ <div class="cpu"><p>The following installation instructions have been tested on OSX Sierra and El Capitan.</p>
+<div class="virtualenv">
+<br/><p><strong>Step 1</strong> Install prerequisites - Homebrew, python development tools.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="c1"># Install Homebrew</span>
+$ /usr/bin/ruby -e <span class="s2">"</span><span class="k">$(</span>curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install<span class="k">)</span><span class="s2">"</span>
+$ <span class="nb">export</span> <span class="nv">PATH</span><span class="o">=</span>/usr/local/bin:/usr/local/sbin:<span class="nv">$PATH</span>
+
+<span class="c1"># Install python development tools - python2.7, pip, python-setuptools</span>
+$ brew install python
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install virtualenv for macOS.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install virtualenv
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Create and activate virtualenv environment for MXNet.</p>
+<p>Following command creates a virtualenv environment at <code class="docutils literal"><span class="pre">~/mxnet</span></code> directory. However, you can choose any directory by replacing <code class="docutils literal"><span class="pre">~/mxnet</span></code> with a directory of your choice.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ virtualenv --system-site-packages ~/mxnet
+</pre></div>
+</div>
+<p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Install MXNet in the active virtualenv environment.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install --upgrade pip
+<span class="o">(</span>mxnet<span class="o">)</span>$ pip install --upgrade setuptools
+</pre></div>
+</div>
+<p>Install <em>MXNet</em> with OpenBLAS acceleration.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet==1.2.0
+</pre></div>
+</div>
+<p><strong>Step 5</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ brew install graphviz
+<span class="o">(</span>mxnet<span class="o">)</span>$ pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 6</strong> Validate the installation by running simple <em>MXNet</em> code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Note</strong> You can read more about virtualenv <a class="reference external" href="https://virtualenv.pypa.io/en/stable/userguide/">here</a>.</p>
+</div><div class="pip">
+<br/><p><strong>Step 1</strong> Install prerequisites - Homebrew, python development tools.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="c1"># Install Homebrew</span>
+$ /usr/bin/ruby -e <span class="s2">"</span><span class="k">$(</span>curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install<span class="k">)</span><span class="s2">"</span>
+$ <span class="nb">export</span> <span class="nv">PATH</span><span class="o">=</span>/usr/local/bin:/usr/local/sbin:<span class="nv">$PATH</span>
+
+<span class="c1"># Install python development tools - python2.7, pip, python-setuptools</span>
+$ brew install python
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install MXNet with OpenBLAS acceleration.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install --upgrade pip
+$ pip install --upgrade setuptools
+</pre></div>
+</div>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet==1.2.0
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ brew install graphviz
+$ pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div><div class="docker">
+<br/><p>Docker images with <em>MXNet</em> are available at <a class="reference external" href="https://hub.docker.com/r/mxnet/">Docker Hub</a>.</p>
+<p><strong>Step 1</strong> Install Docker on your machine by following the <a class="reference external" href="https://docs.docker.com/docker-for-mac/install/#install-and-run-docker-for-mac">docker installation instructions</a>.</p>
+<p><em>Note</em> - You can install Community Edition (CE) to get started with <em>MXNet</em>.</p>
+<p><strong>Step 2</strong> Pull the MXNet docker image.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker pull mxnet/python
+</pre></div>
+</div>
+<p>You can list docker images to see if mxnet/python docker image pull was successful.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker images
+
+REPOSITORY TAG IMAGE ID CREATED SIZE
+mxnet/python latest 00d026968b3c <span class="m">3</span> weeks ago <span class="m">1</span>.41 GB
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div><div class="build-from-source">
+<br/><p><strong>Prerequisites</strong></p>
+<p>If not already installed, <a class="reference external" href="https://developer.apple.com/xcode/">download and install Xcode</a> (or <a class="reference external" href="https://itunes.apple.com/us/app/xcode/id497799835">insall it from the App Store</a>) for macOS. <a class="reference external" href="https://en.wikipedia.org/wiki/Xcode">Xcode</a> is an integrated development environment for macOS containing a suite of software development tools like C/C++ compilers, BLAS library and more.</p>
+<p><br/></p>
+<p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings. Example - Python bindings, Scala bindings.</li>
+</ol>
+<p>Make sure you have installed Xcode before proceeding further.</p>
+<p><br/></p>
+<p>All the instructions to build <em>MXNet</em> core shared library and <em>MXNet</em> Python bindings are compiled as one helper <em>bash</em> script. You can use <a class="reference external" href="https://raw.githubusercontent.com/dmlc/mxnet/master/setup-utils/install-mxnet-osx-python.sh">this bash script</a> to build <em>MXNet</em> for Python, from source, on macOS.</p>
+<p><strong>Step 1</strong> Download the bash script for building MXNet from source.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ curl -O https://raw.githubusercontent.com/dmlc/mxnet/master/setup-utils/install-mxnet-osx-python.sh
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Run the script to get latest MXNet source and build.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="c1"># Make the script executable</span>
+$ chmod <span class="m">744</span> install-mxnet-osx-python.sh
+
+<span class="c1"># Run the script. It takes around 5 mins.</span>
+$ bash install-mxnet-osx-python.sh
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div> <!-- End of source build --></div>
+</div>
+</div><!-- END - Mac OS Python CPU Installation Instructions --><!-- START - Mac OS Python GPU Installation Instructions --><div class="macos">
+ <div class="python">
+ <div class="gpu"><div class="pip virtualenv docker">
+</br><p>Try the <strong>Build from Source</strong> option for now.</p>
+</div><div class="build-from-source"><p><strong>Step 1</strong> Install prerequisites - Homebrew, python development tools.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="c1"># Install Homebrew</span>
+$ /usr/bin/ruby -e <span class="s2">"</span><span class="k">$(</span>curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install<span class="k">)</span><span class="s2">"</span>
+$ <span class="nb">export</span> <span class="nv">PATH</span><span class="o">=</span>/usr/local/bin:/usr/local/sbin:<span class="nv">$PATH</span>
+
+<span class="c1"># Install python development tools - python2.7, pip, python-setuptools</span>
+$ brew install python pkg-config graphviz
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install optional components - OpenCV</p>
+<p>If you want to use OpenCV you should install it first, then build MXNet with the <code class="docutils literal"><span class="pre">USE_OPENCV=1</span></code> option in the later steps.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>brew tap homebrew/science
+brew install opencv
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install CUDA and cuDNN</p>
+<p>The following instructions are for CUDA 9.1 and cuDNN 7 for macOS 10.12+ and a CUDA-capable GPU. They summarize confirmed successful builds in <a class="reference external" href="https://github.com/apache/incubator-mxnet/issues/9217">#9217</a>.
+Alternatively, you may follow the <a class="reference external" href="https://docs.nvidia.com/cuda/cuda-installation-guide-mac-os-x/index.html">CUDA installation instructions for macOS</a>.</p>
+<ol class="simple">
+<li><a class="reference external" href="https://developer.apple.com/download/more/">Download Xcode 8.3.3 from Apple</a>. This is the version <a class="reference external" href="https://docs.nvidia.com/cuda/cuda-installation-guide-mac-os-x/index.html">NVIDIA specifies in its instructions for macOS</a>. Unzip and rename to <code class="docutils literal"><span class="pre">Xcode8.3.3.app</span></code>.</li>
+<li>Run <code class="docutils literal"><span class="pre">sudo</span> <span class="pre">xcode-select</span> <span class="pre">-s</span> <span class="pre">/Applications/Xcode8.3.3.app</span></code> or to wherever you have placed Xcode.</li>
+<li>Run <code class="docutils literal"><span class="pre">xcode-select</span> <span class="pre">--install</span></code> to install all command line tools, compilers, etc.</li>
+<li>Run <code class="docutils literal"><span class="pre">sudo</span> <span class="pre">xcodebuild</span> <span class="pre">-license</span> <span class="pre">accept</span></code> to accept Xcode’s licensing terms.</li>
+<li>Install CUDA for macOS. Specific steps are provided in NVIDIA’s <a class="reference external" href="https://docs.nvidia.com/cuda/cuda-installation-guide-mac-os-x/index.html#installation">CUDA installation instructions</a>.</li>
+<li><a class="reference external" href="http://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html#download-mac">Download</a> and <a class="reference external" href="http://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html#installmac">install</a> cuDNN for macOS. You will need to <a class="reference external" href="https://developer.nvidia.com/accelerated-computing-developer">create a free developer account</a> with NVIDIA prior to getting the download link.</li>
+</ol>
+<p><strong>Step 4</strong> Build MXNet</p>
+<ol>
+<li><p class="first">Run <code class="docutils literal"><span class="pre">git</span> <span class="pre">clone</span> <span class="pre">--recursive</span> <span class="pre">https://github.com/apache/incubator-mxnet.git</span> <span class="pre">mxnet</span></code> to get the latest version.</p>
+</li>
+<li><p class="first">Run <code class="docutils literal"><span class="pre">cd</span> <span class="pre">mxnet</span></code>.</p>
+</li>
+<li><p class="first">Edit the <code class="docutils literal"><span class="pre">make/osx.mk</span></code> file to set the following parameters:</p>
+<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">USE_CUDA</span> <span class="o">=</span> <span class="mi">1</span>
+<span class="n">USE_CUDA_PATH</span> <span class="o">=</span> <span class="o">/</span><span class="n">usr</span><span class="o">/</span><span class="n">local</span><span class="o">/</span><span class="n">cuda</span>
+<span class="n">USE_CUDNN</span> <span class="o">=</span> <span class="mi">1</span>
+<span class="n">USE_OPENCV</span> <span class="o">=</span> <span class="mi">0</span> <span class="c1"># set to 1 if you want to build with OpenCV</span>
+</pre></div>
+</div>
+</li>
+<li><p class="first">Copy the <code class="docutils literal"><span class="pre">make/osx.mk</span></code> to <code class="docutils literal"><span class="pre">config.mk</span></code></p>
+</li>
+<li><p class="first">Run <code class="docutils literal"><span class="pre">make</span></code>. If you previously attempted to compile you might want to do <code class="docutils literal"><span class="pre">make</span> <span class="pre">clean_all</span></code> first. You can also run <code class="docutils literal"><span class="pre">make</span> <span class="pre">-j</span></code> with the number of processors you have to compile with multithreading. There’ll be plenty of warnings, but there should be no errors.</p>
+</li>
+<li><p class="first">Once finished, you should have a file called <code class="docutils literal"><span class="pre">libmxnet.so</span></code> in <code class="docutils literal"><span class="pre">lib/</span></code>.</p>
+</li>
+<li><p class="first">Do <code class="docutils literal"><span class="pre">cd</span> <span class="pre">python</span></code>.</p>
+</li>
+<li><p class="first">Run <code class="docutils literal"><span class="pre">sudo</span> <span class="pre">pip</span> <span class="pre">install</span> <span class="pre">-e</span> <span class="pre">.</span></code> <strong>Note</strong>: the <code class="docutils literal"><span class="pre">.</span></code> is part of the command.</p>
+</li>
+</ol>
+</div>
+</div>
+</div>
+</div><!-- END - Mac OS Python GPU Installation Instructions --><!-- START - Cloud Python Installation Instructions --><div class="cloud"><p>AWS Marketplace distributes Deep Learning AMIs (Amazon Machine Image) with MXNet pre-installed. You can launch one of these Deep Learning AMIs by following instructions in the <a class="reference external" href="http://docs.aws.amazon.com/dlami/latest/devguide/what-is-dlami.html">AWS Deep Learning AMI Developer Guide</a>.</p>
+<p>You can also run distributed deep learning with <em>MXNet</em> on AWS using <a class="reference external" href="https://github.com/awslabs/deeplearning-cfn/blob/master/README.md">Cloudformation Template</a>.</p>
+</div><!-- END - Cloud Python Installation Instructions --><!-- START - MacOS R CPU Installation Instructions --><div class="macos">
+ <div class="r">
+ <div class="cpu"><p>The CPU version of MXNet R package can be installed in R like other packages</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span>cran <span class="o"><-</span> <span class="kp">getOption</span><span class="p">(</span><span class="s">"repos"</span><span class="p">)</span>
+cran<span class="p">[</span><span class="s">"dmlc"</span><span class="p">]</span> <span class="o"><-</span> <span class="s">"https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/R/CRAN/"</span>
+<span class="kp">options</span><span class="p">(</span>repos <span class="o">=</span> cran<span class="p">)</span>
+install.packages<span class="p">(</span><span class="s">"mxnet"</span><span class="p">)</span>
+</pre></div>
+</div>
+</div><div class="gpu"><p>Will be available soon.</p>
+</div></div>
+</div>
+<!-- END - MacOS R CPU Installation Instructions --><div class="linux">
+ <div class="r">
+ <div class="cpu">
+<br/><p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings.</li>
+</ol>
+<p><strong>Minimum Requirements</strong></p>
+<ol class="simple">
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">GCC 4.8</a> or later to compile C++ 11.</li>
+<li><a class="reference external" href="https://www.gnu.org/software/make/">GNU Make</a></li>
+</ol>
+<p><br/></p>
+<p><strong>Build the MXNet core shared library</strong></p>
+<p><strong>Step 1</strong> Install build tools and git.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y build-essential git
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install OpenBLAS.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="https://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprograms">BLAS</a> and <a class="reference external" href="https://en.wikipedia.org/wiki/LAPACK">LAPACK</a> libraries for accelerated numerical computations on CPU machine. There are several flavors of BLAS/LAPACK libraries - <a class="reference external" href="http://www.openblas.net/">OpenBLAS</a>, <a class="reference external" href="http://math-atlas.sourceforge.net/">ATLAS</a> and <a class="reference external" href="https://software.intel.com/en-us/intel-mkl">MKL</a>. In this step we install OpenBLAS. You can choose to install ATLAS or MKL.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopenblas-dev liblapack-dev
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install OpenCV.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="http://opencv.org/">OpenCV</a> for efficient image loading and augmentation operations.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopencv-dev
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Download MXNet sources and build MXNet core shared library. You can clone the repository as described in the following code block, or you may try the <a href="download.html">download links</a> for your desired MXNet version.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ git clone --recursive https://github.com/apache/incubator-mxnet
+$ <span class="nb">cd</span> incubator-mxnet
+$ make -j <span class="k">$(</span>nproc<span class="k">)</span> <span class="nv">USE_OPENCV</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_BLAS</span><span class="o">=</span>openblas
+</pre></div>
+</div>
+<p><em>Note</em> - USE_OPENCV and USE_BLAS are make file flags to set compilation options to use OpenCV and BLAS library. You can explore and use more compilation options in <code class="docutils literal"><span class="pre">make/config.mk</span></code>.</p>
+<p><br/></p>
+<p><strong>Build and install the MXNet R binding</strong></p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ make rpkg
+$ R CMD INSTALL mxnet_current_r.tar.gz
+</pre></div>
+</div>
+</div><div class="gpu"><p>The following installation instructions have been tested on Ubuntu 14.04 and 16.04.</p>
+<p><strong>Prerequisites</strong></p>
+<p>Install the following NVIDIA libraries to setup <em>MXNet</em> with GPU support:</p>
+<ol class="simple">
+<li>Install CUDA 9.0 following the NVIDIA’s <a class="reference external" href="http://docs.nvidia.com/cuda/cuda-installation-guide-linux/">installation guide</a>.</li>
+<li>Install cuDNN 7 for CUDA 9.0 following the NVIDIA’s <a class="reference external" href="https://developer.nvidia.com/cudnn">installation guide</a>. You may need to register with NVIDIA for downloading the cuDNN library.</li>
+</ol>
+<p><strong>Note:</strong> Make sure to add CUDA install path to <code class="docutils literal"><span class="pre">LD_LIBRARY_PATH</span></code>.</p>
+<p>Example - <em>export LD_LIBRARY_PATH=/usr/local/cuda/lib64/:$LD_LIBRARY_PATH</em></p>
+<p><br/></p>
+<p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings.</li>
+</ol>
+<p><strong>Minimum Requirements</strong></p>
+<ol class="simple">
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">GCC 4.8</a> or later to compile C++ 11.</li>
+<li><a class="reference external" href="https://www.gnu.org/software/make/">GNU Make</a></li>
+</ol>
+<p><br/></p>
+<p><strong>Build the MXNet core shared library</strong></p>
+<p><strong>Step 1</strong> Install build tools and git.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y build-essential git
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install OpenBLAS.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="https://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprograms">BLAS</a> and <a class="reference external" href="https://en.wikipedia.org/wiki/LAPACK">LAPACK</a> libraries for accelerated numerical computations on CPU machine. There are several flavors of BLAS/LAPACK libraries - <a class="reference external" href="http://www.openblas.net/">OpenBLAS</a>, <a class="reference external" href="http://math-atlas.sourceforge.net/">ATLAS</a> and <a class="reference external" href="https://software.intel.com/en-us/intel-mkl">MKL</a>. In this step we install OpenBLAS. You can choose to install ATLAS or MKL.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopenblas-dev liblapack-dev
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install OpenCV.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="http://opencv.org/">OpenCV</a> for efficient image loading and augmentation operations.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopencv-dev
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Download MXNet sources and build MXNet core shared library. You can clone the repository as described in the following code block, or you may try the <a href="download.html">download links</a> for your desired MXNet version.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ git clone --recursive https://github.com/apache/incubator-mxnet
+$ <span class="nb">cd</span> incubator-mxnet
+$ make -j <span class="k">$(</span>nproc<span class="k">)</span> <span class="nv">USE_OPENCV</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_BLAS</span><span class="o">=</span>openblas <span class="nv">USE_CUDA</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_CUDA_PATH</span><span class="o">=</span>/usr/local/cuda <span class="nv">USE_CUDNN</span><span class="o">=</span><span class="m">1</span>
+</pre></div>
+</div>
+<p><em>Note</em> - USE_OPENCV, USE_BLAS, USE_CUDA, USE_CUDA_PATH AND USE_CUDNN are make file flags to set compilation options to use OpenCV, OpenBLAS, CUDA and cuDNN libraries. You can explore and use more compilation options in <code class="docutils literal"><span class="pre">make/config.mk</span></code>. Make sure to set USE_CUDA_PATH to right CUDA installation path. In most cases it is - <em>/usr/local/cuda</em>.</p>
+<p><br/></p>
+<p><strong>Build and install the MXNet R binding</strong></p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ make rpkg
+$ R CMD INSTALL mxnet_current_r.tar.gz
+</pre></div>
+</div>
+</div></div>
+</div><!-- START - Windows R CPU Installation Instructions --><div class="windows">
+ <div class="r">
+ <div class="cpu"><p>The CPU version of MXNet R package can be installed in R like other packages</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span>cran <span class="o"><-</span> <span class="kp">getOption</span><span class="p">(</span><span class="s">"repos"</span><span class="p">)</span>
+cran<span class="p">[</span><span class="s">"dmlc"</span><span class="p">]</span> <span class="o"><-</span> <span class="s">"https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/R/CRAN/"</span>
+<span class="kp">options</span><span class="p">(</span>repos <span class="o">=</span> cran<span class="p">)</span>
+install.packages<span class="p">(</span><span class="s">"mxnet"</span><span class="p">)</span>
+</pre></div>
+</div>
+</div><!-- END - Windows R CPU Installation Instructions --><div class="gpu"><p>The GPU version of MXNet R package can be installed in R like other packages</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span>cran <span class="o"><-</span> <span class="kp">getOption</span><span class="p">(</span><span class="s">"repos"</span><span class="p">)</span>
+cran<span class="p">[</span><span class="s">"dmlc"</span><span class="p">]</span> <span class="o"><-</span> <span class="s">"https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/R/CRAN/GPU"</span>
+<span class="kp">options</span><span class="p">(</span>repos <span class="o">=</span> cran<span class="p">)</span>
+install.packages<span class="p">(</span><span class="s">"mxnet"</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>Alternatively, You can also follow the installation instructions <a class="reference internal" href="windows_setup.html"><span class="doc">in this guide</span></a> to build MXNet from source.</p>
+</div>
+</div>
+</div><!-- END - Windows R GPU Installation Instructions --><div class="linux">
+ <div class="scala julia perl">
+ <div class="cpu gpu"><p>Follow the installation instructions <a class="reference internal" href="ubuntu_setup.html"><span class="doc">in this guide</span></a> to set up MXNet.</p>
+</div>
+</div>
+</div><div class="macos">
+ <div class="scala julia perl">
+ <div class="cpu gpu"><p>Follow the installation instructions <a class="reference internal" href="osx_setup.html"><span class="doc">in this guide</span></a> to set up MXNet.</p>
+</div>
+</div>
+</div><div class="windows">
+ <div class="python">
+ <div class="cpu">
+ <div class="pip">
+<br/><p><strong>Step 1</strong> Install Python.</p>
+<p><a class="reference external" href="https://www.anaconda.com/download/">Anaconda</a> is recommended.</p>
+<p><strong>Step 2</strong> Install <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet==1.2.0
+</pre></div>
+</div>
+</div>
+</div> <div class="gpu"><div class="pip">
+<br/><p><strong>Step 1</strong> Install Python.</p>
+<p><a class="reference external" href="https://www.anaconda.com/download/">Anaconda</a> is recommended.</p>
+<p><strong>Step 2</strong> Install <em>MXNet</em> with GPU support using CUDA 9.0.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet-cu90==1.2.0
+</pre></div>
+</div>
+<p>Refer to <a class="reference external" href="https://github.com/apache/incubator-mxnet/issues/8671">#8671</a> for status on CUDA 9.1 support.</p>
+</div>
+<div class="build-from-source">
+<br/>
+To build and install MXNet yourself, you need the following dependencies. Install the required dependencies:<ol class="simple">
+<li>If <a class="reference external" href="https://www.visualstudio.com/downloads/">Microsoft Visual Studio 2015</a> is not already installed, download and install it. You can download and install the free community edition.</li>
+<li>Download and install <a class="reference external" href="https://cmake.org/">CMake</a> if it is not already installed.</li>
+<li>Download and install <a class="reference external" href="http://sourceforge.net/projects/opencvlibrary/files/opencv-win/3.0.0/opencv-3.0.0.exe/download">OpenCV</a>.</li>
+<li>Unzip the OpenCV package.</li>
+<li>Set the environment variable <code class="docutils literal"><span class="pre">OpenCV_DIR</span></code> to point to the <code class="docutils literal"><span class="pre">OpenCV</span> <span class="pre">build</span> <span class="pre">directory</span></code>.</li>
+<li>If you don’t have the Intel Math Kernel Library (MKL) installed, download and install <a class="reference external" href="http://sourceforge.net/projects/openblas/files/v0.2.14/">OpenBlas</a>.</li>
+<li>Set the environment variable <code class="docutils literal"><span class="pre">OpenBLAS_HOME</span></code> to point to the <code class="docutils literal"><span class="pre">OpenBLAS</span></code> directory that contains the <code class="docutils literal"><span class="pre">include</span></code> and <code class="docutils literal"><span class="pre">lib</span></code> directories. Typically, you can find the directory in <code class="docutils literal"><span class="pre">C:\Program</span> <span class="pre">files</span> <span class="pre">(x86)\OpenBLAS\</span></code>.</li>
+<li>Download and install <a class="reference external" href="https://developer.nvidia.com/cuda-downloads?target_os=Windows&target_arch=x86_64">CUDA</a> and <a class="reference external" href="https://developer.nvidia.com/cudnn">cuDNN</a>. To get access to the download link, register as an NVIDIA community user.</li>
+</ol>
+<p>After you have installed all of the required dependencies, build the MXNet source code:</p>
+<ol class="simple">
+<li>Download the MXNet source code from <a class="reference external" href="https://github.com/apache/incubator-mxnet">GitHub</a>.</li>
+<li>Use <a class="reference external" href="https://cmake.org/">CMake</a> to create a Visual Studio solution in <code class="docutils literal"><span class="pre">./build</span></code>.</li>
+<li>In Visual Studio, open the solution file,<code class="docutils literal"><span class="pre">.sln</span></code>, and compile it.
+These commands produce a library called <code class="docutils literal"><span class="pre">mxnet.dll</span></code> in the <code class="docutils literal"><span class="pre">./build/Release/</span></code> or <code class="docutils literal"><span class="pre">./build/Debug</span></code> folder.</li>
+</ol>
+<p><p> </p>
+
+Next, we install the <code class="docutils literal"><span class="pre">graphviz</span></code> library that we use for visualizing network graphs that you build on MXNet. We will also install <a class="reference external" href="http://jupyter.readthedocs.io/">Jupyter Notebook</a> which is used for running MXNet tutorials and examples.</p>
+<ul class="simple">
+<li>Install the <code class="docutils literal"><span class="pre">graphviz</span></code> by downloading the installer from the <a class="reference external" href="https://graphviz.gitlab.io/_pages/Download/Download_windows.html">Graphviz Download Page</a>.
+<strong>Note</strong> Make sure to add the <code class="docutils literal"><span class="pre">graphviz</span></code> executable path to the PATH environment variable. Refer <a class="reference external" href="http://stackoverflow.com/questions/35064304/runtimeerror-make-sure-the-graphviz-executables-are-on-your-systems-path-aft">here for more details</a></li>
+</ul>
+<p><p> </p>
+</p>
+</div>
+</div>
+</div><div class="windows">
+ <div class="scala julia perl">
+ <div class="cpu gpu"><p>Follow the installation instructions <a class="reference internal" href="windows_setup.html"><span class="doc">in this guide</span></a> to set up MXNet.</p>
+</div>
+</div>
+</div><div class="windows">
+ <div class="python">
+ <div class="cpu">
+ <div class="virtualenv docker build-from-source"><p>Follow the installation instructions <a class="reference internal" href="windows_setup.html"><span class="doc">in this guide</span></a> to set up MXNet.</p>
+</div>
+</div>
+</div>
+</div><div class="devices">
+ <div class="raspberry-pi"><p>MXNet supports the Debian based Raspbian ARM based operating system so you can run MXNet on Raspberry Pi Devices.</p>
+<p>These instructions will walk through how to build MXNet for the Raspberry Pi and install the Python bindings for the library.</p>
+<p>The complete MXNet library and its requirements can take almost 200MB of RAM, and loading large models with the library can take over 1GB of RAM. Because of this, we recommend running MXNet on the Raspberry Pi 3 or an equivalent device that has more than 1 GB of RAM and a Secure Digital (SD) card that has at least 4 GB of free memory.</p>
+<p><strong>Install MXNet</strong></p>
+<p>Installing MXNet is a two-step process:</p>
+<ol class="simple">
+<li>Build the shared library from the MXNet C++ source code.</li>
+<li>Install the supported language-specific packages for MXNet.</li>
+</ol>
+<p><strong>Step 1</strong> Build the Shared Library</p>
+<p>On Raspbian versions Wheezy and later, you need the following dependencies:</p>
+<ul class="simple">
+<li>Git (to pull code from GitHub)</li>
+<li>libblas (for linear algebraic operations)</li>
+<li>libopencv (for computer vision operations. This is optional if you want to save RAM and Disk Space)</li>
+<li>A C++ compiler that supports C++ 11. The C++ compiler compiles and builds MXNet source code. Supported compilers include the following:</li>
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">G++ (4.8 or later)</a></li>
+</ul>
+<p>Install these dependencies using the following commands in any directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> sudo apt-get update
+ sudo apt-get -y install git cmake build-essential g++-4.8 c++-4.8 liblapack* libblas* libopencv*
+</pre></div>
+</div>
+<p>Clone the MXNet source code repository using the following <code class="docutils literal"><span class="pre">git</span></code> command in your home directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> git clone https://github.com/apache/incubator-mxnet.git --recursive
+ <span class="nb">cd</span> incubator-mxnet
+</pre></div>
+</div>
+<p>If you aren’t processing images with MXNet on the Raspberry Pi, you can minimize the size of the compiled library by building MXNet without the Open Source Computer Vision (OpenCV) library with the following commands:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> <span class="nb">export</span> <span class="nv">USE_OPENCV</span> <span class="o">=</span> <span class="m">0</span>
+ make
+</pre></div>
+</div>
+<p>Otherwise, you can build the complete MXNet library with the following command:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> make
+</pre></div>
+</div>
+<p>Executing either of these commands start the build process, which can take up to a couple hours, and creates a file called <code class="docutils literal"><span class="pre">libmxnet.so</span></code> in the mxnet/lib directory.</p>
+<p>If you are getting build errors in which the compiler is being killed, it is likely that the compiler is running out of memory (especially if you are on Raspberry Pi 1, 2 or Zero, which have less than 1GB of RAM), this can often be rectified by increasing the swapfile size on the Pi by editing the file /etc/dphys-swapfile and changing the line CONF_SWAPSIZE=100 to CONF_SWAPSIZE=1024, then running:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> sudo /etc/init.d/dphys-swapfile stop
+ sudo /etc/init.d/dphys-swapfile start
+ free -m <span class="c1"># to verify the swapfile size has been increased</span>
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install MXNet Python Bindings</p>
+<p>To install Python bindings run the following commands in the MXNet directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> <span class="nb">cd</span> python
+ pip install --upgrade pip
+ pip install -e .
+</pre></div>
+</div>
+<p>Note that the <code class="docutils literal"><span class="pre">-e</span></code> flag is optional. It is equivalent to <code class="docutils literal"><span class="pre">--editable</span></code> and means that if you edit the source files, these changes will be reflected in the package installed.</p>
+<p>You are now ready to run MXNet on your Raspberry Pi device. You can get started by following the tutorial on <a class="reference external" href="http://mxnet.io/tutorials/embedded/wine_detector.html">Real-time Object Detection with MXNet On The Raspberry Pi</a>.</p>
+<p><em>Note - Because the complete MXNet library takes up a significant amount of the Raspberry Pi’s limited RAM, when loading training data or large models into memory, you might have to turn off the GUI and terminate running processes to free RAM.</em></p>
+</div><div class="nvidia-jetson-tx2"><p>MXNet supports the Ubuntu Arch64 based operating system so you can run MXNet on NVIDIA Jetson Devices.</p>
+<p>These instructions will walk through how to build MXNet for the Pascal based <a class="reference external" href="http://www.nvidia.com/object/embedded-systems-dev-kits-modules.html">NVIDIA Jetson TX2</a> and install the corresponding python language bindings.</p>
+<p>For the purposes of this install guide we will assume that CUDA is already installed on your Jetson device.</p>
+<p><strong>Install MXNet</strong></p>
+<p>Installing MXNet is a two-step process:</p>
+<ol class="simple">
+<li>Build the shared library from the MXNet C++ source code.</li>
+<li>Install the supported language-specific packages for MXNet.</li>
+</ol>
+<p><strong>Step 1</strong> Build the Shared Library</p>
+<p>You need the following additional dependencies:</p>
+<ul class="simple">
+<li>Git (to pull code from GitHub)</li>
+<li>libatlas (for linear algebraic operations)</li>
+<li>libopencv (for computer vision operations)</li>
+<li>python pip (to load relevant python packages for our language bindings)</li>
+</ul>
+<p>Install these dependencies using the following commands in any directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> sudo apt-get update
+ sudo apt-get -y install git build-essential libatlas-base-dev libopencv-dev graphviz python-pip
+ sudo pip install pip --upgrade
+ sudo pip install setuptools numpy --upgrade
+ sudo pip install graphviz jupyter
+</pre></div>
+</div>
+<p>Clone the MXNet source code repository using the following <code class="docutils literal"><span class="pre">git</span></code> command in your home directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> git clone https://github.com/apache/incubator-mxnet.git --recursive
+ <span class="nb">cd</span> incubator-mxnet
+</pre></div>
+</div>
+<p>Edit the Makefile to install the MXNet with CUDA bindings to leverage the GPU on the Jetson:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> cp make/config.mk .
+ <span class="nb">echo</span> <span class="s2">"USE_CUDA=1"</span> >> config.mk
+ <span class="nb">echo</span> <span class="s2">"USE_CUDA_PATH=/usr/local/cuda"</span> >> config.mk
+ <span class="nb">echo</span> <span class="s2">"USE_CUDNN=1"</span> >> config.mk
+</pre></div>
+</div>
+<p>Edit the Mshadow Makefile to ensure MXNet builds with Pascal’s hardware level low precision acceleration by editing mshadow/make/mshadow.mk and adding the following after line 122:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="nv">MSHADOW_CFLAGS</span> <span class="o">+=</span> -DMSHADOW_USE_PASCAL<span class="o">=</span><span class="m">1</span>
+</pre></div>
+</div>
+<p>Now you can build the complete MXNet library with the following command:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> make -j <span class="k">$(</span>nproc<span class="k">)</span>
+</pre></div>
+</div>
+<p>Executing this command creates a file called <code class="docutils literal"><span class="pre">libmxnet.so</span></code> in the mxnet/lib directory.</p>
+<p><strong>Step 2</strong> Install MXNet Python Bindings</p>
+<p>To install Python bindings run the following commands in the MXNet directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> <span class="nb">cd</span> python
+ pip install --upgrade pip
+ pip install -e .
+</pre></div>
+</div>
+<p>Note that the <code class="docutils literal"><span class="pre">-e</span></code> flag is optional. It is equivalent to <code class="docutils literal"><span class="pre">--editable</span></code> and means that if you edit the source files, these changes will be reflected in the package installed.</p>
+<p>Add the mxnet folder to the path:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> <span class="nb">cd</span> ..
+ <span class="nb">export</span> <span class="nv">MXNET_HOME</span><span class="o">=</span><span class="k">$(</span><span class="nb">pwd</span><span class="k">)</span>
+ <span class="nb">echo</span> <span class="s2">"export PYTHONPATH=</span><span class="nv">$MXNET_HOME</span><span class="s2">/python:</span><span class="nv">$PYTHONPATH</span><span class="s2">"</span> >> ~/.bashrc
+ <span class="nb">source</span> ~/.bashrc
+</pre></div>
+</div>
+<p>You are now ready to run MXNet on your NVIDIA Jetson TX2 device.</p>
+</div>
+</div><p><br/></p>
+<!-- This # tag restarts the page and allows reuse
+ of the div classes for validation sections, etc --></div>
+<div class="section" id="validate-mxnet-installation">
+<span id="validate-mxnet-installation"></span><h1>Validate MXNet Installation<a class="headerlink" href="#validate-mxnet-installation" title="Permalink to this headline">¶</a></h1>
+<div class="linux macos">
+ <div class="python">
+ <div class="cpu"><div class="pip build-from-source"><p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+</div><div class="docker"><p>Launch a Docker container with <code class="docutils literal"><span class="pre">mxnet/python</span></code> image and run example <em>MXNet</em> python program on the terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker run -it mxnet/python bash <span class="c1"># Use sudo if you skip Step 2 in the installation instruction</span>
+
+<span class="c1"># Start a python terminal</span>
+root@4919c4f58cac:/# python
+</pre></div>
+</div>
+</div><div class="virtualenv"><p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+</div><p>Run a short <em>MXNet</em> python program to create a 2X3 matrix of ones, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">mxnet</span> <span class="kn">as</span> <span class="nn">mx</span>
+<span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">))</span>
+<span class="gp">>>> </span><span class="n">b</span> <span class="o">=</span> <span class="n">a</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+<span class="gp">>>> </span><span class="n">b</span><span class="o">.</span><span class="n">asnumpy</span><span class="p">()</span>
+<span class="go">array([[ 3., 3., 3.],</span>
+<span class="go"> [ 3., 3., 3.]], dtype=float32)</span>
+</pre></div>
+</div>
+</div>
+</div>
+</div><!-- Validate Windows CPU pip install --><div class="windows">
+ <div class="python">
+ <div class="cpu">
+ <div class="pip"><p>Run a short <em>MXNet</em> python program to create a 2X3 matrix of ones, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">mxnet</span> <span class="kn">as</span> <span class="nn">mx</span>
+<span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">))</span>
+<span class="gp">>>> </span><span class="n">b</span> <span class="o">=</span> <span class="n">a</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+<span class="gp">>>> </span><span class="n">b</span><span class="o">.</span><span class="n">asnumpy</span><span class="p">()</span>
+<span class="go">array([[ 3., 3., 3.],</span>
+<span class="go"> [ 3., 3., 3.]], dtype=float32)</span>
+</pre></div>
+</div>
+</div>
+</div>
+</div>
+</div><!-- Mac OS GPU installation validation --><div class="macos">
+ <div class="python">
+ <div class="gpu"><div class="pip virtualenv docker">
+</br><p>Will be available soon.</p>
+</div><div class="build-from-source">
+</br><p>From the MXNet root directory run: <code class="docutils literal"><span class="pre">python</span> <span class="pre">example/image-classification/train_mnist.py</span> <span class="pre">--network</span> <span class="pre">lenet</span> <span class="pre">--gpus</span> <span class="pre">0</span></code> to test GPU training.</p>
+</div>
+</div>
+</div>
+</div><!-- Windows GPU installation validation --><div class="windows">
+ <div class="python">
+ <div class="gpu"><div class="virtualenv docker">
+</br><p>Will be available soon.</p>
+</div><div class="pip build-from-source">
+</br><p>From the MXNet root directory run: <code class="docutils literal"><span class="pre">python</span> <span class="pre">example/image-classification/train_mnist.py</span> <span class="pre">--network</span> <span class="pre">lenet</span> <span class="pre">--gpus</span> <span class="pre">0</span></code> to test GPU training.</p>
+</div>
+</div>
+</div>
+</div><!-- Validation for GPU machines --><div class="linux">
+ <div class="python">
+ <div class="gpu"><div class="pip build-from-source"><p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+</div><div class="docker"><p>Launch a NVIDIA Docker container with <code class="docutils literal"><span class="pre">mxnet/python:gpu</span></code> image and run example <em>MXNet</em> python program on the terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ nvidia-docker run -it mxnet/python:gpu bash <span class="c1"># Use sudo if you skip Step 2 in the installation instruction</span>
+
+<span class="c1"># Start a python terminal</span>
+root@4919c4f58cac:/# python
+</pre></div>
+</div>
+</div><div class="virtualenv"><p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+</div><p>Run a short <em>MXNet</em> python program to create a 2X3 matrix of ones <em>a</em> on a <em>GPU</em>, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3. We use <em>mx.gpu()</em>, to set <em>MXNet</em> context to be GPUs.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">mxnet</span> <span class="kn">as</span> <span class="nn">mx</span>
+<span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">mx</span><span class="o">.</span><span class="n">gpu</span><span class="p">())</span>
+<span class="gp">>>> </span><span class="n">b</span> <span class="o">=</span> <span class="n">a</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+<span class="gp">>>> </span><span class="n">b</span><span class="o">.</span><span class="n">asnumpy</span><span class="p">()</span>
+<span class="go">array([[ 3., 3., 3.],</span>
+<span class="go"> [ 3., 3., 3.]], dtype=float32)</span>
+</pre></div>
+</div>
+</div>
+</div>
+</div><!-- Linux Clean up -->
+<div class="linux">
+ <div class="python">
+ <div class="cpu"><div class="pip build-from-source"><p>Exit the Python terminal.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><div class="virtualenv"><p>Exit the Python terminal and Deactivate the virtualenv <em>MXNet</em> environment.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">(mxnet)$ deactivate</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><div class="docker"><p>Exit the Python terminal and mxnet/python docker container.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">root@4919c4f58cac:/# exit</span>
+</pre></div>
+</div>
+</div></div>
+</div>
+</div><!-- MacOS Clean up -->
+<div class="macos">
+ <div class="python">
+ <div class="cpu"><div class="pip build-from-source"><p>Exit the Python terminal.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><div class="virtualenv"><p>Exit the Python terminal and Deactivate the virtualenv <em>MXNet</em> environment.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">(mxnet)$ deactivate</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><div class="docker"><p>Exit the Python terminal and then the docker container.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">root@4919c4f58cac:/# exit</span>
+</pre></div>
+</div>
+</div></div>
+</div>
+</div><!-- Validation for cloud installation --><div class="cloud"><p>Login to the cloud instance you launched, with pre-installed <em>MXNet</em>, following the guide by corresponding cloud provider.</p>
+<p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+<!-- Example Python code for CPU --><div class="cpu"><p>Run a short <em>MXNet</em> python program to create a 2X3 matrix of ones, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">mxnet</span> <span class="kn">as</span> <span class="nn">mx</span>
+<span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">))</span>
+<span class="gp">>>> </span><span class="n">b</span> <span class="o">=</span> <span class="n">a</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+<span class="gp">>>> </span><span class="n">b</span><span class="o">.</span><span class="n">asnumpy</span><span class="p">()</span>
+<span class="go">array([[ 3., 3., 3.],</span>
+<span class="go"> [ 3., 3., 3.]], dtype=float32)</span>
+</pre></div>
+</div>
+<p>Exit the Python terminal.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><!-- Example Python code for CPU --><div class="gpu"><p>Run a short <em>MXNet</em> python program to create a 2X3 matrix of ones <em>a</em> on a <em>GPU</em>, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3. We use <em>mx.gpu()</em>, to set <em>MXNet</em> context to be GPUs.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">mxnet</span> <span class="kn">as</span> <span class="nn">mx</span>
+<span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">mx</span><span class="o">.</span><span class="n">gpu</span><span class="p">())</span>
+<span class="gp">>>> </span><span class="n">b</span> <span class="o">=</span> <span class="n">a</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+<span class="gp">>>> </span><span class="n">b</span><span class="o">.</span><span class="n">asnumpy</span><span class="p">()</span>
+<span class="go">array([[ 3., 3., 3.],</span>
+<span class="go"> [ 3., 3., 3.]], dtype=float32)</span>
+</pre></div>
+</div>
+</div></div><!-- Example R code for CPU --><div class="linux macos windows">
+ <div class="r">
+ <div class="cpu"><p>Run a short <em>MXNet</em> R program to create a 2X3 matrix of ones, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3.</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span><span class="kn">library</span><span class="p">(</span>mxnet<span class="p">)</span>
+a <span class="o"><-</span> mx.nd.ones<span class="p">(</span><span class="kt">c</span><span class="p">(</span><span class="m">2</span><span class="p">,</span><span class="m">3</span><span class="p">),</span> ctx <span class="o">=</span> mx.cpu<span class="p">())</span>
+b <span class="o"><-</span> a <span class="o">*</span> <span class="m">2</span> <span class="o">+</span> <span class="m">1</span>
+b
+</pre></div>
+</div>
+</div>
+</div>
+</div><!-- Example R code for GPU --><div class="linux macos windows">
+ <div class="r">
+ <div class="gpu"><p>Run a short <em>MXNet</em> R program to create a 2X3 matrix of ones <em>a</em> on a <em>GPU</em>, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3. We use <em>mx.gpu()</em>, to set <em>MXNet</em> context to be GPUs.</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span><span class="kn">library</span><span class="p">(</span>mxnet<span class="p">)</span>
+a <span class="o"><-</span> mx.nd.ones<span class="p">(</span><span class="kt">c</span><span class="p">(</span><span class="m">2</span><span class="p">,</span><span class="m">3</span><span class="p">),</span> ctx <span class="o">=</span> mx.gpu<span class="p">())</span>
+b <span class="o"><-</span> a <span class="o">*</span> <span class="m">2</span> <span class="o">+</span> <span class="m">1</span>
+b
+</pre></div>
+</div>
+</div>
+</div>
+</div><div class="linux">
+ <div class="scala julia perl">
+ <div class="cpu gpu"><p>Will be available soon.</p>
+</div>
+</div>
+</div><div class="macos">
+ <div class="scala julia perl">
+ <div class="cpu gpu"><p>Will be available soon.</p>
+</div>
+</div>
+</div><!-- Windows MXNet Installation validation -->
+<div class="windows">
+ <div class="python">
+ <div class="cpu"><div class="build-from-source virtualenv docker">
+<br/>
+Will be available soon.
+</div></div>
+</div><div class="scala julia perl">
+<div class="cpu gpu"><p>Will be available soon.</p>
+</div>
+</div>
+</div>
+<!-- End Windows Installation validation --><div class="devices">
+ <div class="raspberry-pi"><p>Will be available soon.</p>
+</div>
+<div class="nvidia-jetson-tx2"><p>Will be available soon.</p>
+</div>
+</div></div>
+<div class="section" id="download-source-package">
+<span id="download-source-package"></span><h1>Download Source Package<a class="headerlink" href="#download-source-package" title="Permalink to this headline">¶</a></h1>
+</div>
+
+
+ </div>
+ </div>
+
+ <div class="sphinxsidebar rightsidebar" role="navigation" aria-label="main navigation">
+ <div class="sphinxsidebarwrapper">
+ <h3><a href="../index.html">Table Of Contents</a></h3>
+ <ul>
+<li><a class="reference internal" href="#">Installing MXNet</a></li>
+<li><a class="reference internal" href="#validate-mxnet-installation">Validate MXNet Installation</a></li>
+<li><a class="reference internal" href="#download-source-package">Download Source Package</a></li>
+</ul>
+
+ </div>
+ </div>
+ </div><div class="footer">
+<div class="section-disclaimer">
+<div class="container">
+ <div>
+ <img src="https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/apache_incubator_logo.png" height=60>
+ <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 © 2017-2018, 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 src="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.6/js/bootstrap.min.js" integrity="sha384-0mSbJDEHialfmuBBQP6A4Qrprq5OVfW37PRR3j5ELqxss1yVqOtnepnHVP9aJ7xS" crossorigin="anonymous"></script>
+ <script type="text/javascript" src="../_static/js/sidebar.js"></script>
+ <script type="text/javascript" src="../_static/js/search.js"></script>
+ <script type="text/javascript" src="../_static/js/navbar.js"></script>
+ <script type="text/javascript" src="../_static/js/clipboard.min.js"></script>
+ <script type="text/javascript" src="../_static/js/copycode.js"></script>
+ <script type="text/javascript" src="../_static/js/page.js"></script>
+ <script type="text/javascript">
+ $('body').ready(function () {
+ $('body').css('visibility', 'visible');
+ });
+ </script>
+ </body>
+</html>
diff --git a/docs/build_version_doc/artifacts/index.html b/docs/build_version_doc/artifacts/index.html
new file mode 100644
index 00000000000..51eced8f45f
--- /dev/null
+++ b/docs/build_version_doc/artifacts/index.html
@@ -0,0 +1,73 @@
+<div id="splash">
+ <div class="container">
+ <div class="row">
+ <div class="col-lg-12">
+ <div id="banner-title" class='col-sm-6 col-xs-12'><span>Apache MXNet (Incubating)</span>
+ <p id="landing-title">A flexible and efficient library for deep learning.</p>
+ <div id='landing-btn-blk'>
+ <div id="install_blk">
+ <a href="install/index.html" id="install_btn">Install</a>
+ </div>
+ <div id="why_mxnet">
+ <a href="get_started/why_mxnet.html" id="why_mxnet_btn">Learn More</a>
+ </div>
+ </div>
+ </div>
+ </div>
+ </div>
+ </div>
+</div>
+<div class="section-tout">
+ <div class="container">
+ <div class="row">
+ <div class="col-lg-4 col-sm-12">
+ <h3>MXNet 1.2.0.rc0 Released</h3>
+ <p>We're excited to announce the release of MXNet 1.2.0.rc0! Check out the release notes for latest updates.</p>
+ <a href="https://github.com/apache/incubator-mxnet/releases/tag/1.2.0.rc0">Learn More</a>
+ </div>
+ <div class="col-lg-4 col-sm-12">
+ <h3>Improved ONNX Support</h3>
+ <p>MXNet now supports importing ONNX models natively with the new ONNX-MXNet API. Try out the super resolution example, or a tutorial on fine-tuning an ONXX model with Gluon.
+ </p>
+ <a href="api/python/contrib/onnx.html#onnx-tutorials">Learn More</a>
+ </div>
+ <div class="col-lg-4 col-sm-12">
+ <h3>Introducing the Scala Inference API</h3>
+ <p>A model loading and inference API is now available for Scala developers. Try out the examples for single shot detection and loading models for image classification.
+ </p>
+ <a href="api/scala/index.html">Learn More</a>
+ </div>
+ </div>
+ </div>
+</div>
+
+<div class="section-util">
+ <div class="container">
+ <div class="row">
+ <div id="model-zoo-blk" class="col-lg-4 col-sm-12">
+ <span class="glyphicon glyphicon-folder-open"></span>
+ <h2>Gluon Model Zoo</h2>
+ <p>One-click pre-trained models, included in Gluon. Fast implementations of many state-of-the-art models, for plug-and-play effortless use.</p>
+ <div class='util-btn'>
+ <a id="model-zoo-link" href="api/python/gluon/model_zoo.html">Gluon model zoo</a>
+ </div>
+ </div>
+ <div id="example-blk" class="col-lg-4 col-sm-12">
+ <span class="glyphicon glyphicon-list-alt"></span>
+ <h2>Examples</h2>
+ <p>Explore projects from simple demos, examples, tutorials to state-of-the-art research.</p>
+ <div class='util-btn'>
+ <a id="example-link" href="https://github.com/dmlc/mxnet/tree/master/example">MXNet examples</a>
+ </div>
+ </div>
+ <div id="tutorial-blk" class="col-lg-4 col-sm-12">
+ <span class="glyphicon glyphicon-ok-circle"></span>
+ <h2>Tutorials</h2>
+ <p>These tutorials introduce a few fundamental concepts in deep learning and how to implement them in MXNet.</p>
+ <div class='util-btn'>
+ <a id="tutorial-link" href="tutorials/index.html">MXNet tutorials</a>
+ </div>
+ </div>
+ </div>
+ </div>
+</div>
diff --git a/docs/build_version_doc/artifacts/master.index.html b/docs/build_version_doc/artifacts/master.index.html
new file mode 100644
index 00000000000..e9fb9716dc1
--- /dev/null
+++ b/docs/build_version_doc/artifacts/master.index.html
@@ -0,0 +1,1371 @@
+<!DOCTYPE html>
+
+<html lang="en">
+ <head>
+ <meta charset="utf-8">
+ <meta http-equiv="X-UA-Compatible" content="IE=edge">
+ <meta name="viewport" content="width=device-width, initial-scale=1">
+
+
+
+ <title>Installing MXNet — mxnet documentation</title>
+
+ <link rel="stylesheet" href="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.6/css/bootstrap.min.css" integrity="sha384-1q8mTJOASx8j1Au+a5WDVnPi2lkFfwwEAa8hDDdjZlpLegxhjVME1fgjWPGmkzs7" crossorigin="anonymous">
+ <link rel="stylesheet" href="https://maxcdn.bootstrapcdn.com/font-awesome/4.5.0/css/font-awesome.min.css">
+
+
+ <link rel="stylesheet" href="../_static/basic.css" type="text/css" />
+ <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+
+ <link rel="stylesheet" href="../_static/mxnet.css" type="text/css">
+
+ <script type="text/javascript">
+ var DOCUMENTATION_OPTIONS = {
+ URL_ROOT: '../',
+ VERSION: '',
+ COLLAPSE_INDEX: false,
+ FILE_SUFFIX: '.html',
+ HAS_SOURCE: true,
+ SOURCELINK_SUFFIX: '.txt'
+ };
+ </script>
+
+ <script type="text/javascript" src="https://code.jquery.com/jquery-1.11.1.min.js"></script>
+
+ <script type="text/javascript" src="../_static/underscore.js"></script>
+
+ <script type="text/javascript" src="../_static/searchtools_custom.js"></script>
+
+ <script type="text/javascript" src="../_static/doctools.js"></script>
+
+ <script type="text/javascript" src="../_static/selectlang.js"></script>
+
+
+ <script type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.1/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></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://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.0/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script> -->
+ <!-- -->
+ <link rel="index" title="Index" href="../genindex.html" />
+ <link rel="search" title="Search" href="../search.html" />
+
+ <link rel="icon" type="image/png" href="https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/mxnet-icon.png">
+ </head>
+ <body role="document" background="https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/mxnet-background-compressed.jpeg">
+ <div class='content-block'><div class="navbar navbar-fixed-top">
+ <div class="container" id="navContainer">
+ <div id="header-inner" class="innder">
+ <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 id="main-nav" class='nav-bar'>
+ <a class="main-nav-link" href="../install/index.html">Install</a>
+
+ <span id="dropdown-menu-position-anchor">
+ <a href="#" class="main-nav-link dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">Gluon <span class="caret"></span></a>
+ <ul id="package-dropdown-menu" class="dropdown-menu navbar-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 href="#" class="main-nav-link dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">API <span class="caret"></span></a>
+ <ul id="package-dropdown-menu" class="dropdown-menu navbar-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 href="#" class="main-nav-link dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">Docs <span class="caret"></span></a>
+ <ul id="package-dropdown-menu-docs" class="dropdown-menu navbar-menu">
+ <li><a class="main-nav-link" href="../tutorials/index.html">Tutorials</a>
+ <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/master/example">Examples</a></li>
+ <li><a class="main-nav-link" href="../api/python/gluon/model_zoo.html">Gluon 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 href="#" class="main-nav-link dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">Community <span class="caret"></span></a>
+ <ul id="package-dropdown-menu-community" class="dropdown-menu navbar-menu">
+ <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>
+ <li><a class="main-nav-link" href="http://discuss.mxnet.io">Discuss</a></li>
+ </ul>
+ </span>
+ </nav>
+ <script> function getRootPath(){ return "../" } </script>
+ <div class="burgerIcon dropdown">
+ <a href="#" class="dropdown-toggle" data-toggle="dropdown" role="button">☰</a>
+ <ul id="burgerMenu" class="dropdown-menu">
+ <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 tabindex="-1" href="../community/index.html">Community</a></li>
+ <li><a tabindex="-1" href="../community/contribute.html">Contribute</a></li>
+ <li><a tabindex="-1" href="../community/powered_by.html">Powered By</a></li>
+ </ul>
+ </li>
+
+ <li class="dropdown-submenu">
+ <a href="#" tabindex="-1">API</a>
+ <ul class="dropdown-menu">
+
+ <li><a tabindex="-1" href="../api/python/index.html">Python</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/scala/index.html">Scala</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/r/index.html">R</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/julia/index.html">Julia</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/c++/index.html">C++</a>
+ </li>
+
+ <li><a tabindex="-1" href="../api/perl/index.html">Perl</a>
+ </li>
+
+ </ul>
+ </li>
+
+ <li class="dropdown-submenu">
+ <a href="#" tabindex="-1">Docs</a>
+ <ul class="dropdown-menu">
+ <li><a tabindex="-1" href="../tutorials/index.html">Tutorials</a></li>
+ <li><a tabindex="-1" href="../faq/index.html">FAQ</a></li>
+ <li><a tabindex="-1" href="../architecture/index.html">Architecture</a></li>
+ <li><a tabindex="-1" href="https://github.com/apache/incubator-mxnet/tree/master/example">Examples</a></li>
+ <li><a tabindex="-1" href="../api/python/gluon/model_zoo.html">Gluon 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>
+ </ul>
+ </div>
+ <div class="plusIcon dropdown">
+ <a href="#" class="dropdown-toggle" data-toggle="dropdown" role="button"><span class="glyphicon glyphicon-plus" aria-hidden="true"></span></a>
+ <ul id="plusMenu" class="dropdown-menu dropdown-menu-right"></ul>
+ </div>
+ <div id="search-input-wrap">
+
+<form class="" role="search" action="../search.html" method="get" autocomplete="off">
+ <div class="form-group inner-addon left-addon">
+ <i class="glyphicon glyphicon-search"></i>
+ <input type="text" name="q" class="form-control" placeholder="Search">
+ </div>
+ <input type="hidden" name="check_keywords" value="yes" />
+ <input type="hidden" name="area" value="default" />
+
+</form>
+ <div id="search-preview"></div>
+ </div>
+ <div id='searchIcon'>
+ <span class="glyphicon glyphicon-search" aria-hidden="true"></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 class="sphinxsidebar leftsidebar" role="navigation" aria-label="main 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>
+
+ <div class="section" id="installing-mxnet">
+<span id="installing-mxnet"></span><h1>Installing MXNet<a class="headerlink" href="#installing-mxnet" title="Permalink to this headline">¶</a></h1>
+<p>Indicate your preferred configuration. Then, follow the customized commands to install <em>MXNet</em>.</p>
+<script type="text/javascript" src='../_static/js/options.js'></script><!-- START - OS Menu --><div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">Linux</button>
+ <button type="button" class="btn btn-default opt">MacOS</button>
+ <button type="button" class="btn btn-default opt">Windows</button>
+ <button type="button" class="btn btn-default opt">Cloud</button>
+ <button type="button" class="btn btn-default opt">Devices</button>
+</div><!-- START - Language Menu --><div class="linux macos windows">
+<div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">Python</button>
+ <button type="button" class="btn btn-default opt">Scala</button>
+ <button type="button" class="btn btn-default opt">R</button>
+ <button type="button" class="btn btn-default opt">Julia</button>
+ <button type="button" class="btn btn-default opt">Perl</button>
+</div>
+</div><!-- No CPU GPU for other Devices -->
+<div class="linux macos windows cloud">
+<div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">CPU</button>
+ <button type="button" class="btn btn-default opt">GPU</button>
+</div>
+</div><!-- other devices -->
+<div class="devices">
+<div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">Raspberry Pi</button>
+ <button type="button" class="btn btn-default opt">NVIDIA Jetson TX2</button>
+</div>
+</div><!-- Linux Python GPU Options --><div class="linux macos windows">
+<div class="python">
+<div class="cpu gpu">
+<div class="btn-group opt-group" role="group">
+ <button type="button" class="btn btn-default opt active">Pip</button>
+ <button type="button" class="btn btn-default opt">Virtualenv</button>
+ <button type="button" class="btn btn-default opt">Docker</button>
+ <button type="button" class="btn btn-default opt">Build from Source</button>
+</div>
+</div>
+</div>
+</div><!-- END - Main Menu --><!-- START - Linux Python CPU Installation Instructions --><div class="linux">
+<div class="python">
+<div class="cpu"><p>The following installation instructions have been tested on Ubuntu 14.04 and 16.04.</p>
+<div class="pip">
+<br/><p><strong>Step 1</strong> Install prerequisites - wget and latest pip.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command in the terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y wget python gcc
+$ wget https://bootstrap.pypa.io/get-pip.py <span class="o">&&</span> sudo python get-pip.py
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install MXNet with OpenBLAS acceleration.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet --pre
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Experimental Choice</strong> If You would like to install mxnet with Intel MKL, try the experimental pip package with MKL:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet-mkl --pre
+</pre></div>
+</div>
+</div> <!-- End of pip --><div class="virtualenv">
+<br/><p><strong>Step 1</strong> Install virtualenv for Ubuntu.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y python-dev python-virtualenv
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Create and activate virtualenv environment for MXNet.</p>
+<p>Following command creates a virtualenv environment at <code class="docutils literal"><span class="pre">~/mxnet</span></code> directory. However, you can choose any directory by replacing <code class="docutils literal"><span class="pre">~/mxnet</span></code> with a directory of your choice.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ virtualenv --system-site-packages ~/mxnet
+</pre></div>
+</div>
+<p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install MXNet in the active virtualenv environment.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install --upgrade pip
+</pre></div>
+</div>
+<p>Install <em>MXNet</em> with OpenBLAS acceleration.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet --pre
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 5</strong> Validate the installation by running simple <em>MXNet</em> code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Note</strong> You can read more about virtualenv <a class="reference external" href="https://virtualenv.pypa.io/en/stable/userguide/">here</a>.</p>
+</div> <!-- END of virtualenv --><div class="docker">
+<br/><p>Docker images with <em>MXNet</em> are available at <a class="reference external" href="https://hub.docker.com/r/mxnet/">Docker Hub</a>.</p>
+<p><strong>Step 1</strong> Install Docker on your machine by following the <a class="reference external" href="https://docs.docker.com/engine/installation/linux/ubuntu/#install-using-the-repository">docker installation instructions</a>.</p>
+<p><em>Note</em> - You can install Community Edition (CE) to get started with <em>MXNet</em>.</p>
+<p><strong>Step 2</strong> [Optional] Post installation steps to manage Docker as a non-root user.</p>
+<p>Follow the four steps in this <a class="reference external" href="https://docs.docker.com/engine/installation/linux/linux-postinstall/#manage-docker-as-a-non-root-user">docker documentation</a> to allow managing docker containers without <em>sudo</em>.</p>
+<p>If you skip this step, you need to use <em>sudo</em> each time you invoke Docker.</p>
+<p><strong>Step 3</strong> Pull the MXNet docker image.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker pull mxnet/python <span class="c1"># Use sudo if you skip Step 2</span>
+</pre></div>
+</div>
+<p>You can list docker images to see if mxnet/python docker image pull was successful.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker images <span class="c1"># Use sudo if you skip Step 2</span>
+
+REPOSITORY TAG IMAGE ID CREATED SIZE
+mxnet/python latest 00d026968b3c <span class="m">3</span> weeks ago <span class="m">1</span>.41 GB
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div> <!-- END of docker --><div class="build-from-source">
+<br/><p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings. Example - Python bindings, Scala bindings.</li>
+</ol>
+<p><strong>Minimum Requirements</strong></p>
+<ol class="simple">
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">GCC 4.8</a> or later to compile C++ 11.</li>
+<li><a class="reference external" href="https://www.gnu.org/software/make/">GNU Make</a></li>
+</ol>
+<p><br/></p>
+<p><strong>Build the MXNet core shared library</strong></p>
+<p><strong>Step 1</strong> Install build tools and git.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y build-essential git
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install OpenBLAS.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="https://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprograms">BLAS</a> and <a class="reference external" href="https://en.wikipedia.org/wiki/LAPACK">LAPACK</a> libraries for accelerated numerical computations on CPU machine. There are several flavors of BLAS/LAPACK libraries - <a class="reference external" href="http://www.openblas.net/">OpenBLAS</a>, <a class="reference external" href="http://math-atlas.sourceforge.net/">ATLAS</a> and <a class="reference external" href="https://software.intel.com/en-us/intel-mkl">MKL</a>. In this step we install OpenBLAS. You can choose to install ATLAS or MKL.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopenblas-dev liblapack-dev
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install OpenCV.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="http://opencv.org/">OpenCV</a> for efficient image loading and augmentation operations.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopencv-dev
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Download MXNet sources and build MXNet core shared library. You can clone the repository as described in the following code block, or you may try the <a href="download.html">download links</a> for your desired MXNet version.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ git clone --recursive https://github.com/apache/incubator-mxnet
+$ <span class="nb">cd</span> incubator-mxnet
+$ make -j <span class="k">$(</span>nproc<span class="k">)</span> <span class="nv">USE_OPENCV</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_BLAS</span><span class="o">=</span>openblas
+</pre></div>
+</div>
+<p><em>Note</em> - USE_OPENCV and USE_BLAS are make file flags to set compilation options to use OpenCV and BLAS library. You can explore and use more compilation options in <code class="docutils literal"><span class="pre">make/config.mk</span></code>.</p>
+<p><br/></p>
+<p><strong>Build the MXNet Python binding</strong></p>
+<p><strong>Step 1</strong> Install prerequisites - python, setup-tools, python-pip and libfortran (required for Numpy).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y python-dev python-setuptools python-pip libgfortran3
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install the MXNet Python binding.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">cd</span> python
+$ pip install --upgrade pip
+$ pip install -e .
+</pre></div>
+</div>
+<p>Note that the <code class="docutils literal"><span class="pre">-e</span></code> flag is optional. It is equivalent to <code class="docutils literal"><span class="pre">--editable</span></code> and means that if you edit the source files, these changes will be reflected in the package installed.</p>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div><!-- END of build from source -->
+</div><!-- END of CPU -->
+<!-- END - Linux Python CPU Installation Instructions --><!-- START - Linux Python GPU Installation Instructions --><div class="gpu"><p>The following installation instructions have been tested on Ubuntu 14.04 and 16.04.</p>
+<p><strong>Prerequisites</strong></p>
+<p>Install the following NVIDIA libraries to setup <em>MXNet</em> with GPU support:</p>
+<ol class="simple">
+<li>Install CUDA 9.0 following the NVIDIA’s <a class="reference external" href="http://docs.nvidia.com/cuda/cuda-installation-guide-linux/">installation guide</a>.</li>
+<li>Install cuDNN 7 for CUDA 9.0 following the NVIDIA’s <a class="reference external" href="https://developer.nvidia.com/cudnn">installation guide</a>. You may need to register with NVIDIA for downloading the cuDNN library.</li>
+</ol>
+<p><strong>Note:</strong> Make sure to add CUDA install path to <code class="docutils literal"><span class="pre">LD_LIBRARY_PATH</span></code>.</p>
+<p>Example - <em>export LD_LIBRARY_PATH=/usr/local/cuda/lib64/:$LD_LIBRARY_PATH</em></p>
+<div class="pip">
+<br/><p><strong>Step 1</strong> Install prerequisites - wget and latest pip.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command in the terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y wget python
+$ wget https://bootstrap.pypa.io/get-pip.py <span class="o">&&</span> sudo python get-pip.py
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install <em>MXNet</em> with GPU support using CUDA 9.0</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet-cu90 --pre
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Experimental Choice</strong> If You would like to install mxnet with Intel MKL, try the experimental pip package with MKL:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet-cu90mkl --pre
+</pre></div>
+</div>
+</div> <!-- END of pip --><div class="virtualenv"><p><br/></p>
+<p><strong>Step 1</strong> Install virtualenv for Ubuntu.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y python-dev python-virtualenv
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Create and activate virtualenv environment for MXNet.</p>
+<p>Following command creates a virtualenv environment at <code class="docutils literal"><span class="pre">~/mxnet</span></code> directory. However, you can choose any directory by replacing <code class="docutils literal"><span class="pre">~/mxnet</span></code> with a directory of your choice.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ virtualenv --system-site-packages ~/mxnet
+</pre></div>
+</div>
+<p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install MXNet in the active virtualenv environment.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install --upgrade pip
+</pre></div>
+</div>
+<p>Install <em>MXNet</em> with GPU support using CUDA 9.0.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet-cu90 --pre
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 5</strong> Validate the installation by running simple <em>MXNet</em> code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Note</strong> You can read more about virtualenv <a class="reference external" href="https://virtualenv.pypa.io/en/stable/userguide/">here</a>.</p>
+</div> <!-- END of virtualenv --><div class="docker"><p><br/></p>
+<p>Docker images with <em>MXNet</em> are available at <a class="reference external" href="https://hub.docker.com/r/mxnet/">Docker Hub</a>.</p>
+<p><strong>Step 1</strong> Install Docker on your machine by following the <a class="reference external" href="https://docs.docker.com/engine/installation/linux/ubuntu/#install-using-the-repository">docker installation instructions</a>.</p>
+<p><em>Note</em> - You can install Community Edition (CE) to get started with <em>MXNet</em>.</p>
+<p><strong>Step 2</strong> [Optional] Post installation steps to manage Docker as a non-root user.</p>
+<p>Follow the four steps in this <a class="reference external" href="https://docs.docker.com/engine/installation/linux/linux-postinstall/#manage-docker-as-a-non-root-user">docker documentation</a> to allow managing docker containers without <em>sudo</em>.</p>
+<p>If you skip this step, you need to use <em>sudo</em> each time you invoke Docker.</p>
+<p><strong>Step 3</strong> Install <em>nvidia-docker-plugin</em> following the <a class="reference external" href="https://github.com/NVIDIA/nvidia-docker/wiki/Installation">installation instructions</a>. <em>nvidia-docker-plugin</em> is required to enable the usage of GPUs from the docker containers.</p>
+<p><strong>Step 4</strong> Pull the MXNet docker image.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker pull mxnet/python:gpu <span class="c1"># Use sudo if you skip Step 2</span>
+</pre></div>
+</div>
+<p>You can list docker images to see if mxnet/python docker image pull was successful.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker images <span class="c1"># Use sudo if you skip Step 2</span>
+
+REPOSITORY TAG IMAGE ID CREATED SIZE
+mxnet/python gpu 493b2683c269 <span class="m">3</span> weeks ago <span class="m">4</span>.77 GB
+</pre></div>
+</div>
+<p><strong>Step 5</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div> <!-- END of docker --><div class="build-from-source"><p><br/></p>
+<p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings. Example - Python bindings, Scala bindings.</li>
+</ol>
+<p><strong>Minimum Requirements</strong></p>
+<ol class="simple">
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">GCC 4.8</a> or later to compile C++ 11.</li>
+<li><a class="reference external" href="https://www.gnu.org/software/make/">GNU Make</a></li>
+</ol>
+<p><br/></p>
+<p><strong>Build the MXNet core shared library</strong></p>
+<p><strong>Step 1</strong> Install build tools and git.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y build-essential git
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install OpenBLAS.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="https://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprograms">BLAS</a> and <a class="reference external" href="https://en.wikipedia.org/wiki/LAPACK">LAPACK</a> libraries for accelerated numerical computations on CPU machine. There are several flavors of BLAS/LAPACK libraries - <a class="reference external" href="http://www.openblas.net/">OpenBLAS</a>, <a class="reference external" href="http://math-atlas.sourceforge.net/">ATLAS</a> and <a class="reference external" href="https://software.intel.com/en-us/intel-mkl">MKL</a>. In this step we install OpenBLAS. You can choose to install ATLAS or MKL.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopenblas-dev liblapack-dev
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install OpenCV.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="http://opencv.org/">OpenCV</a> for efficient image loading and augmentation operations.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopencv-dev
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Download MXNet sources and build MXNet core shared library. You can clone the repository as described in the following code block, or you may try the <a href="download.html">download links</a> for your desired MXNet version.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ git clone --recursive https://github.com/apache/incubator-mxnet
+$ <span class="nb">cd</span> incubator-mxnet
+$ make -j <span class="k">$(</span>nproc<span class="k">)</span> <span class="nv">USE_OPENCV</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_BLAS</span><span class="o">=</span>openblas <span class="nv">USE_CUDA</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_CUDA_PATH</span><span class="o">=</span>/usr/local/cuda <span class="nv">USE_CUDNN</span><span class="o">=</span><span class="m">1</span>
+</pre></div>
+</div>
+<p><em>Note</em> - USE_OPENCV, USE_BLAS, USE_CUDA, USE_CUDA_PATH AND USE_CUDNN are make file flags to set compilation options to use OpenCV, OpenBLAS, CUDA and cuDNN libraries. You can explore and use more compilation options in <code class="docutils literal"><span class="pre">make/config.mk</span></code>. Make sure to set USE_CUDA_PATH to right CUDA installation path. In most cases it is - <em>/usr/local/cuda</em>.</p>
+<p><br/></p>
+<p><strong>Install the MXNet Python binding</strong></p>
+<p><strong>Step 1</strong> Install prerequisites - python, setup-tools, python-pip and libfortran (required for Numpy)..</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y python-dev python-setuptools python-pip libgfortran3
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install the MXNet Python binding.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">cd</span> python
+$ pip install --upgrade pip
+$ pip install -e .
+</pre></div>
+</div>
+<p>Note that the <code class="docutils literal"><span class="pre">-e</span></code> flag is optional. It is equivalent to <code class="docutils literal"><span class="pre">--editable</span></code> and means that if you edit the source files, these changes will be reflected in the package installed.</p>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>sudo apt-get install graphviz
+pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div> <!-- END of build from source -->
+</div> <!-- END of GPU -->
+</div> <!-- END of Python -->
+-<!-- END - Linux Python GPU Installation Instructions --><div class="r">
+<div class="cpu"><p><br/></p>
+<p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings.</li>
+</ol>
+<p><strong>Minimum Requirements</strong></p>
+<ol class="simple">
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">GCC 4.8</a> or later to compile C++ 11.</li>
+<li><a class="reference external" href="https://www.gnu.org/software/make/">GNU Make</a></li>
+</ol>
+<p><br/></p>
+<p><strong>Build the MXNet core shared library</strong></p>
+<p><strong>Step 1</strong> Install build tools and git.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y build-essential git
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install OpenBLAS.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="https://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprograms">BLAS</a> and <a class="reference external" href="https://en.wikipedia.org/wiki/LAPACK">LAPACK</a> libraries for accelerated numerical computations on CPU machine. There are several flavors of BLAS/LAPACK libraries - <a class="reference external" href="http://www.openblas.net/">OpenBLAS</a>, <a class="reference external" href="http://math-atlas.sourceforge.net/">ATLAS</a> and <a class="reference external" href="https://software.intel.com/en-us/intel-mkl">MKL</a>. In this step we install OpenBLAS. You can choose to install ATLAS or MKL.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopenblas-dev liblapack-dev
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install OpenCV.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="http://opencv.org/">OpenCV</a> for efficient image loading and augmentation operations.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopencv-dev
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Download MXNet sources and build MXNet core shared library. You can clone the repository as described in the following code block, or you may try the <a href="download.html">download links</a> for your desired MXNet version.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ git clone --recursive https://github.com/apache/incubator-mxnet
+$ <span class="nb">cd</span> incubator-mxnet
+$ make -j <span class="k">$(</span>nproc<span class="k">)</span> <span class="nv">USE_OPENCV</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_BLAS</span><span class="o">=</span>openblas
+</pre></div>
+</div>
+<p><em>Note</em> - USE_OPENCV and USE_BLAS are make file flags to set compilation options to use OpenCV and BLAS library. You can explore and use more compilation options in <code class="docutils literal"><span class="pre">make/config.mk</span></code>.</p>
+<p><br/></p>
+<p><strong>Build and install the MXNet R binding</strong></p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ make rpkg
+$ R CMD INSTALL mxnet_current_r.tar.gz
+</pre></div>
+</div>
+</div> <!-- END of CPU --><div class="gpu"><p>The following installation instructions have been tested on Ubuntu 14.04 and 16.04.</p>
+<p><strong>Prerequisites</strong></p>
+<p>Install the following NVIDIA libraries to setup <em>MXNet</em> with GPU support:</p>
+<ol class="simple">
+<li>Install CUDA 9.0 following the NVIDIA’s <a class="reference external" href="http://docs.nvidia.com/cuda/cuda-installation-guide-linux/">installation guide</a>.</li>
+<li>Install cuDNN 7 for CUDA 9.0 following the NVIDIA’s <a class="reference external" href="https://developer.nvidia.com/cudnn">installation guide</a>. You may need to register with NVIDIA for downloading the cuDNN library.</li>
+</ol>
+<p><strong>Note:</strong> Make sure to add CUDA install path to <code class="docutils literal"><span class="pre">LD_LIBRARY_PATH</span></code>.</p>
+<p>Example - <em>export LD_LIBRARY_PATH=/usr/local/cuda/lib64/:$LD_LIBRARY_PATH</em></p>
+<p><br/></p>
+<p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings.</li>
+</ol>
+<p><strong>Minimum Requirements</strong></p>
+<ol class="simple">
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">GCC 4.8</a> or later to compile C++ 11.</li>
+<li><a class="reference external" href="https://www.gnu.org/software/make/">GNU Make</a></li>
+</ol>
+<p><br/></p>
+<p><strong>Build the MXNet core shared library</strong></p>
+<p><strong>Step 1</strong> Install build tools and git.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get update
+$ sudo apt-get install -y build-essential git
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install OpenBLAS.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="https://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprograms">BLAS</a> and <a class="reference external" href="https://en.wikipedia.org/wiki/LAPACK">LAPACK</a> libraries for accelerated numerical computations on CPU machine. There are several flavors of BLAS/LAPACK libraries - <a class="reference external" href="http://www.openblas.net/">OpenBLAS</a>, <a class="reference external" href="http://math-atlas.sourceforge.net/">ATLAS</a> and <a class="reference external" href="https://software.intel.com/en-us/intel-mkl">MKL</a>. In this step we install OpenBLAS. You can choose to install ATLAS or MKL.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopenblas-dev liblapack-dev
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install OpenCV.</p>
+<p><em>MXNet</em> uses <a class="reference external" href="http://opencv.org/">OpenCV</a> for efficient image loading and augmentation operations.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ sudo apt-get install -y libopencv-dev
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Download MXNet sources and build MXNet core shared library. You can clone the repository as described in the following code block, or you may try the <a href="download.html">download links</a> for your desired MXNet version.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ git clone --recursive https://github.com/apache/incubator-mxnet
+$ <span class="nb">cd</span> incubator-mxnet
+$ make -j <span class="k">$(</span>nproc<span class="k">)</span> <span class="nv">USE_OPENCV</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_BLAS</span><span class="o">=</span>openblas <span class="nv">USE_CUDA</span><span class="o">=</span><span class="m">1</span> <span class="nv">USE_CUDA_PATH</span><span class="o">=</span>/usr/local/cuda <span class="nv">USE_CUDNN</span><span class="o">=</span><span class="m">1</span>
+</pre></div>
+</div>
+<p><em>Note</em> - USE_OPENCV, USE_BLAS, USE_CUDA, USE_CUDA_PATH AND USE_CUDNN are make file flags to set compilation options to use OpenCV, OpenBLAS, CUDA and cuDNN libraries. You can explore and use more compilation options in <code class="docutils literal"><span class="pre">make/config.mk</span></code>. Make sure to set USE_CUDA_PATH to right CUDA installation path. In most cases it is - <em>/usr/local/cuda</em>.</p>
+<p><br/></p>
+<p><strong>Build and install the MXNet R binding</strong></p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ make rpkg
+$ R CMD INSTALL mxnet_current_r.tar.gz
+</pre></div>
+</div>
+</div> <!-- END of GPU -->
+</div> <!-- END of R --><div class="scala julia perl">
+<div class="cpu gpu"><p>Follow the installation instructions <a class="reference internal" href="ubuntu_setup.html"><span class="doc">in this guide</span></a> to set up MXNet.</p>
+</div> <!-- End of cpu gpu -->
+</div> <!-- End of scala julia perl -->
+</div> <!-- END - Linux --><!-- START - MacOS Python CPU Installation Instructions --><div class="macos">
+<div class="python">
+<div class="cpu"><p>The following installation instructions have been tested on OSX Sierra and El Capitan.</p>
+<div class="pip">
+<br/><p><strong>Step 1</strong> Install prerequisites - Homebrew, python development tools.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="c1"># Install Homebrew</span>
+$ /usr/bin/ruby -e <span class="s2">"</span><span class="k">$(</span>curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install<span class="k">)</span><span class="s2">"</span>
+$ <span class="nb">export</span> <span class="nv">PATH</span><span class="o">=</span>/usr/local/bin:/usr/local/sbin:<span class="nv">$PATH</span>
+
+<span class="c1"># Install python development tools - python2.7, pip, python-setuptools</span>
+$ brew install python
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install MXNet with OpenBLAS acceleration.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install --upgrade pip
+$ pip install --upgrade setuptools
+</pre></div>
+</div>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet --pre
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ brew install graphviz
+$ pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div> <!-- END of pip --><div class="virtualenv">
+<br/><p><strong>Step 1</strong> Install prerequisites - Homebrew, python development tools.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="c1"># Install Homebrew</span>
+$ /usr/bin/ruby -e <span class="s2">"</span><span class="k">$(</span>curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install<span class="k">)</span><span class="s2">"</span>
+$ <span class="nb">export</span> <span class="nv">PATH</span><span class="o">=</span>/usr/local/bin:/usr/local/sbin:<span class="nv">$PATH</span>
+
+<span class="c1"># Install python development tools - python2.7, pip, python-setuptools</span>
+$ brew install python
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install virtualenv for macOS.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install virtualenv
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Create and activate virtualenv environment for MXNet.</p>
+<p>Following command creates a virtualenv environment at <code class="docutils literal"><span class="pre">~/mxnet</span></code> directory. However, you can choose any directory by replacing <code class="docutils literal"><span class="pre">~/mxnet</span></code> with a directory of your choice.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ virtualenv --system-site-packages ~/mxnet
+</pre></div>
+</div>
+<p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Install MXNet in the active virtualenv environment.</p>
+<p>Installing <em>MXNet</em> with pip requires a latest version of <code class="docutils literal"><span class="pre">pip</span></code>. Install the latest version of <code class="docutils literal"><span class="pre">pip</span></code> by issuing the following command.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install --upgrade pip
+<span class="o">(</span>mxnet<span class="o">)</span>$ pip install --upgrade setuptools
+</pre></div>
+</div>
+<p>Install <em>MXNet</em> with OpenBLAS acceleration.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$ pip install mxnet --pre
+</pre></div>
+</div>
+<p><strong>Step 5</strong> Install <a class="reference external" href="http://www.graphviz.org/">Graphviz</a>. (Optional, needed for graph visualization using <code class="docutils literal"><span class="pre">mxnet.viz</span></code> package).</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ brew install graphviz
+<span class="o">(</span>mxnet<span class="o">)</span>$ pip install graphviz
+</pre></div>
+</div>
+<p><strong>Step 6</strong> Validate the installation by running simple <em>MXNet</em> code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+<p><strong>Note</strong> You can read more about virtualenv <a class="reference external" href="https://virtualenv.pypa.io/en/stable/userguide/">here</a>.</p>
+</div> <!-- End of virtualenv --><div class="docker">
+<br/><p>Docker images with <em>MXNet</em> are available at <a class="reference external" href="https://hub.docker.com/r/mxnet/">Docker Hub</a>.</p>
+<p><strong>Step 1</strong> Install Docker on your machine by following the <a class="reference external" href="https://docs.docker.com/docker-for-mac/install/#install-and-run-docker-for-mac">docker installation instructions</a>.</p>
+<p><em>Note</em> - You can install Community Edition (CE) to get started with <em>MXNet</em>.</p>
+<p><strong>Step 2</strong> Pull the MXNet docker image.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker pull mxnet/python
+</pre></div>
+</div>
+<p>You can list docker images to see if mxnet/python docker image pull was successful.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker images
+
+REPOSITORY TAG IMAGE ID CREATED SIZE
+mxnet/python latest 00d026968b3c <span class="m">3</span> weeks ago <span class="m">1</span>.41 GB
+</pre></div>
+</div>
+<p><strong>Step 4</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div> <!-- END of docker --><div class="build-from-source">
+<br/><p><strong>Prerequisites</strong></p>
+<p>If not already installed, <a class="reference external" href="https://developer.apple.com/xcode/">download and install Xcode</a> (or <a class="reference external" href="https://itunes.apple.com/us/app/xcode/id497799835">insall it from the App Store</a>) for macOS. <a class="reference external" href="https://en.wikipedia.org/wiki/Xcode">Xcode</a> is an integrated development environment for macOS containing a suite of software development tools like C/C++ compilers, BLAS library and more.</p>
+<p><br/></p>
+<p>Building <em>MXNet</em> from source is a 2 step process.</p>
+<ol class="simple">
+<li>Build the <em>MXNet</em> core shared library, <code class="docutils literal"><span class="pre">libmxnet.so</span></code>, from the C++ sources.</li>
+<li>Build the language specific bindings. Example - Python bindings, Scala bindings.</li>
+</ol>
+<p>Make sure you have installed Xcode before proceeding further.</p>
+<p><br/></p>
+<p>All the instructions to build <em>MXNet</em> core shared library and <em>MXNet</em> Python bindings are compiled as one helper <em>bash</em> script. You can use <a class="reference external" href="https://raw.githubusercontent.com/dmlc/mxnet/master/setup-utils/install-mxnet-osx-python.sh">this bash script</a> to build <em>MXNet</em> for Python, from source, on macOS.</p>
+<p><strong>Step 1</strong> Download the bash script for building MXNet from source.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ curl -O https://raw.githubusercontent.com/dmlc/mxnet/master/setup-utils/install-mxnet-osx-python.sh
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Run the script to get latest MXNet source and build.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="c1"># Make the script executable</span>
+$ chmod <span class="m">744</span> install-mxnet-osx-python.sh
+
+<span class="c1"># Run the script. It takes around 5 mins.</span>
+$ bash install-mxnet-osx-python.sh
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Validate the installation by running simple MXNet code described <a class="reference external" href="#validate-mxnet-installation">here</a>.</p>
+</div> <!-- END of build from source -->
+</div> <!-- END of CPU --><!-- START - Mac OS Python GPU Installation Instructions -->
+<div class="gpu">
+<div class="pip virtualenv docker">
+</br><p>Try the <strong>Build from Source</strong> option for now.</p>
+</div><div class="build-from-source"><p><strong>Step 1</strong> Install prerequisites - Homebrew, python development tools.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="c1"># Install Homebrew</span>
+$ /usr/bin/ruby -e <span class="s2">"</span><span class="k">$(</span>curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install<span class="k">)</span><span class="s2">"</span>
+$ <span class="nb">export</span> <span class="nv">PATH</span><span class="o">=</span>/usr/local/bin:/usr/local/sbin:<span class="nv">$PATH</span>
+
+<span class="c1"># Install python development tools - python2.7, pip, python-setuptools</span>
+$ brew install python pkg-config graphviz
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install optional components - OpenCV</p>
+<p>If you want to use OpenCV you should install it first, then build MXNet with the <code class="docutils literal"><span class="pre">USE_OPENCV=1</span></code> option in the later steps.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>brew tap homebrew/science
+brew install opencv
+</pre></div>
+</div>
+<p><strong>Step 3</strong> Install CUDA and cuDNN</p>
+<p>The following instructions are for CUDA 9.1 and cuDNN 7 for macOS 10.12+ and a CUDA-capable GPU. They summarize confirmed successful builds in <a class="reference external" href="https://github.com/apache/incubator-mxnet/issues/9217">#9217</a>.
+Alternatively, you may follow the <a class="reference external" href="https://docs.nvidia.com/cuda/cuda-installation-guide-mac-os-x/index.html">CUDA installation instructions for macOS</a>.</p>
+<ol class="simple">
+<li><a class="reference external" href="https://developer.apple.com/download/more/">Download Xcode 8.3.3 from Apple</a>. This is the version <a class="reference external" href="https://docs.nvidia.com/cuda/cuda-installation-guide-mac-os-x/index.html">NVIDIA specifies in its instructions for macOS</a>. Unzip and rename to <code class="docutils literal"><span class="pre">Xcode8.3.3.app</span></code>.</li>
+<li>Run <code class="docutils literal"><span class="pre">sudo</span> <span class="pre">xcode-select</span> <span class="pre">-s</span> <span class="pre">/Applications/Xcode8.3.3.app</span></code> or to wherever you have placed Xcode.</li>
+<li>Run <code class="docutils literal"><span class="pre">xcode-select</span> <span class="pre">--install</span></code> to install all command line tools, compilers, etc.</li>
+<li>Run <code class="docutils literal"><span class="pre">sudo</span> <span class="pre">xcodebuild</span> <span class="pre">-license</span> <span class="pre">accept</span></code> to accept Xcode’s licensing terms.</li>
+<li>Install CUDA for macOS. Specific steps are provided in NVIDIA’s <a class="reference external" href="https://docs.nvidia.com/cuda/cuda-installation-guide-mac-os-x/index.html#installation">CUDA installation instructions</a>.</li>
+<li><a class="reference external" href="http://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html#download-mac">Download</a> and <a class="reference external" href="http://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html#installmac">install</a> cuDNN for macOS. You will need to <a class="reference external" href="https://developer.nvidia.com/accelerated-computing-developer">create a free developer account</a> with NVIDIA prior to getting the download link.</li>
+</ol>
+<p><strong>Step 4</strong> Build MXNet</p>
+<ol>
+<li><p class="first">Run <code class="docutils literal"><span class="pre">git</span> <span class="pre">clone</span> <span class="pre">--recursive</span> <span class="pre">https://github.com/apache/incubator-mxnet.git</span> <span class="pre">mxnet</span></code> to get the latest version.</p>
+</li>
+<li><p class="first">Run <code class="docutils literal"><span class="pre">cd</span> <span class="pre">mxnet</span></code>.</p>
+</li>
+<li><p class="first">Edit the <code class="docutils literal"><span class="pre">make/osx.mk</span></code> file to set the following parameters:</p>
+<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">USE_CUDA</span> <span class="o">=</span> <span class="mi">1</span>
+<span class="n">USE_CUDA_PATH</span> <span class="o">=</span> <span class="o">/</span><span class="n">usr</span><span class="o">/</span><span class="n">local</span><span class="o">/</span><span class="n">cuda</span>
+<span class="n">USE_CUDNN</span> <span class="o">=</span> <span class="mi">1</span>
+<span class="n">USE_OPENCV</span> <span class="o">=</span> <span class="mi">0</span> <span class="c1"># set to 1 if you want to build with OpenCV</span>
+</pre></div>
+</div>
+</li>
+<li><p class="first">Copy the <code class="docutils literal"><span class="pre">make/osx.mk</span></code> to <code class="docutils literal"><span class="pre">config.mk</span></code></p>
+</li>
+<li><p class="first">Run <code class="docutils literal"><span class="pre">make</span></code>. If you previously attempted to compile you might want to do <code class="docutils literal"><span class="pre">make</span> <span class="pre">clean_all</span></code> first. You can also run <code class="docutils literal"><span class="pre">make</span> <span class="pre">-j</span></code> with the number of processors you have to compile with multithreading. There’ll be plenty of warnings, but there should be no errors.</p>
+</li>
+<li><p class="first">Once finished, you should have a file called <code class="docutils literal"><span class="pre">libmxnet.so</span></code> in <code class="docutils literal"><span class="pre">lib/</span></code>.</p>
+</li>
+<li><p class="first">Do <code class="docutils literal"><span class="pre">cd</span> <span class="pre">python</span></code>.</p>
+</li>
+<li><p class="first">Run <code class="docutils literal"><span class="pre">sudo</span> <span class="pre">pip</span> <span class="pre">install</span> <span class="pre">-e</span> <span class="pre">.</span></code> <strong>Note</strong>: the <code class="docutils literal"><span class="pre">.</span></code> is part of the command.</p>
+</li>
+</ol>
+</div> <!-- END of build from source -->
+</div> <!-- END of GPU -->
+</div> <!-- END of Python --><!-- START - MacOS R CPU Installation Instructions --><div class="r">
+<div class="cpu"><p>The CPU version of MXNet R package can be installed in R like other packages</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span>cran <span class="o"><-</span> <span class="kp">getOption</span><span class="p">(</span><span class="s">"repos"</span><span class="p">)</span>
+cran<span class="p">[</span><span class="s">"dmlc"</span><span class="p">]</span> <span class="o"><-</span> <span class="s">"https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/R/CRAN/"</span>
+<span class="kp">options</span><span class="p">(</span>repos <span class="o">=</span> cran<span class="p">)</span>
+install.packages<span class="p">(</span><span class="s">"mxnet"</span><span class="p">)</span>
+</pre></div>
+</div>
+</div> <!-- END of CPU --><div class="gpu"><p>Will be available soon.</p>
+</div> <!-- END of GPU -->
+</div> <!-- END of R --><div class="scala julia perl">
+<div class="cpu gpu"><p>Follow the installation instructions <a class="reference internal" href="osx_setup.html"><span class="doc">in this guide</span></a> to set up MXNet.</p>
+</div> <!-- End of cpu gpu -->
+</div> <!-- End of scala julia perl -->
+</div> <!-- END - Mac OS --><div class="windows">
+<div class="python">
+<div class="cpu">
+<div class="pip"><p><br/></p>
+<p><strong>Step 1</strong> Install Python.</p>
+<p><a class="reference external" href="https://www.anaconda.com/download/">Anaconda</a> is recommended.</p>
+<p><strong>Step 2</strong> Install <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet --pre
+</pre></div>
+</div>
+</div> <!-- End of pip --><div class="virtualenv docker build-from-source"><p>Follow the installation instructions <a class="reference internal" href="windows_setup.html"><span class="doc">in this guide</span></a> to set up MXNet.</p>
+</div> <!-- End of virtualenv docker build-from-source -->
+</div> <!-- End of CPU --><div class="gpu">
+<div class="pip"><p><br/></p>
+<p><strong>Step 1</strong> Install Python.</p>
+<p><a class="reference external" href="https://www.anaconda.com/download/">Anaconda</a> is recommended.</p>
+<p><strong>Step 2</strong> Install <em>MXNet</em> with GPU support using CUDA 9.0.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet-cu90 --pre
+</pre></div>
+</div>
+<p>Refer to <a class="reference external" href="https://github.com/apache/incubator-mxnet/issues/8671">#8671</a> for status on CUDA 9.1 support.</p>
+</div>
+<div class="build-from-source">
+<br/><p>We provide both options to build and install MXNet yourself using <a class="reference external" href="https://www.visualstudio.com/downloads/">Microsoft Visual Studio 2017</a>, and <a class="reference external" href="https://www.visualstudio.com/vs/older-downloads/">Microsoft Visual Studio 2015</a>.</p>
+<p><strong>Option 1</strong></p>
+<p>To build and install MXNet yourself using <a class="reference external" href="https://www.visualstudio.com/downloads/">Microsoft Visual Studio 2017</a>, you need the following dependencies. Install the required dependencies:</p>
+<ol class="simple">
+<li>If <a class="reference external" href="https://www.visualstudio.com/downloads/">Microsoft Visual Studio 2017</a> is not already installed, download and install it. You can download and install the free community edition.</li>
+<li>Download and install <a class="reference external" href="https://cmake.org/files/v3.11/cmake-3.11.0-rc4-win64-x64.msi">CMake</a> if it is not already installed.</li>
+<li>Download and install <a class="reference external" href="https://sourceforge.net/projects/opencvlibrary/files/opencv-win/3.4.1/opencv-3.4.1-vc14_vc15.exe/download">OpenCV</a>.</li>
+<li>Unzip the OpenCV package.</li>
+<li>Set the environment variable <code class="docutils literal"><span class="pre">OpenCV_DIR</span></code> to point to the <code class="docutils literal"><span class="pre">OpenCV</span> <span class="pre">build</span> <span class="pre">directory</span></code> (e.g., <code class="docutils literal"><span class="pre">OpenCV_DIR</span> <span class="pre">=</span> <span class="pre">C:\utils\opencv\build</span></code>).</li>
+<li>If you don’t have the Intel Math Kernel Library (MKL) installed, download and install <a class="reference external" href="https://sourceforge.net/projects/openblas/files/v0.2.20/OpenBLAS%200.2.20%20version.zip/download">OpenBlas</a>.</li>
+<li>Set the environment variable <code class="docutils literal"><span class="pre">OpenBLAS_HOME</span></code> to point to the <code class="docutils literal"><span class="pre">OpenBLAS</span></code> directory that contains the <code class="docutils literal"><span class="pre">include</span></code> and <code class="docutils literal"><span class="pre">lib</span></code> directories (e.g., <code class="docutils literal"><span class="pre">OpenBLAS_HOME</span> <span class="pre">=</span> <span class="pre">C:\utils\OpenBLAS</span></code>).</li>
+<li>Download and install CUDA: Install <a class="reference external" href="https://developer.nvidia.com/cuda-downloads?target_os=Windows&target_arch=x86_64&target_version=10&target_type=exelocal">CUDA</a>, and Download the base installer (e.g., <code class="docutils literal"><span class="pre">cuda_9.1.85_win10.exe</span></code>).</li>
+<li>Download and install cuDNN. To get access to the download link, register as an NVIDIA community user. Then Follow the <a class="reference external" href="http://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html#install-windows">link</a> to install the cuDNN.</li>
+<li>Download and install <a class="reference external" href="https://git-for-windows.github.io/">git</a>.</li>
+</ol>
+<p>After you have installed all of the required dependencies, build the MXNet source code:</p>
+<ol class="simple">
+<li>Start <code class="docutils literal"><span class="pre">cmd</span></code> in windows.</li>
+<li>Download the MXNet source code from GitHub by using following command:</li>
+</ol>
+<div class="highlight-r"><div class="highlight"><pre><span></span>cd C<span class="o">:</span>\
+git clone https<span class="o">://</span>github.com<span class="o">/</span>apache<span class="o">/</span>incubator<span class="o">-</span>mxnet.git <span class="o">--</span>recursive
+</pre></div>
+</div>
+<ol class="simple">
+<li>Follow <a class="reference external" href="https://docs.microsoft.com/en-us/visualstudio/install/modify-visual-studio">this link</a> to modify <code class="docutils literal"><span class="pre">Individual</span> <span class="pre">components</span></code>, and check <code class="docutils literal"><span class="pre">VC++</span> <span class="pre">2017</span> <span class="pre">version</span> <span class="pre">15.4</span> <span class="pre">v14.11</span> <span class="pre">toolset</span></code>, and click <code class="docutils literal"><span class="pre">Modify</span></code>.</li>
+<li>Change the version of the Visual studio 2017 to v14.11 using the following command (by default the VS2017 is installed in the following path):</li>
+</ol>
+<div class="highlight-r"><div class="highlight"><pre><span></span><span class="s">"C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Auxiliary\Build\vcvars64.bat"</span> <span class="o">-</span>vcvars_ver<span class="o">=</span><span class="m">14.11</span>
+</pre></div>
+</div>
+<ol class="simple">
+<li>Create a build dir using the following command and go to the directory, for example:</li>
+</ol>
+<div class="highlight-r"><div class="highlight"><pre><span></span>mkdir C<span class="o">:</span>\build
+cd C<span class="o">:</span>\build
+</pre></div>
+</div>
+<ol class="simple">
+<li>CMake the MXNet source code by using following command:</li>
+</ol>
+<div class="highlight-r"><div class="highlight"><pre><span></span>cmake <span class="o">-</span>G <span class="s">"Visual Studio 15 2017 Win64"</span> <span class="o">-</span><span class="bp">T</span> cuda<span class="o">=</span><span class="m">9.1</span><span class="p">,</span>host<span class="o">=</span>x64 <span class="o">-</span>DUSE_CUDA<span class="o">=</span><span class="m">1</span> <span class="o">-</span>DUSE_CUDNN<span class="o">=</span><span class="m">1</span> <span class="o">-</span>DUSE_NVRTC<span class="o">=</span><span class="m">1</span> <span class="o">-</span>DUSE_OPENCV<span class="o">=</span><span class="m">1</span> <span class="o">-</span>DUSE_OPENMP<span class="o">=</span><span class="m">1</span> <span class="o">-</span>DUSE_BLAS<span class="o">=</span>open <span class="o">-</span>DUSE_LAPACK<span class="o">=</span><span class="m">1</span> <span class="o">-</span>DUSE_DIST_KVSTORE<span class="o">=</span><span class="m">0</span> <span class="o">-</span>DCUDA_ARCH_LIST<span class="o">=</span>Common <span class="o">-</span>DCUDA_TOOLSET<span class="o">=</span><span class="m">9.1</span> <span class="o">-</span>DCUDNN_INCLUDE<span class="o">=</span>C<span class="o">:</span>\cuda\include <span class="o">-</span>DCUDNN_LIBRARY<span class="o">=</span>C<span class="o">:</span>\cuda\lib\x64\cudnn.lib <span class="s">"C:\incubator-mxnet"</span>
+</pre></div>
+</div>
+<p>NOTE: make sure the DCUDNN_INCLUDE and DCUDNN_LIBRARY pointing to the “include” and “cudnn.lib” of your CUDA installed location, and the <code class="docutils literal"><span class="pre">C:\incubator-mxnet</span></code> is the location of the source code you just git in the previous step</p>
+<ol class="simple">
+<li>After the CMake successfully completed, compile the the MXNet source code by using following command:</li>
+</ol>
+<div class="highlight-r"><div class="highlight"><pre><span></span>msbuild mxnet.sln <span class="o">/</span>p<span class="o">:</span>Configuration<span class="o">=</span>Release<span class="p">;</span>Platform<span class="o">=</span>x64 <span class="o">/</span>maxcpucount
+</pre></div>
+</div>
+<p><strong>Option 2</strong></p>
+<p>To build and install MXNet yourself using <a class="reference external" href="https://www.visualstudio.com/vs/older-downloads/">Microsoft Visual Studio 2015</a>, you need the following dependencies. Install the required dependencies:</p>
+<ol class="simple">
+<li>If <a class="reference external" href="https://www.visualstudio.com/vs/older-downloads/">Microsoft Visual Studio 2015</a> is not already installed, download and install it. You can download and install the free community edition.</li>
+<li>Download and install <a class="reference external" href="https://cmake.org/">CMake</a> if it is not already installed.</li>
+<li>Download and install <a class="reference external" href="http://sourceforge.net/projects/opencvlibrary/files/opencv-win/3.0.0/opencv-3.0.0.exe/download">OpenCV</a>.</li>
+<li>Unzip the OpenCV package.</li>
+<li>Set the environment variable <code class="docutils literal"><span class="pre">OpenCV_DIR</span></code> to point to the <code class="docutils literal"><span class="pre">OpenCV</span> <span class="pre">build</span> <span class="pre">directory</span></code> (<code class="docutils literal"><span class="pre">C:\opencv\build\x64\vc14</span></code> for example). Also, you need to add the OpenCV bin directory (<code class="docutils literal"><span class="pre">C:\opencv\build\x64\vc14\bin</span></code> for example) to the <code class="docutils literal"><span class="pre">PATH</span></code> variable.</li>
+<li>If you don’t have the Intel Math Kernel Library (MKL) installed, download and install <a class="reference external" href="http://sourceforge.net/projects/openblas/files/v0.2.14/">OpenBlas</a>.</li>
+<li>Set the environment variable <code class="docutils literal"><span class="pre">OpenBLAS_HOME</span></code> to point to the <code class="docutils literal"><span class="pre">OpenBLAS</span></code> directory that contains the <code class="docutils literal"><span class="pre">include</span></code> and <code class="docutils literal"><span class="pre">lib</span></code> directories. Typically, you can find the directory in <code class="docutils literal"><span class="pre">C:\Program</span> <span class="pre">files</span> <span class="pre">(x86)\OpenBLAS\</span></code>.</li>
+<li>Download and install <a class="reference external" href="https://developer.nvidia.com/cuda-downloads?target_os=Windows&target_arch=x86_64">CUDA</a> and <a class="reference external" href="https://developer.nvidia.com/cudnn">cuDNN</a>. To get access to the download link, register as an NVIDIA community user.</li>
+</ol>
+<p>After you have installed all of the required dependencies, build the MXNet source code:</p>
+<ol class="simple">
+<li>Download the MXNet source code from <a class="reference external" href="https://github.com/apache/incubator-mxnet">GitHub</a>.</li>
+<li>Use <a class="reference external" href="https://cmake.org/">CMake</a> to create a Visual Studio solution in <code class="docutils literal"><span class="pre">./build</span></code>.</li>
+<li>In Visual Studio, open the solution file,<code class="docutils literal"><span class="pre">.sln</span></code>, and compile it.
+These commands produce a library called <code class="docutils literal"><span class="pre">mxnet.dll</span></code> in the <code class="docutils literal"><span class="pre">./build/Release/</span></code> or <code class="docutils literal"><span class="pre">./build/Debug</span></code> folder.</li>
+</ol>
+<p><p> </p>
+
+Next, we install the <code class="docutils literal"><span class="pre">graphviz</span></code> library that we use for visualizing network graphs that you build on MXNet. We will also install <a class="reference external" href="http://jupyter.readthedocs.io/">Jupyter Notebook</a> which is used for running MXNet tutorials and examples.</p>
+<ul class="simple">
+<li>Install the <code class="docutils literal"><span class="pre">graphviz</span></code> by downloading the installer from the <a class="reference external" href="https://graphviz.gitlab.io/_pages/Download/Download_windows.html">Graphviz Download Page</a>.
+<strong>Note</strong> Make sure to add the <code class="docutils literal"><span class="pre">graphviz</span></code> executable path to the PATH environment variable. Refer <a class="reference external" href="http://stackoverflow.com/questions/35064304/runtimeerror-make-sure-the-graphviz-executables-are-on-your-systems-path-aft">here for more details</a></li>
+</ul>
+<p><p> </p>
+</p>
+</div> <!-- End of pip -->
+</div> <!-- End of GPU -->
+</div> <!-- End of Python --><!-- START - Windows R CPU Installation Instructions --><div class="r">
+<div class="cpu"><p>The CPU version of MXNet R package can be installed in R like other packages</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span>cran <span class="o"><-</span> <span class="kp">getOption</span><span class="p">(</span><span class="s">"repos"</span><span class="p">)</span>
+cran<span class="p">[</span><span class="s">"dmlc"</span><span class="p">]</span> <span class="o"><-</span> <span class="s">"https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/R/CRAN/"</span>
+<span class="kp">options</span><span class="p">(</span>repos <span class="o">=</span> cran<span class="p">)</span>
+install.packages<span class="p">(</span><span class="s">"mxnet"</span><span class="p">)</span>
+</pre></div>
+</div>
+</div> <!-- END - Windows R CPU --><div class="gpu"><p>The GPU version of MXNet R package can be installed in R like other packages</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span>cran <span class="o"><-</span> <span class="kp">getOption</span><span class="p">(</span><span class="s">"repos"</span><span class="p">)</span>
+cran<span class="p">[</span><span class="s">"dmlc"</span><span class="p">]</span> <span class="o"><-</span> <span class="s">"https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/R/CRAN/GPU"</span>
+<span class="kp">options</span><span class="p">(</span>repos <span class="o">=</span> cran<span class="p">)</span>
+install.packages<span class="p">(</span><span class="s">"mxnet"</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>Alternatively, You can also follow the installation instructions <a class="reference internal" href="windows_setup.html"><span class="doc">in this guide</span></a> to build MXNet from source.</p>
+</div> <!-- END of GPU -->
+</div> <!-- END - Windows R --><div class="scala julia perl">
+<div class="cpu gpu"><p>Follow the installation instructions <a class="reference internal" href="windows_setup.html"><span class="doc">in this guide</span></a> to set up MXNet.</p>
+</div> <!-- End of cpu gpu -->
+</div> <!-- End of scala julia perl -->
+</div> <!-- End of Windows --><!-- START - Cloud Python Installation Instructions --><div class="cloud"><p>AWS Marketplace distributes Deep Learning AMIs (Amazon Machine Image) with MXNet pre-installed. You can launch one of these Deep Learning AMIs by following instructions in the <a class="reference external" href="http://docs.aws.amazon.com/dlami/latest/devguide/what-is-dlami.html">AWS Deep Learning AMI Developer Guide</a>.</p>
+<p>You can also run distributed deep learning with <em>MXNet</em> on AWS using <a class="reference external" href="https://github.com/awslabs/deeplearning-cfn/blob/master/README.md">Cloudformation Template</a>.</p>
+</div> <!-- END - Cloud Python Installation Instructions --><div class="devices">
+ <div class="raspberry-pi"><p>MXNet supports the Debian based Raspbian ARM based operating system so you can run MXNet on Raspberry Pi Devices.</p>
+<p>These instructions will walk through how to build MXNet for the Raspberry Pi and install the Python bindings for the library.</p>
+<p>The complete MXNet library and its requirements can take almost 200MB of RAM, and loading large models with the library can take over 1GB of RAM. Because of this, we recommend running MXNet on the Raspberry Pi 3 or an equivalent device that has more than 1 GB of RAM and a Secure Digital (SD) card that has at least 4 GB of free memory.</p>
+<p><strong>Install MXNet</strong></p>
+<p>Installing MXNet is a two-step process:</p>
+<ol class="simple">
+<li>Build the shared library from the MXNet C++ source code.</li>
+<li>Install the supported language-specific packages for MXNet.</li>
+</ol>
+<p><strong>Step 1</strong> Build the Shared Library</p>
+<p>On Raspbian versions Wheezy and later, you need the following dependencies:</p>
+<ul class="simple">
+<li>Git (to pull code from GitHub)</li>
+<li>libblas (for linear algebraic operations)</li>
+<li>libopencv (for computer vision operations. This is optional if you want to save RAM and Disk Space)</li>
+<li>A C++ compiler that supports C++ 11. The C++ compiler compiles and builds MXNet source code. Supported compilers include the following:</li>
+<li><a class="reference external" href="https://gcc.gnu.org/gcc-4.8/">G++ (4.8 or later)</a></li>
+</ul>
+<p>Install these dependencies using the following commands in any directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> sudo apt-get update
+ sudo apt-get -y install git cmake build-essential g++-4.8 c++-4.8 liblapack* libblas* libopencv*
+</pre></div>
+</div>
+<p>Clone the MXNet source code repository using the following <code class="docutils literal"><span class="pre">git</span></code> command in your home directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> git clone https://github.com/apache/incubator-mxnet.git --recursive
+ <span class="nb">cd</span> incubator-mxnet
+</pre></div>
+</div>
+<p>If you aren’t processing images with MXNet on the Raspberry Pi, you can minimize the size of the compiled library by building MXNet without the Open Source Computer Vision (OpenCV) library with the following commands:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> <span class="nb">export</span> <span class="nv">USE_OPENCV</span> <span class="o">=</span> <span class="m">0</span>
+ make
+</pre></div>
+</div>
+<p>Otherwise, you can build the complete MXNet library with the following command:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> make
+</pre></div>
+</div>
+<p>Executing either of these commands start the build process, which can take up to a couple hours, and creates a file called <code class="docutils literal"><span class="pre">libmxnet.so</span></code> in the mxnet/lib directory.</p>
+<p>If you are getting build errors in which the compiler is being killed, it is likely that the compiler is running out of memory (especially if you are on Raspberry Pi 1, 2 or Zero, which have less than 1GB of RAM), this can often be rectified by increasing the swapfile size on the Pi by editing the file /etc/dphys-swapfile and changing the line CONF_SWAPSIZE=100 to CONF_SWAPSIZE=1024, then running:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> sudo /etc/init.d/dphys-swapfile stop
+ sudo /etc/init.d/dphys-swapfile start
+ free -m <span class="c1"># to verify the swapfile size has been increased</span>
+</pre></div>
+</div>
+<p><strong>Step 2</strong> Install MXNet Python Bindings</p>
+<p>To install Python bindings run the following commands in the MXNet directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> <span class="nb">cd</span> python
+ pip install --upgrade pip
+ pip install -e .
+</pre></div>
+</div>
+<p>Note that the <code class="docutils literal"><span class="pre">-e</span></code> flag is optional. It is equivalent to <code class="docutils literal"><span class="pre">--editable</span></code> and means that if you edit the source files, these changes will be reflected in the package installed.</p>
+<p>You are now ready to run MXNet on your Raspberry Pi device. You can get started by following the tutorial on <a class="reference external" href="http://mxnet.io/tutorials/embedded/wine_detector.html">Real-time Object Detection with MXNet On The Raspberry Pi</a>.</p>
+<p><em>Note - Because the complete MXNet library takes up a significant amount of the Raspberry Pi’s limited RAM, when loading training data or large models into memory, you might have to turn off the GUI and terminate running processes to free RAM.</em></p>
+</div> <!-- End of raspberry pi --><div class="nvidia-jetson-tx2"><p>MXNet supports the Ubuntu Arch64 based operating system so you can run MXNet on NVIDIA Jetson Devices.</p>
+<p>These instructions will walk through how to build MXNet for the Pascal based <a class="reference external" href="http://www.nvidia.com/object/embedded-systems-dev-kits-modules.html">NVIDIA Jetson TX2</a> and install the corresponding python language bindings.</p>
+<p>For the purposes of this install guide we will assume that CUDA is already installed on your Jetson device.</p>
+<p><strong>Install MXNet</strong></p>
+<p>Installing MXNet is a two-step process:</p>
+<ol class="simple">
+<li>Build the shared library from the MXNet C++ source code.</li>
+<li>Install the supported language-specific packages for MXNet.</li>
+</ol>
+<p><strong>Step 1</strong> Build the Shared Library</p>
+<p>You need the following additional dependencies:</p>
+<ul class="simple">
+<li>Git (to pull code from GitHub)</li>
+<li>libatlas (for linear algebraic operations)</li>
+<li>libopencv (for computer vision operations)</li>
+<li>python pip (to load relevant python packages for our language bindings)</li>
+</ul>
+<p>Install these dependencies using the following commands in any directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> sudo apt-get update
+ sudo apt-get -y install git build-essential libatlas-base-dev libopencv-dev graphviz python-pip
+ sudo pip install pip --upgrade
+ sudo pip install setuptools numpy --upgrade
+ sudo pip install graphviz jupyter
+</pre></div>
+</div>
+<p>Clone the MXNet source code repository using the following <code class="docutils literal"><span class="pre">git</span></code> command in your home directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> git clone https://github.com/apache/incubator-mxnet.git --recursive
+ <span class="nb">cd</span> incubator-mxnet
+</pre></div>
+</div>
+<p>Edit the Makefile to install the MXNet with CUDA bindings to leverage the GPU on the Jetson:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> cp make/config.mk .
+ <span class="nb">echo</span> <span class="s2">"USE_CUDA=1"</span> >> config.mk
+ <span class="nb">echo</span> <span class="s2">"USE_CUDA_PATH=/usr/local/cuda"</span> >> config.mk
+ <span class="nb">echo</span> <span class="s2">"USE_CUDNN=1"</span> >> config.mk
+</pre></div>
+</div>
+<p>Edit the Mshadow Makefile to ensure MXNet builds with Pascal’s hardware level low precision acceleration by editing 3rdparty/mshadow/make/mshadow.mk and adding the following after line 122:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="nv">MSHADOW_CFLAGS</span> <span class="o">+=</span> -DMSHADOW_USE_PASCAL<span class="o">=</span><span class="m">1</span>
+</pre></div>
+</div>
+<p>Now you can build the complete MXNet library with the following command:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> make -j <span class="k">$(</span>nproc<span class="k">)</span>
+</pre></div>
+</div>
+<p>Executing this command creates a file called <code class="docutils literal"><span class="pre">libmxnet.so</span></code> in the mxnet/lib directory.</p>
+<p><strong>Step 2</strong> Install MXNet Python Bindings</p>
+<p>To install Python bindings run the following commands in the MXNet directory:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> <span class="nb">cd</span> python
+ pip install --upgrade pip
+ pip install -e .
+</pre></div>
+</div>
+<p>Note that the <code class="docutils literal"><span class="pre">-e</span></code> flag is optional. It is equivalent to <code class="docutils literal"><span class="pre">--editable</span></code> and means that if you edit the source files, these changes will be reflected in the package installed.</p>
+<p>Add the mxnet folder to the path:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span> <span class="nb">cd</span> ..
+ <span class="nb">export</span> <span class="nv">MXNET_HOME</span><span class="o">=</span><span class="k">$(</span><span class="nb">pwd</span><span class="k">)</span>
+ <span class="nb">echo</span> <span class="s2">"export PYTHONPATH=</span><span class="nv">$MXNET_HOME</span><span class="s2">/python:</span><span class="nv">$PYTHONPATH</span><span class="s2">"</span> >> ~/.bashrc
+ <span class="nb">source</span> ~/.bashrc
+</pre></div>
+</div>
+<p>You are now ready to run MXNet on your NVIDIA Jetson TX2 device.</p>
+</div> <!-- End of jetson -->
+</div> <!-- End of devices --><!-- This # tag restarts the page and allows reuse
+ of the div classes for validation sections, etc --></div>
+<div class="section" id="validate-mxnet-installation">
+<span id="validate-mxnet-installation"></span><h1>Validate MXNet Installation<a class="headerlink" href="#validate-mxnet-installation" title="Permalink to this headline">¶</a></h1>
+<div class="linux macos">
+ <div class="python">
+ <div class="cpu"><div class="pip build-from-source"><p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+</div><div class="docker"><p>Launch a Docker container with <code class="docutils literal"><span class="pre">mxnet/python</span></code> image and run example <em>MXNet</em> python program on the terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ docker run -it mxnet/python bash <span class="c1"># Use sudo if you skip Step 2 in the installation instruction</span>
+
+<span class="c1"># Start a python terminal</span>
+root@4919c4f58cac:/# python
+</pre></div>
+</div>
+</div><div class="virtualenv"><p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+</div><p>Run a short <em>MXNet</em> python program to create a 2X3 matrix of ones, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">mxnet</span> <span class="kn">as</span> <span class="nn">mx</span>
+<span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">))</span>
+<span class="gp">>>> </span><span class="n">b</span> <span class="o">=</span> <span class="n">a</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+<span class="gp">>>> </span><span class="n">b</span><span class="o">.</span><span class="n">asnumpy</span><span class="p">()</span>
+<span class="go">array([[ 3., 3., 3.],</span>
+<span class="go"> [ 3., 3., 3.]], dtype=float32)</span>
+</pre></div>
+</div>
+</div>
+</div>
+</div><!-- Validate Windows CPU pip install --><div class="windows">
+ <div class="python">
+ <div class="cpu">
+ <div class="pip"><p>Run a short <em>MXNet</em> python program to create a 2X3 matrix of ones, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">mxnet</span> <span class="kn">as</span> <span class="nn">mx</span>
+<span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">))</span>
+<span class="gp">>>> </span><span class="n">b</span> <span class="o">=</span> <span class="n">a</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+<span class="gp">>>> </span><span class="n">b</span><span class="o">.</span><span class="n">asnumpy</span><span class="p">()</span>
+<span class="go">array([[ 3., 3., 3.],</span>
+<span class="go"> [ 3., 3., 3.]], dtype=float32)</span>
+</pre></div>
+</div>
+</div>
+</div>
+</div>
+</div><!-- Mac OS GPU installation validation --><div class="macos">
+ <div class="python">
+ <div class="gpu"><div class="pip virtualenv docker">
+</br><p>Will be available soon.</p>
+</div><div class="build-from-source">
+</br><p>From the MXNet root directory run: <code class="docutils literal"><span class="pre">python</span> <span class="pre">example/image-classification/train_mnist.py</span> <span class="pre">--network</span> <span class="pre">lenet</span> <span class="pre">--gpus</span> <span class="pre">0</span></code> to test GPU training.</p>
+</div>
+</div>
+</div>
+</div><!-- Windows GPU installation validation --><div class="windows">
+ <div class="python">
+ <div class="gpu"><div class="virtualenv docker">
+</br><p>Will be available soon.</p>
+</div><div class="pip build-from-source">
+</br><p>From the MXNet root directory run: <code class="docutils literal"><span class="pre">python</span> <span class="pre">example/image-classification/train_mnist.py</span> <span class="pre">--network</span> <span class="pre">lenet</span> <span class="pre">--gpus</span> <span class="pre">0</span></code> to test GPU training.</p>
+</div>
+</div>
+</div>
+</div><!-- Validation for GPU machines --><div class="linux">
+ <div class="python">
+ <div class="gpu"><div class="pip build-from-source"><p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+</div><div class="docker"><p>Launch a NVIDIA Docker container with <code class="docutils literal"><span class="pre">mxnet/python:gpu</span></code> image and run example <em>MXNet</em> python program on the terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ nvidia-docker run -it mxnet/python:gpu bash <span class="c1"># Use sudo if you skip Step 2 in the installation instruction</span>
+
+<span class="c1"># Start a python terminal</span>
+root@4919c4f58cac:/# python
+</pre></div>
+</div>
+</div><div class="virtualenv"><p>Activate the virtualenv environment created for <em>MXNet</em>.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">source</span> ~/mxnet/bin/activate
+</pre></div>
+</div>
+<p>After activating the environment, you should see the prompt as below.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">(</span>mxnet<span class="o">)</span>$
+</pre></div>
+</div>
+<p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+</div><p>Run a short <em>MXNet</em> python program to create a 2X3 matrix of ones <em>a</em> on a <em>GPU</em>, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3. We use <em>mx.gpu()</em>, to set <em>MXNet</em> context to be GPUs.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">mxnet</span> <span class="kn">as</span> <span class="nn">mx</span>
+<span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">mx</span><span class="o">.</span><span class="n">gpu</span><span class="p">())</span>
+<span class="gp">>>> </span><span class="n">b</span> <span class="o">=</span> <span class="n">a</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+<span class="gp">>>> </span><span class="n">b</span><span class="o">.</span><span class="n">asnumpy</span><span class="p">()</span>
+<span class="go">array([[ 3., 3., 3.],</span>
+<span class="go"> [ 3., 3., 3.]], dtype=float32)</span>
+</pre></div>
+</div>
+</div>
+</div>
+</div><!-- Linux Clean up -->
+<div class="linux">
+ <div class="python">
+ <div class="cpu"><div class="pip build-from-source"><p>Exit the Python terminal.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><div class="virtualenv"><p>Exit the Python terminal and Deactivate the virtualenv <em>MXNet</em> environment.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">(mxnet)$ deactivate</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><div class="docker"><p>Exit the Python terminal and mxnet/python docker container.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">root@4919c4f58cac:/# exit</span>
+</pre></div>
+</div>
+</div></div>
+</div>
+</div><!-- MacOS Clean up -->
+<div class="macos">
+ <div class="python">
+ <div class="cpu"><div class="pip build-from-source"><p>Exit the Python terminal.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><div class="virtualenv"><p>Exit the Python terminal and Deactivate the virtualenv <em>MXNet</em> environment.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">(mxnet)$ deactivate</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><div class="docker"><p>Exit the Python terminal and then the docker container.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">root@4919c4f58cac:/# exit</span>
+</pre></div>
+</div>
+</div></div>
+</div>
+</div><!-- Validation for cloud installation --><div class="cloud"><p>Login to the cloud instance you launched, with pre-installed <em>MXNet</em>, following the guide by corresponding cloud provider.</p>
+<p>Start the python terminal.</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ python
+</pre></div>
+</div>
+<!-- Example Python code for CPU --><div class="cpu"><p>Run a short <em>MXNet</em> python program to create a 2X3 matrix of ones, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">mxnet</span> <span class="kn">as</span> <span class="nn">mx</span>
+<span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">))</span>
+<span class="gp">>>> </span><span class="n">b</span> <span class="o">=</span> <span class="n">a</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+<span class="gp">>>> </span><span class="n">b</span><span class="o">.</span><span class="n">asnumpy</span><span class="p">()</span>
+<span class="go">array([[ 3., 3., 3.],</span>
+<span class="go"> [ 3., 3., 3.]], dtype=float32)</span>
+</pre></div>
+</div>
+<p>Exit the Python terminal.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="nb">exit</span><span class="p">()</span>
+<span class="go">$</span>
+</pre></div>
+</div>
+</div><!-- Example Python code for CPU --><div class="gpu"><p>Run a short <em>MXNet</em> python program to create a 2X3 matrix of ones <em>a</em> on a <em>GPU</em>, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3. We use <em>mx.gpu()</em>, to set <em>MXNet</em> context to be GPUs.</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">mxnet</span> <span class="kn">as</span> <span class="nn">mx</span>
+<span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">mx</span><span class="o">.</span><span class="n">gpu</span><span class="p">())</span>
+<span class="gp">>>> </span><span class="n">b</span> <span class="o">=</span> <span class="n">a</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+<span class="gp">>>> </span><span class="n">b</span><span class="o">.</span><span class="n">asnumpy</span><span class="p">()</span>
+<span class="go">array([[ 3., 3., 3.],</span>
+<span class="go"> [ 3., 3., 3.]], dtype=float32)</span>
+</pre></div>
+</div>
+</div></div><!-- Example R code for CPU --><div class="linux macos windows">
+ <div class="r">
+ <div class="cpu"><p>Run a short <em>MXNet</em> R program to create a 2X3 matrix of ones, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3.</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span><span class="kn">library</span><span class="p">(</span>mxnet<span class="p">)</span>
+a <span class="o"><-</span> mx.nd.ones<span class="p">(</span><span class="kt">c</span><span class="p">(</span><span class="m">2</span><span class="p">,</span><span class="m">3</span><span class="p">),</span> ctx <span class="o">=</span> mx.cpu<span class="p">())</span>
+b <span class="o"><-</span> a <span class="o">*</span> <span class="m">2</span> <span class="o">+</span> <span class="m">1</span>
+b
+</pre></div>
+</div>
+</div>
+</div>
+</div><!-- Example R code for GPU --><div class="linux macos windows">
+ <div class="r">
+ <div class="gpu"><p>Run a short <em>MXNet</em> R program to create a 2X3 matrix of ones <em>a</em> on a <em>GPU</em>, multiply each element in the matrix by 2 followed by adding 1. We expect the output to be a 2X3 matrix with all elements being 3. We use <em>mx.gpu()</em>, to set <em>MXNet</em> context to be GPUs.</p>
+<div class="highlight-r"><div class="highlight"><pre><span></span><span class="kn">library</span><span class="p">(</span>mxnet<span class="p">)</span>
+a <span class="o"><-</span> mx.nd.ones<span class="p">(</span><span class="kt">c</span><span class="p">(</span><span class="m">2</span><span class="p">,</span><span class="m">3</span><span class="p">),</span> ctx <span class="o">=</span> mx.gpu<span class="p">())</span>
+b <span class="o"><-</span> a <span class="o">*</span> <span class="m">2</span> <span class="o">+</span> <span class="m">1</span>
+b
+</pre></div>
+</div>
+</div>
+</div>
+</div><div class="linux">
+ <div class="scala julia perl">
+ <div class="cpu gpu"><p>Will be available soon.</p>
+</div>
+</div>
+</div><div class="macos">
+ <div class="scala julia perl">
+ <div class="cpu gpu"><p>Will be available soon.</p>
+</div>
+</div>
+</div><!-- Windows MXNet Installation validation -->
+<div class="windows">
+ <div class="python">
+ <div class="cpu"><div class="build-from-source virtualenv docker">
+<br/>
+Will be available soon.
+</div></div>
+</div><div class="scala julia perl">
+<div class="cpu gpu"><p>Will be available soon.</p>
+</div>
+</div>
+</div>
+<!-- End Windows Installation validation --><div class="devices">
+ <div class="raspberry-pi"><p>Will be available soon.</p>
+</div>
+<div class="nvidia-jetson-tx2"><p>Will be available soon.</p>
+</div>
+</div></div>
+<div class="section" id="download-source-package">
+<span id="download-source-package"></span><h1>Download Source Package<a class="headerlink" href="#download-source-package" title="Permalink to this headline">¶</a></h1>
+</div>
+
+
+ </div>
+ </div>
+
+ <div class="sphinxsidebar rightsidebar" role="navigation" aria-label="main navigation">
+ <div class="sphinxsidebarwrapper">
+ <h3><a href="../index.html">Table Of Contents</a></h3>
+ <ul>
+<li><a class="reference internal" href="#">Installing MXNet</a></li>
+<li><a class="reference internal" href="#validate-mxnet-installation">Validate MXNet Installation</a></li>
+<li><a class="reference internal" href="#download-source-package">Download Source Package</a></li>
+</ul>
+
+ </div>
+ </div>
+ </div><div class="footer">
+<div class="section-disclaimer">
+<div class="container">
+ <div>
+ <img src="https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/apache_incubator_logo.png" height=60>
+ <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 © 2017-2018, 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 src="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.6/js/bootstrap.min.js" integrity="sha384-0mSbJDEHialfmuBBQP6A4Qrprq5OVfW37PRR3j5ELqxss1yVqOtnepnHVP9aJ7xS" crossorigin="anonymous"></script>
+ <script type="text/javascript" src="../_static/js/sidebar.js"></script>
+ <script type="text/javascript" src="../_static/js/search.js"></script>
+ <script type="text/javascript" src="../_static/js/navbar.js"></script>
+ <script type="text/javascript" src="../_static/js/clipboard.min.js"></script>
+ <script type="text/javascript" src="../_static/js/copycode.js"></script>
+ <script type="text/javascript" src="../_static/js/page.js"></script>
+ <script type="text/javascript">
+ $('body').ready(function () {
+ $('body').css('visibility', 'visible');
+ });
+ </script>
+ </body>
+</html>
diff --git a/docs/build_version_doc/build_all_version.sh b/docs/build_version_doc/build_all_version.sh
index e7283728b96..56b80e3a0fc 100755
--- a/docs/build_version_doc/build_all_version.sh
+++ b/docs/build_version_doc/build_all_version.sh
@@ -33,8 +33,10 @@ if [ -z "$1" ]
echo "Please provide a list of version tags you wish to run."
exit 1
else
- tag_list="$1"
- echo "Using these tags: $1"
+ IFS=$';'
+ tag_list=$1
+ echo "Using these tags: $tag_list"
+ for tag in $tag_list; do echo $tag; done
fi
mxnet_url="https://github.com/apache/incubator-mxnet.git"
@@ -60,15 +62,11 @@ for tag in $tag_list; do
git checkout master
git pull
else
- git checkout "tags/$tag"
+ git checkout "v$tag"
fi
- # this gets around the Python 3 support issue in old versions of mxdoc.py
- if [ $tag == '0.11.0' ]
- then
- git checkout master -- docs/mxdoc.py
- fi
- git submodule update || exit 1
- git submodule update --init --recursive
+
+ git submodule update --init --recursive || exit 1
+
make clean
cd docs
make clean
diff --git a/docs/build_version_doc/update_all_version.sh b/docs/build_version_doc/update_all_version.sh
index 90d4fa68d7c..5467ad2157f 100755
--- a/docs/build_version_doc/update_all_version.sh
+++ b/docs/build_version_doc/update_all_version.sh
@@ -17,9 +17,9 @@
# specific language governing permissions and limitations
# under the License.
-# This script will update the html content from building
+# This script will update the html content from building
# different tags.
-# It assumes you have already run build_all_version.sh for
+# It assumes you have already run build_all_version.sh for
# the tags you want to update.
# Takes three arguments:
@@ -39,20 +39,23 @@ MXNET_THEME_DIR="_static/mxnet-theme"
BUILD_HTML_DIR="_build/html"
if [ -z "$1" ]
- then
- echo "Please provide a list of version tags you wish to run. Ex : \"1.1.0 1.0.0 master\""
+ then
+ echo "Please provide a list of version tags you wish to run."
exit 1
else
+ IFS=$';'
tag_list=$1
-fi
+ echo "Using these tags: $tag_list"
+ for tag in $tag_list; do echo $tag; done
+fi
if [ -z "$2" ]
- then
+ then
echo "Please pick a version to use as a default for the website. Ex: 1.1.0"
exit 1
else
tag_default=$2
-fi
+fi
if [ -z "$3" ]
then
@@ -82,14 +85,23 @@ function update_mxnet_css {
# All fixes are done on the master branch of mxnet-incubator repository
# During a nightly build, these fixes will be patched to all the versions in the asf-site repository including the master folder under versions directory.
# copy <master folder location> <version folder location>
-
+
echo "Copying mxnet.css from master branch to all versions...."
cp "$MASTER_SOURCE_DIR/$STATIC_FILES_DIR/mxnet.css" "$built/versions/$tag/_static"
echo "Update fixes complete.."
}
+function update_install {
+ tag=$1
+ echo "Updating installation page for $1..."
+ cp "artifacts/$tag.index.html" "$built/versions/$tag/install/index.html"
+}
+
+
# Update the specified tags with the Versions dropdown
+# Add various artifacts depending on the version
+
for tag in $tag_list; do
# This Python script is expecting the tag_list.txt and it will use that as the entries to populate
@@ -97,22 +109,35 @@ for tag in $tag_list; do
# Patch any fixes to all versions except 0.11.0.
# Version 0.11.0 has old theme and does not make use of the current mxnet.css
- if [ $tag != '0.11.0' ]
- then
- update_mxnet_css $tag
+ # It also has its install page in /getting_started, so we skip updating that
+ if [ $tag != '0.11.0' ]; then
+ if [ -d $built/versions/$tag ]; then
+ echo "The $tag is going to be updated with new css and install pages."
+ update_mxnet_css $tag
+ update_install $tag
+ fi
fi
-
+
# Update all the files that are required to go into the root folder or live version
if [ $tag == $tag_default ]
then
cp -a "$built/versions/$tag/." "$built"
-
+ echo "Updating default site's installation page."
+ cp "artifacts/$tag_default.index.html" "$built/install/index.html"
echo "Copying .htaccess from default branch to root folder...."
- cp "$MASTER_SOURCE_DIR/.htaccess" "$built"
+ cp "artifacts/.htaccess" "$built"
else
file_loc="$built/versions/$tag"
fi
+
+ # Copy the latest README.md from master
+ if [ $tag == 'master' ]; then
+ cd $mxnet_folder
+ git checkout master
+ cp README.md ../$built
+ cd ..
+ fi
done
-
+
echo "The output of this process can be found in the VersionedWeb folder."
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services