Explanation:
The question asks for three ways to reconfigure an Azure ML compute cluster's node settings. Based on the community discussion and Azure ML documentation: A (Azure ML Studio) is correct as it provides a UI for managing compute resources. B (update method of AmlCompute class) is correct as it directly modifies ScaleSettings programmatically. C (Azure portal) is debated but historically supported; however, recent comments indicate this option was removed, making it unreliable. D (Azure ML designer) is incorrect as it's for pipeline creation, not compute management. E (refresh_state() of BatchCompute) is incorrect as it only refreshes state without reconfiguration. The consensus from highly upvoted comments (e.g., 23 upvotes for ABC initially, but 17 upvotes clarifying C is no longer valid) supports A and B as definitive answers, with C being questionable. Thus, A and B are the optimal choices based on current functionality and best practices.
Ultimate access to all questions.
You create an Azure Machine Learning compute cluster with the following configuration:
You need to reconfigure the cluster to use these values:
What are three possible ways to reconfigure the compute cluster? Each correct answer presents a complete solution.
A
Use the Azure Machine Learning studio.
B
Run the update method of the AmlCompute class in the Python SDK.
C
Use the Azure portal.
D
Use the Azure Machine Learning designer.
E
Run the refresh_state() method of the BatchCompute class in the Python SDK.
No comments yet.