This module builds custom kernel for SageMaker studio from a Dockerfile.
ecr-repo-name
: Name of the ECR repo for the image.studio-domain-id
: SageMaker studio domain to attach the kernel to.studio-domain-name
: SageMaker studio name to attach the kernel to.sagemaker-image-name
: Name of the sagemaker image. This variable is also used to find the Dockerfile. The docker build script will be looking for file insidemodules/mlops/custom-kernel/docker/{sagemaker_image_name}
. 1 Dockerfile is added already:pytorch-10
.studio-execution-role-arn
: SageMaker Studio Domain execution role. Required to associate custom kernel with SageMaker Studio Domain.
app-image-config-name
: Name of the app image config. Defaults toidf-{deployment_name}-app-config
kernel-user-uuid
: Default Unix User ID, defaults to: 1000kernel-user-guid
: Default Unix Group ID, defaults to 100kernel-user-mount-path
: # Path to mount in SageMaker Studio, defaults to/home/sagemaker-user
ECRRepositoryName
: ECR repository nameCustomKernelImageName
: Image nameCustomKernelImageURI
: Image URIAppImageConfigName
: AppConfig image nameSageMakerCustomKernelRoleArn
: Role for custom kernel
{
"ECRRepositoryName": "default",
"CustomKernelImageName": "echo-kernel",
"CustomKernelImageURI": "<account>.dkr.ecr.us-east-1.amazonaws.com/default:latest",
"AppImageConfigName": "echo-kernel-app-config",
"SageMakerCustomKernelRoleArn": "arn:aws:iam::<account>:role/idf-shared-infra-kernels-addfsharedinfrakernelske-9O6FZXGI0MM8",
}