R Deep Learning Essentials - Second Edition

Creating a deep learning AMI in AWS

In the previous example, we used an Amazon Machine Image (AMI) that was built by RStudio. In AWS, you can also create your own AMI's. When you create an AMI, you can install the software you want, load data onto it, and set it up as you wish. This section will show you how to use an AMI to use MXNet on AWS.

1. The first step in creating an AMI is to select the base image that you are going to use. We could start with a base image that just has the operating system installed, but instead we are going to use the RStudio Server with Tensorflow-GPU for AWS that we used previously and add the MXNet package to it.
2. The instructions to install MXNet were adapted from https://mxnet.incubator.apache.org/install/index.html ...