QuestionQ79

VerifiedScaling prototypes into ML models

You are developing a linear model with more than 100 input features, each having values between –1 and 1. You believe that many features are non-informative. You want to remove non-informative features from the model while retaining informative features in their original form. Which technique should you use?

Explanation

L1 regularization (such as Lasso) yields sparse linear models by setting many coefficients exactly to zero. The zero-coefficient features can be removed, while retained features remain the original input features rather than transformed components.

Learn more

Community Discussion

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