DP-100 Question #409: Real Exam Question with Answer & Explanation

Question

You are developing a two-step Azure Machine Learning pipeline by using the Azure Machine Learning SDK for Python. You need to register the output of the pipeline as a new version of a named dataset after the run has been completed. What should you implement?

Options

• Athe as_input method of the OutputDatasetConfig class
• Bthe register_on_complete method of the OutputDatasetConfig class
• Cthe as_mount method of the DatasetConsumptionConfig class
• Dthe as_download method of the DatasetConsumptionConfig class

Explanation

{"question_number": 3, "correct_answer": "B", "explanation": "The register_on_complete method of the OutputDatasetConfig class is specifically designed to automatically register the pipeline step's output as a named dataset version in Azure Machine Learning once the run completes. This enables dataset versioning and makes the output reusable by other pipelines or users without manual registration steps. The as_input method configures how an OutputDatasetConfig is passed as input to a subsequent step (not for registering). The as_mount and as_download methods belong to DatasetConsumptionConfig and control how a dataset is accessed during a run (mounted as a virtual file system or downloaded locally), not how outputs are registered.", "generated_by": "claude-sonnet", "llm_judge_score": 3}

No community discussion yet for this question.