Question 1
Domain 1: Design and prepare a machine learning solutionYou are using Azure ML SDK v2 and want to deploy a model to a local endpoint for testing. Which method should you use?
Correct answer: C
Explanation
Azure ML SDK v2 uses local online endpoints for testing, and the method named "invoke_local()" is the one that invokes a deployed model on a local endpoint. The answer matches the SDK pattern for local endpoint invocation rather than cloud deployment.
Why each option is right or wrong
A. ml_client.online_endpoints.create_or_update()
B. ml_client.batch_endpoints.create_or_update()
C. ml_client.online_endpoints.invoke_local()
Azure ML SDK v2 exposes local online endpoint testing through the `online_endpoints` operations group, and the local invocation path is the `invoke_local()` method on `ml_client.online_endpoints`. In this scenario, no cloud endpoint resource, scoring URI, or deployment target is needed; the call is made directly against the local endpoint runtime for validation before any remote deployment.
D. ml_client.models.download()