🐢

How to setup Linux server

21 Mar 2021 | server linux

This post is about how to setup Linux server for machine learning.

Install essentials

sudo apt update && sudo apt upgrade

sudo apt install build-essential cmake pkg-config libjpeg8-dev libtiff5-dev libjasper-dev libpng12-dev libgtk-3-dev libavcodec-dev libavformat-dev libswscale-dev libv4l-dev libatlas-base-dev gfortran screen tmux
sudo apt-get install python2.7 python-pip python3-pip python-dev python-virtualenv git dkms linux-headers-generic openssh-server vim build-essential python3-tk nodejs npm nodejs-legacy

Install python libraries

sudo pip3 install -U pip
sudo pip3 install -U matplotlib scipy numpy scikit-learn atpy astropy pandas readchar JSAnimation opencv-python jupyter jupyterlab pillow

sudo apt update

Install CUDA Toolkit

Download CUDA Toolkit from Archive.

CUDA Toolkit 9.0 
wget https://developer.nvidia.com/compute/cuda/9.0/Prod/local_installers/cuda-repo-ubuntu1604-9-0-local_9.0.176-1_amd64-deb
sudo dpkg -i cuda-repo-ubuntu1604-9-0-local_9.0.176-1_amd64.deb
sudo apt-key add /var/cuda-repo-9-0-local/7fa2af80.pub
sudo apt-get update
sudo apt-get install cuda

export PATH=$PATH:/usr/local/cuda-9.0/bin
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda-9.0/lib64
CUDA Toolkit 10.2 
wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64/cuda-ubuntu1604.pin
sudo mv cuda-ubuntu1604.pin /etc/apt/preferences.d/cuda-repository-pin-600
wget http://developer.download.nvidia.com/compute/cuda/10.2/Prod/local_installers/cuda-repo-ubuntu1604-10-2-local-10.2.89-440.33.01_1.0-1_amd64.deb
sudo dpkg -i cuda-repo-ubuntu1604-10-2-local-10.2.89-440.33.01_1.0-1_amd64.deb
sudo apt-key add /var/cuda-repo-10-2-local-10.2.89-440.33.01/7fa2af80.pub
sudo apt-get update
sudo apt-get -y install cuda

Check CUDA installed

nvcc --version

Install cuDNN

Download cuDNN from Archive. (Log-in and download)

tar xvzf cudnn-*.tgz
sudo cp -P cuda/include/cudnn.h /usr/local/cuda/include
sudo cp -P cuda/lib64/libcudnn* /usr/local/cuda/lib64
sudo chmod a+r /usr/local/cuda/include/cudnn.h /usr/local/cuda/lib64/libcudnn*
sudo apt-get install libcupti-dev
reboot

Check cuDNN installed

cat /usr/local/cuda/include/cudnn.h | grep CUDNN_MAJOR -A 2

Install Tensorflow

https://www.tensorflow.org/install/install_linux https://github.com/tensorflow/tensorflow/releases

ex) sudo pip3 install tensorflow-gpu==1.4.1

Install PyTorch

https://pytorch.org/ https://github.com/pytorch/pytorch/releases

ex) sudo pip3 install torch torchvision

Related Posts

Comments