Designing and Implementing a Data Science Solution on Azure — Question 57

You are preparing to train a regression model via automated machine learning. The data available to you has features with missing values, as well as categorical features with little discrete values.
You want to make sure that automated machine learning is configured as follows:
✑ missing values must be automatically imputed.
✑ categorical features must be encoded as part of the training task.
Which of the following actions should you take?

Answer options

• A. You should make use of the featurization parameter with the 'auto' value pair.
• B. You should make use of the featurization parameter with the 'off' value pair.
• C. You should make use of the featurization parameter with the 'on' value pair.
• D. You should make use of the featurization parameter with the 'FeaturizationConfig' value pair.

Correct answer: A

Explanation

The correct answer is A because setting the featurization parameter to 'auto' allows automated machine learning to automatically impute missing values and encode categorical features as needed. Option B does not enable any processing, while C may not guarantee automatic handling of missing data and encoding. Option D requires additional configuration and does not directly imply automatic handling.