You use the Azure Machine Learning Python SDK to create a batch inference pipeline. You must publish the batch inference pipeline so that business groups in your organization can use the pipeline…

Question

Options

• ACreate multiple endpoints for the published pipeline service and have each business group
• BDefine a PipelineParameter object for the pipeline and use it to specify the business group-
• CDefine a OutputFileDatasetConfig object for the pipeline and use the object to specify the
• DHave each business group run the pipeline on local compute and use a local file for the input

• A
  17% (7)
• B
  71% (29)
• C
  10% (4)
• D
  2% (1)

Explanation

A PipelineParameter object allows you to define a named, typed parameter in a published Azure Machine Learning pipeline that callers can supply at runtime when triggering the pipeline via its REST endpoint. By defining a PipelineParameter for the input data path, each business group can POST to the same published pipeline endpoint and pass their own specific data location as a parameter value in the request body. This approach requires only one published endpoint and avoids code duplication. Creating multiple endpoints (A) adds unnecessary overhead. OutputFileDatasetConfig (C) is for specifying pipeline outputs, not runtime input parameters. Running locally (D) defeats the purpose of a centrally published pipeline and cannot scale across business groups.

No community discussion yet for this question.