register a model to sagemaker model group

regiter_model.py

import boto3


sm_client = boto3.client('sagemaker', region_name='us-east-1')

model_url = "s3://sagemaker-us-east-1-.../sagemaker-scikit-learn-.../output/model.tar.gz"

modelpackage_inference_specification =  {
    "InferenceSpecification": {
      "Containers": [
         {
            "Environment" : {
                "SAGEMAKER_CONTAINER_LOG_LEVEL" : "20",
                "SAGEMAKER_PROGRAM" : "sklearn-titanic-train-job.py",
                "SAGEMAKER_REGION" : "us-east-1",
                "SAGEMAKER_SUBMIT_DIRECTORY" : "s3://sagemaker-us-east-1-.../titanic_source_dir/sourcedir.tar.gz"
            },
            "Image": '683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-scikit-learn:0.20.0-cpu-py3',
            "ModelDataUrl": model_url
         }
      ],
      "SupportedContentTypes": [ "text/csv", "application/x-npy" ],
      "SupportedResponseMIMETypes": [ "text/csv" ],
   }
 }

# Alternatively, you can specify the model source like this:
# modelpackage_inference_specification["InferenceSpecification"]["Containers"][0]["ModelDataUrl"]=model_url

create_model_package_input_dict = {
    "ModelPackageGroupName" : "Titanic-survival",
    "ModelPackageDescription" : "predict titanic deaths",
    "ModelApprovalStatus" : "PendingManualApproval"
}
create_model_package_input_dict.update(modelpackage_inference_specification)

create_model_package_response = sm_client.create_model_package(**create_model_package_input_dict)
model_package_arn = create_model_package_response["ModelPackageArn"]
print('ModelPackage Version ARN : {}'.format(model_package_arn))