Skip to content

Instantly share code, notes, and snippets.

@piiswrong
Last active December 12, 2017 07:58
Show Gist options
  • Save piiswrong/87ef716a421f0405e5d09415e54fe8cd to your computer and use it in GitHub Desktop.
Save piiswrong/87ef716a421f0405e5d09415e54fe8cd to your computer and use it in GitHub Desktop.
install mxnet on ubuntu
#!/usr/bin/env bash
set -e
# install cuda-7.5
wget http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1404/x86_64/cuda-repo-ubuntu1404_7.5-18_amd64.deb
sudo dpkg -i cuda-repo-ubuntu1404_7.5-18_amd64.deb
sudo apt-get update
sudo apt-get install -y linux-image-extra-`uname -r` linux-headers-`uname -r` linux-image-`uname -r`
sudo apt-get install -y cuda-7-5
echo "export LD_LIBRARY_PATH=/usr/local/cuda/lib64/:\$LD_LIBRARY_PATH" | tee -a ~/.profile | tee -a ~/.bashrc
# install cudnn5
echo "deb http://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1404/x86_64 /" | sudo tee --append /etc/apt/sources.list.d/nvidia-ml.list
sudo apt-get update
sudo apt-get install -y --force-yes --no-install-recommends libcudnn5-dev=5.1.3-1+cuda7.5
# install dependencies
sudo apt-get update
sudo apt-get install -y build-essential libatlas-base-dev libopencv-dev git unzip
# build
cd ~
git clone https://github.com/dmlc/mxnet.git --recursive
cd mxnet
cp make/config.mk .
echo 'USE_CUDA=1' >> config.mk
echo 'USE_CUDA_PATH=/usr/local/cuda' >> config.mk
echo 'USE_CUDNN=1' >> config.mk
echo 'USE_MKL2017=1' >> config.mk
echo 'USE_MKL2017_EXPERIMENTAL=1' >> config.mk
echo 'USE_DIST_KVSTORE=1' >> config.mk
make -j$(nproc)
echo "export PYTHONPATH=~/mxnet/python:\$PYTHONPATH" >> ~/.bashrc
echo "export LD_LIBRARY_PATH=~/mxnet/mklml_lnx_2017.0.1.20161005/lib:\$LD_LIBRARY_PATH" >> ~/.bashrc
# python and jupyter
sudo apt-get install -y python-pip python-dev python-numpy python-opencv graphviz python-scipy python-sklearn python-nose
sudo pip install jupyter graphviz pandas bokeh
jupyter notebook --generate-config
source ~/.bashrc
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment