When we deploy a model as an ACIWebService in Azure Machine Learning Service, we do not need to specify any deployment_target
.
According to the AzureML documentation for azureml.core.model.model
class,
deployment_target
ComputeTarget
default value: None
A ComputeTarget to deploy the Webservice to. As Azure Container Instances has no associated ComputeTarget, leave this parameter as None to deploy to Azure Container Instances.
What does Microsoft mean by
As Azure Container Instances has no associated ComputeTarget
In which "Compute Target" is an ACIWebService deployed?
Azure Container Instances itself is the compute platform. It spins up a container in a serverless-fashion.