alberto-re / ec2-deeplearning

Automated setup of a deeplearning-ready EC2 instance (Tensorflow)

README

Automated provisioning of a deeplearning-ready EC2 instance.

This project puts together a CloudFormation template and some helper scripts providing an easy way to get a running virtual machine with all the basic software needed in order to run deep learning applications written in Python on AWS.

USAGE

INSTANCE SETUP

First of all create a copy of the configuration example file and call it 'config.sh':

cp config.sh.default config.sh

Then edit the newly created file properly.

Use the helper script 'mng_ec2.sh' to create the EC2 instance:

sh mng_ec2 launch

After a while the instance public IP address will become available by running the command:

sh mng_ec2 status

SOFTWARE SETUP

SSH into your newly created instance and clone this repository locally:

sudo yum update -y
sudo yum install -y git
git clone https://github.com/proud/ec2-deeplearning.git

Now run the provisioning script 'setup_aws_p2.sh' to install Tensorflow and it's dependencies plus some commonly used data science libraries:

cd ec2-deeplearning
sudo sh setup_aws_p2.sh