This repository provides an easy hands-on way to get started with AWS Inferentia. A demonstration of this hands-on can be seen in the AWS Innovate 2023 - AIML Edition session. Please refer to the AWS Neuron SDK official developer guide for details.
Before you begin, you'll need to prepare at least one of the instances below.
- AWS Inf1 EC2 instances (https://aws.amazon.com/ko/ec2/instance-types/inf1/)
- AWS Trn1 EC2 instances (https://aws.amazon.com/ko/ec2/instance-types/trn1/)
- AWS Inf2 EC2 instances (https://aws.amazon.com/ko/ec2/instance-types/inf2/)
A CPU instance (e.g., c5.xlarge) is also possible as long as you only compile the model with the Neuron SDK.
This sample code is provided under the MIT-0 license. See the LICENSE file.