DP-100 Question #317: Real Exam Question with Answer & Explanation

Question

You manage an Azure Machine Learning workspace named Workspace1. You plan to create a pipeline in the Azure Machine Learning Studio designer. The pipeline must include a custom component. You need to ensure the custom component can be used in the pipeline. What should you do first?

Options

• ACreate a pipeline endpoint.
• BAdd a linked service to Workspace1.
• CUpload a .json file to Workspace1.
• DCreate a datastore.
• EUpload a .yaml file to Workspace1.

Explanation

To use a custom component in Azure ML Studio designer, you must first upload a YAML specification file that defines the component's interface, environment, and implementation.

Common mistakes.

• A. A pipeline endpoint is used to publish and version a completed pipeline for batch or real-time inference, not to define or register custom components.
• B. A linked service connects the Azure ML workspace to external compute services such as Azure Synapse Analytics, and has no role in registering custom pipeline components.
• C. Azure ML custom component specifications must be authored in YAML format, not JSON, so uploading a .json file would not register a valid component.
• D. A datastore defines a connection to an external storage service for data access and is unrelated to defining or registering component logic.

Concept tested. Registering custom components in Azure ML designer

Reference. https://learn.microsoft.com/en-us/azure/machine-learning/how-to-create-component-pipelines-ui

No community discussion yet for this question.