TensorFlow™ AMD Setup Prerequisites

This guide was created for versions: v0.1.0 - Latest

Ubuntu Users

  • Ubuntu 14.04 with kernel 3.19.0-79-generic or Ubuntu 16.04 with default kernel.
    • You can check the kernel version using this command
    • > hostnamectl status
  • JDK 8 and bazel
    > sudo add-apt-repository ppa:webupd8team/java
    > sudo apt-get update
    > sudo apt-get install oracle-java8-installer
    > echo "deb [arch=amd64] http://storage.googleapis.com/bazel-apt stable jdk1.8" | sudo tee /etc/apt/sources.list.d/bazel.list
    > curl https://bazel.build/bazel-release.pub.gpg | sudo apt-key add -
    > sudo apt-get update && sudo apt-get install bazel
    > sudo apt-get upgrade bazel
  • gcc or clang
    > sudo apt-get install gcc
  • build-essential
    > sudo apt-get install build-essential
  • git
    > sudo apt install git
  • clinfo
    > sudo apt install clinfo
  • Install Python Packages
    > sudo apt-get install python-dev python-pip
    > sudo pip install numpy
  • OpenCL drivers for your hardware, we have tested the following drivers
    • Intel drivers are available from the Intel website
      • Once you have downloaded the package
      • Intel GPU
        > tar -xvf ./intel_sdk_for_opencl_2016_ubuntu_6.3.0.1904_x64.tgz
        > cd ./intel_sdk_for_opencl_2016_ubuntu_6.3.0.1904_x64/rpm
        > sudo apt-get install dkms
        > cd ..
        > sudo ./install.sh
      • Intel CPU
        > sudo apt-get install opencl-headers lsb-core
        > wget http://registrationcenter-download.intel.com/akdlm/irc_nas/9019/opencl_runtime_16.1.1_x64_ubuntu_6.4.0.25.tgz
        > tar xvf opencl_runtime_16.1.1_x64_ubuntu_6.4.0.25.tgz
        > cd opencl_runtime_16.1.1_x64_ubuntu_6.4.0.25/
        > sudo ./install.sh
    • AMD drivers

      Go to the AMD support website to download and install the correct drivers for your hardware http://support.amd.com/en-us/download.