You create a machine learning model by using the Azure Machine Learning designer. You publish the model as a real-time service on an Azure Kubernetes Service (AKS) inference compute cluster. You…

Question

Options

• AThe name of the AKS cluster where the endpoint is hosted.
• BThe name of the inference pipeline for the endpoint.
• CThe URL of the endpoint.
• DThe run ID of the inference pipeline experiment for the endpoint.
• EThe key for the endpoint.

• A
  4% (1)
• B
  4% (1)
• C
  93% (26)

Explanation

To call a deployed Azure ML real-time endpoint, a client application needs exactly two things: (1) The REST endpoint URL (C) - this is the HTTP address developers send POST requests to in order to get predictions; and (2) The authentication key (E) - by default, AKS real-time endpoints are protected with key-based authentication, so every request must include this key in the Authorization header. The AKS cluster name (A) is an infrastructure detail not needed by app developers. The inference pipeline name (B) and the experiment run ID (D) are internal Azure ML metadata used for auditing and reproducibility, not for making inference calls. Therefore the correct pair is C and E.

No community discussion yet for this question.