By using AWS re:Post, you agree to the Terms of Use

Possible to experiment with AWS Deep Learning Containers on nonAWS machine?


I'm excited about the prospects of experimenting with the recently announced preconfigured Deep Learning containers. I'd like to set things up on my local Docker server, in order to ensure I waste as little time as possible when I move to AWS for the GPU horsepower. Unfortunately, when I run docker pull on my local machine I get Error response from daemon: no basic auth credentials . Which credentials should I be using, and where should I set them?

Edited by: ivar vasara on Mar 28, 2019 6:55 PM

2 Answers

Hi ivar vasara,

You should be able to do pull the image to your local machine. AWS Deep Learning Container images are public and hosted on Amazon ECR - In order to securely access the repository, proper authentication from the Docker client to the repository is important. This can be done with a docker login command to authenticate to an ECR registry that provides an authorization token valid for 12 hours. If you’re using the AWS CLI, you can use a get-login command which retrieves the token, decodes it, and converts into a docker login command for you. You will need an AWS account and export/configue the credentials of an IAM user with Amazon ECR access to do this.

See here for an example on how to login to the Amazon ECR repository where AWS Deep Learning Containers images are hosted:

Read more about authenticating Amazon ECR repositories here:

AWS docs on ECR registry authentication docs:

Let us know if you have more questions.


answered 3 years ago

Hi ivar vasara,

Let us know if you have additional questions. You can also find support on the DL Containers forum here:


answered 3 years ago

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.

Guidelines for Answering Questions