Skip to content

Instantly share code, notes, and snippets.

@msis

msis/install_nvidia_dependencies.md

Last active September 27, 2019 15:03
Show Gist options
  • Star 3 You must be signed in to star a gist
  • Fork 2 You must be signed in to fork a gist
  • Save msis/108a74d08f55eed48d8521fa968851ea to your computer and use it in GitHub Desktop.
Save msis/108a74d08f55eed48d8521fa968851ea to your computer and use it in GitHub Desktop.
Download ZIP
[setting up tensorflow-gpu] Step-by-step install for tensorflow-gpu #deeplearning #nvidia #shell #machinelearning #ai
Raw
install_nvidia_dependencies.md

Install nvidia repo packages:

  1. Download nvidia machine learning repo package:
wget http://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1604/x86_64/nvidia-machine-learning-repo-ubuntu1604_1.0.0-1_amd64.deb
  1. Install that package:
sudo dpkg -i nvidia-machine-learning-repo-ubuntu1604_1.0.0-1_amd64.deb
  1. Download nvidia cuda repo package:
wget http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64/cuda-repo-ubuntu1604_8.0.44-1_amd64.deb
  1. Install that package:
sudo dpkg -i cuda-repo-ubuntu1604_8.0.44-1_amd64.deb
  1. Update apt list && upgrade the packages:
sudo apt update && sudo apt full-upgrade
  1. Install cuda and libcudnn
sudo apt install cuda-9-0 libcudnn7-dev
Raw
install_tensorflow-gpu.md

Install tensorflow-gpu

I highly recommend installing tensorflow-gpu using either :

  1. virtualenv
  2. anaconda
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment