Created
May 5, 2016 23:23
-
-
Save nobuf/6db4528ec7a38752e7c1316abbbbad9c to your computer and use it in GitHub Desktop.
Install TensorFlow 0.8 GPU enabled on Ubuntu 14.04 EC2 g2.2xlarge
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# you need to sign-up https://developer.nvidia.com/rdp/cudnn-download | |
scp cudnn-7.0-linux-x64-v4.0-prod.tgz ubuntu@your-ec2-instance:~/ | |
ssh ubuntu@your-ec2-instance | |
curl -O 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 | |
# see http://tleyden.github.io/blog/2015/11/22/cuda-7-dot-5-on-aws-gpu-instance-running-ubuntu-14-dot-04/ | |
sudo apt-get install -y linux-image-extra-`uname -r` linux-headers-`uname -r` linux-image-`uname -r` | |
sudo apt-get install cuda | |
tar zxvf cudnn-7.0-linux-x64-v4.0-prod.tgz | |
sudo cp cuda/include/cudnn.h /usr/local/cuda/include/ | |
sudo cp cuda/lib64/libcudnn* /usr/local/cuda/lib64 | |
sudo chmod a+r /usr/local/cuda/lib64/libcudnn* | |
# installing tensorflow | |
sudo apt-get install python-pip python-dev | |
sudo pip install --upgrade https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow-0.8.0-cp27-none-linux_x86_64.whl | |
git clone --recurse-submodules https://github.com/tensorflow/tensorflow | |
cat ~/.bashrc | |
export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/usr/local/cuda/lib64" | |
export CUDA_HOME=/usr/local/cuda | |
# With running the below command, I can see some messages about GPU... | |
# `I tensorflow/core/common_runtime/gpu/gpu_device.cc:755] Creating TensorFlow device (/gpu:0) -> (device: 0, name: GRID K520, pci bus id: 0000:00:03.0)` | |
# It takes about 3 seconds on MacBook Pro 2.6 GHz Intel Core i5, so I was expecting it a lot faster. | |
# Apparently it takes about the same time... maybe the bottleneck is not something GPU can solve. I might be missing something though. | |
cd tensorflow/tensorflow/models/image/imagenet | |
time python classify_image.py |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment