AI-300: Operationalizing Machine Learning and Generative AI SolutionsDemo

By Microsoft · Question Mode

QuestionQ138

Implement machine learning model lifecycle and operations

You are implementing hyperparameter tuning with Bayesian sampling for Azure ML Python SDK v2–based model training from a notebook. The notebook is in an Azure Machine Learning workspace. It uses a training script that runs on a compute cluster with 20 nodes.

The code implements a Bandit termination policy with slackjactor set to 0.2 and a sweep job with max_concurrent_trials set to 10.

You must increase the effectiveness of the tuning process by improving sampling convergence.

You need to choose which sampling convergence setting to use.

What should you select?

  • A Set the value of max_concurrent_trials to 20.
  • B Set the value of slack_factor of early_termination policy to 0.9.
  • C Set the value of max_concurrent_trials to 4.
  • D Set the value of slack_factor of early_termination policy to 0.1.
Explanation

Bayesian sampling uses results from completed trials to choose promising later configurations. Limiting the number of concurrently running trials allows more completed-trial feedback to inform each new configuration, improving convergence. The Bandit policy’s slack factor controls how aggressively underperforming trials are terminated; it does not improve Bayesian sampling convergence.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!