QuestionQ81

VerifiedScaling prototypes into ML models

You are an ML engineer at a global car manufacturer. You must build an ML model that predicts car sales across different cities worldwide. Which features or feature crosses should you use to train city-specific relationships between car type and the number of sales?

Explanation

A city is identified by the conjunction of latitude and longitude, so both location values must be binned and crossed together. Crossing that joint location with the one-hot encoded car type produces separate features for each city-bin and car-type combination, allowing the model to learn city-specific sales relationships. Feature crosses are Cartesian products of categorical or bucketed features and encode their interactions.

Learn more

Community Discussion

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