Deploy Sagemaker trained model locally

Hi
I am trying to run an already trained MXNet model locally. I do not need to perform any training locally
I have called the MXNETModel function with the trained model and then initiallise the deploy with the instance_type = ‘local’.
But I am getting the following error.
ValueError: Must setup local AWS configuration with a region supported by SageMaker.
Could you please let guide me with either an example of how to run only the predict function with the pretrained model or any thoughts for the above error.

You have to configure your local environment to set it up with the right credentials and default region details.

To this simply run

aws configure in the terminal assuming you have the aws cli tool installed.

it will ask you for access key and secret key, if you don’t have one you can create a user for your local machine by following the instructions here: https://docs.aws.amazon.com/IAM/latest/UserGuide/id_users_create.html

Hi
Thanks for you reply and the advise
I manage to solve the problem with your help.
I was expecting to be able to deploy the model even without internet connection but from my understading this is not possible.
Many thanks