The ML.POLYNOMIAL_EXPAND function

This document describes theML.POLYNOMIAL_EXPAND function, which lets youcalculate all polynomial combinations of the input features.

You can use this function with models that supportmanual feature preprocessing. For moreinformation, see the following documents:

• End-to-end user journeys for ML models
• Contribution analysis user journey

Syntax

ML.POLYNOMIAL_EXPAND(struct_numerical_features [, degree])

Arguments

ML.POLYNOMIAL_EXPAND takes the following arguments:

• struct_numerical_features: aSTRUCT value that contains thenumericalinput features to expand. You can specify less than or equal to10input features. Don't specify unnamed features or duplicate features.
• degree: anINT64 value that specifies the highest degree ofall combinations in the range of[1, 4]. The default value is2.

Output

ML.POLYNOMIAL_EXPAND returns aSTRUCT<STRING> value that contain allpolynomial combinations of the numerical input features with a degree no largerthan the passed-in degree, including the original features. The field names ofthe output struct are concatenations of the original feature names.

Example

The following example calculates the polynomial expansion of twonumerical features:

SELECTML.POLYNOMIAL_EXPAND(STRUCT(2ASf1,3ASf2))ASoutput;

The output looks similar to the following:

+-------------------------------------------------------------------+|                              output                               |+-------------------------------------------------------------------+| {"f1":"2.0","f1_f1":"4.0","f1_f2":"6.0","f2":"3.0","f2_f2":"9.0"} |+-------------------------------------------------------------------+

What's next

• For information about feature preprocessing, seeFeature preprocessing overview.