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 2018/01/11 18:40:20 UTC
[GitHub] sandeep-krishnamurthy closed pull request #42: Fix gpu install instructions
sandeep-krishnamurthy closed pull request #42: Fix gpu install instructions
URL: https://github.com/apache/incubator-mxnet-site/pull/42
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/install/index.html b/install/index.html
index 72d5bf18..ee3e399d 100644
--- a/install/index.html
+++ b/install/index.html
@@ -384,8 +384,8 @@ <h1 id="logo-wrap">
<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>
+<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>
@@ -397,8 +397,8 @@ <h1 id="logo-wrap">
$ 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
+<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==1.0.0
</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>
@@ -408,7 +408,7 @@ <h1 id="logo-wrap">
</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
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet-cu90mkl==1.0.0
</pre></div>
</div>
</div><div class="virtualenv"><p><br/></p>
@@ -641,7 +641,7 @@ <h1 id="logo-wrap">
</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>
+<p><em>MXNet</em> is expected to be compatible on macOS with NVIDIA GPUs. Please install CUDA 9.0 and cuDNN 7.0, 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>
@@ -709,8 +709,8 @@ <h1 id="logo-wrap">
<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>
+<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/:$LD_LIBRARY_PATH</em></p>
@@ -1099,15 +1099,15 @@ <h1 id="logo-wrap">
<p><strong>Prerequisites</strong></p>
<p>Install the following NVIDIA libraries to setup <em>MXNet</em> with GPU support:</p>
<ol class="simple">
-<li>Install CUDA 8.0 following the NVIDIA?s <a class="reference external" href="http://docs.nvidia.com/cuda/cuda-installation-guide-microsoft-windows">installation guide</a>.</li>
-<li>Install cuDNN 7 for CUDA 8.0 following the NVIDIA?s <a class="reference external" href="https://developer.nvidia.com/cudnn">installation guide</a>. You may need to register with NVIDIA for downloading the cuDNN library.</li>
+<li>Install CUDA 9.0 following the NVIDIA?s <a class="reference external" href="http://docs.nvidia.com/cuda/cuda-installation-guide-microsoft-windows">installation guide</a>.</li>
+<li>Install cuDNN 7 for CUDA 9.0 following the NVIDIA?s <a class="reference external" href="https://developer.nvidia.com/cudnn">installation guide</a>. You may need to register with NVIDIA for downloading the cuDNN library.</li>
</ol>
<p><strong>Note:</strong> Make sure to add CUDA install path to <code class="docutils literal"><span class="pre">PATH</span></code>.</p>
<div class="pip">
<br/><p><strong>Step 1</strong> Install python.</p>
<p>Recommend install <code class="docutils literal"><span class="pre">Anaconda3</span></code> <a class="reference external" href="https://www.anaconda.com/download/">here</a></p>
-<p><strong>Step 2</strong> Install <em>MXNet</em> with GPU support using CUDA 8.0</p>
-<div class="highlight-bash"><div class="highlight"><pre><span></span>$ pip install mxnet-cu80==1.0.0
+<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==1.0.0
</pre></div>
</div>
</div>
----------------------------------------------------------------
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