gptr

# The following describes how to set up a new image on EC2 with Cuda 7.5, CuDNN, and Tensor Flow
# The default Ubuntu 14.04 kernel needs to be reconfigured in order to work propperly (see below).
# The result of this available as an AMI: ami-ff3a2095 
# Another available AMI: ami-e191b38b 
# Via http://ramhiser.com/2016/01/05/installing-tensorflow-on-an-aws-ec2-instance-with-gpu-support/
# and https://github.com/jasimpson/tensorflow-on-aws
# This works as-is but when upgrading this image to tensor flow 0.8.0 I get the error:
# "/usr/bin/python: libcudart.so.7.5: cannot open shared object file: No such file or directory"

# Launch a new GPU instance on EC2