Skip to content

Instantly share code, notes, and snippets.

@davidnvq
Last active March 4, 2018 00:06
Show Gist options
  • Save davidnvq/9564b42d3d51f756e72dc804a33934fa to your computer and use it in GitHub Desktop.
Save davidnvq/9564b42d3d51f756e72dc804a33934fa to your computer and use it in GitHub Desktop.
Tensorflow GPU Installation
# For more information, visit at
# https://www.tensorflow.org/install/install_linux
#1. NVIDIA requirements
#1.0 Install CUDA Toolkit
http://docs.nvidia.com/cuda/cuda-installation-guide-linux/#axzz4VZnqTJ2A
# Check which version
cat /usr/local/cuda/version.txt
#1.1 Create the CUDA_HOME environment variable at .bash_profile
export CUDA_HOME=/usr/local/cuda-8.0
export PATH=$CUDA_HOME:$PATH
#1.2 Install libcupti-dev library,
# which is the NVIDIA CUDA Profile Tools Interface.
# if CUDA Toolkit >= 8.0
sudo apt-get install cuda-command-line-tools
# if CUDA Toolkit <= 7.5
sudo apt-get install libcupti-dev
# Add its path to LD_LIBRARY_PATH
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda/extras/CUPTI/lib64
#2. Install Tensorflow via Anaconda
# Tensorflow GPU with Python 3.5
pip install --ignore-installed --upgrade https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-1.6.0-cp35-cp35m-linux_x86_64.whl
#3. Verification
# Run the following python script
import tensorflow as tf
print(tf.__version__)
hello = tf.constant('Hello, TensorFlow!')
sess = tf.Session()
print(sess.run(hello))
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment