Question 37
Domain 5: Monitor, retrain, and manage ML lifecycleYou are monitoring data drift in Azure Machine Learning using the SDK v2. Which component should you configure to define the logic for drift detection?
Correct answer: B
Explanation
In Azure Machine Learning SDK v2, a "MonitorSchedule" defines when and how a monitor runs, including the logic used to evaluate data drift over time. It is the scheduling component you configure to set up drift detection for a dataset or model monitoring job.
Why each option is right or wrong
A. DataDriftSchedule
B. MonitorSchedule
Azure Machine Learning SDK v2 uses the `MonitorSchedule` entity to create and configure a monitoring job, including the monitor type, inputs, and the evaluation settings that determine how drift is assessed over time. In the v2 monitoring API, this is the object you submit to define the recurring monitor run; the schedule can be set with a cron expression or recurrence pattern, and the monitor configuration is what drives the drift-detection logic rather than a separate standalone detector class.
C. DriftMetricsLogger
D. RunConfig