Apache MXNet on AWS
Developer Guide

Set Up an EC2 Instance for Deep Learning Using Apache MXNet

If you need to run MXNet on only one EC2 instance, use this procedure to set it up. In this procedure, you launch an EC2 instance using the Deep Learning Amazon Machine Image (AMI) for Ubuntu. The AMI installs MXNet, the deep learning framework of choice at AWS, and other components, such as Python 2 and Python 3, launch configuration tools, and many popular AWS libraries and tools. The Deep Learning AMI is provided on AWS Marketplace. After installing MXNet, you use it with the MNIST dataset—a collection of 60,000 images of handwritten digits—to train a model to recognize handwritten digits.

When working with large training data, you might choose to set up a cluster of EC2 instances to provide the required compute power and memory. To set up a cluster and train a model on multiple EC2 instances, see Set Up a Stack for Distributed Deep Learning Using Apache MXNet .


The AMI also includes other deep learning frameworks, such as TensorFlow, Caffe, and Keras. The AMI is available for both Ubuntu and Amazon Linux. For more information, see Deep Learning AMI Ubuntu Version and Deep Learning AMI Amazon Linux Version.