Install TensorFlow v 0.11 and later on EC2 instances with Ubuntu 16.04. Instances: p2.xlarge, g2.xlarge and bigger

install-tensorflow.sh

# Note – this is not a bash script (some of the steps require reboot)
# I named it .sh just so Github does correct syntax highlighting.
#
# This is also available as an AMI in us-east-1 (virginia): ami-cf5028a5
#
# The CUDA part is mostly based on this excellent blog post:
# http://tleyden.github.io/blog/2014/10/25/cuda-6-dot-5-on-aws-gpu-instance-running-ubuntu-14-dot-04/

# Install various packages
sudo apt-get update
sudo apt-get upgrade -y # choose “install package maintainers version”
sudo apt-get install -y build-essential python-pip python-dev git python-numpy swig python-dev default-jdk zip zlib1g-dev

# Blacklist Noveau which has some kind of conflict with the nvidia driver
echo -e "blacklist nouveau\nblacklist lbm-nouveau\noptions nouveau modeset=0\nalias nouveau off\nalias lbm-nouveau off\n" | sudo tee /etc/modprobe.d/blacklist-nouveau.conf
echo options nouveau modeset=0 | sudo tee -a /etc/modprobe.d/nouveau-kms.conf
sudo update-initramfs -u

# Some other annoying thing we have to do
sudo apt-get install -y linux-image-extra-virtual

# REBOOT!
sudo reboot

# Install latest Linux headers
sudo apt-get install -y linux-source linux-headers-`uname -r` 

# Install CUDA 8.0
mkdir install
cd ./install
wget https://developer.nvidia.com/compute/cuda/8.0/prod/local_installers/cuda_8.0.44_linux-run
chmod +x cuda_8.0.44_linux-run
./cuda_8.0.44_linux-run -extract=`pwd`/nvidia_installers
cd nvidia_installers
sudo ./NVIDIA-Linux-x86_64-367.48.run
sudo modprobe nvidia
sudo ./cuda-linux64-rel-8.0.44-21122537.run
cd

# Install CUDNN 8.0
# YOU NEED TO SCP THIS ONE FROM SOMEWHERE ELSE – it's not available online.
# You need to register and get approved to get a download link. Very annoying.
tar xvzf cudnn-8.0-linux-x64-v5.1.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/include/cudnn.h /usr/local/cuda/lib64/libcudnn*

# At this point the root mount is getting a bit full
# I had a lot of issues where the disk would fill up and then Bazel would end up in this weird state complaining about random things
# Make sure you don't run out of disk space when building Tensorflow!
sudo mkdir /mnt/tmp
sudo chmod 777 /mnt/tmp
sudo rm -rf /tmp
sudo ln -s /mnt/tmp /tmp
# Note that /mnt is not saved when building an AMI, so don't put anything crucial on it

# Install Bazel
# Latest installation manual is here:
# https://bazel.build/versions/master/docs/install.html
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

# Install TensorFlow
export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/usr/local/cuda-8.0/lib64"
export CUDA_HOME=/usr/local/cuda-8.0
git clone --recurse-submodules https://github.com/tensorflow/tensorflow
cd tensorflow

# IMPORTANT! set compability with 3.0 in the next configure step if you are using g2.xlarge. 
# If you using p2.xlarge, just use empty string
# Please note that each additional compute capability significantly increases your build time and binary size.
# [Default is: "3.5,5.2"]: 3.0
TF_UNOFFICIAL_SETTING=1 ./configure

# Build Python package
# Note: you have to specify --config=cuda here - this is not mentioned in the official docs
# https://github.com/tensorflow/tensorflow/issues/25#issuecomment-156173717
# To build with GPU support:
bazel build -c opt --config=cuda //tensorflow/tools/pip_package:build_pip_package
bazel-bin/tensorflow/tools/pip_package/build_pip_package /tmp/tensorflow_pkg
sudo pip install /tmp/tensorflow_pkg/tensorflow-0.11.0rc2-py2-none-any.whl

# Test it!
cd tensorflow/models/image/cifar10/
python cifar10_multi_gpu_train.py 

# On a g2.2xlarge: step 100, loss = 4.50 (325.2 examples/sec; 0.394 sec/batch)
# On a g2.8xlarge: step 100, loss = 4.49 (337.9 examples/sec; 0.379 sec/batch)
# doesn't seem like it is able to use the 4 GPU cards unfortunately :(

# To run tf in ipython after session relaunch yiu have to export some variables
LD_LIBRARY_PATH=/usr/local/cuda/extras/CUPTI/lib64
export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/usr/local/cuda-8.0/lib64"
export CUDA_HOME=/usr/local/cuda-8.0