Amazon EC2 setup - Amazon Deep Learning Containers
Next steps
Services or capabilities described in Amazon Web Services documentation might vary by Region. To see the differences applicable to the China Regions, see Getting Started with Amazon Web Services in China (PDF).

Amazon EC2 setup

In this section, you learn how to set up Amazon Deep Learning Containers with Amazon Elastic Compute Cloud.

Complete the following steps to configure your instance:

  • Create an Amazon Identity and Access Management user or modify an existing user with the following policies. You can search for them by name in the IAM console's policy tab.

    For more information about creating or editing an IAM user, see Adding and Removing IAM Identity Permissions in the IAM user guide.

  • Launch an Amazon Elastic Compute Cloud instance (CPU or GPU), preferably a Deep Learning Base AMI. Other AMIs work, but require relevant GPU drivers.

  • Connect to your instance by using SSH. For more information about connections, see Troubleshooting Connecting to Your Instance in the Amazon EC2 user guide..

  • Ensure your Amazon CLI is up to date using the steps in Installing the current Amazon CLI Version.

  • In your instance, run aws configure and provide the credentials of your created user.

  • In your instance, run the following command to log in to the Amazon ECR repository where Deep Learning Containers images are hosted.

    aws ecr get-login-password --region us-east-1 | docker login --username AWS --password-stdin 763104351884.dkr.ecr.us-east-1.amazonaws.com

For a complete list of Amazon Deep Learning Containers, refer to Deep Learning Containers Images.

Note

MKL users: Read the Amazon Deep Learning Containers Intel Math Kernel Library (MKL) Recommendations to get the best training or inference performance.

Next steps

To learn about training and inference on Amazon EC2 with Deep Learning Containers, see Amazon EC2 Tutorials.