The ML.FEATURE_CROSS function

This document describes theML.FEATURE_CROSS function, which lets you createfeature crosses.

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.FEATURE_CROSS(struct_categorical_features [, degree])

Arguments

ML.FEATURE_CROSS takes the following arguments:

• struct_categorical_features: aSTRUCT<STRING> value that specifiesthe categorical features to cross. The maximum number of input features is 10.Don't specify unnamed features or duplicate features instruct_numerical_features.
• degree: anINT64 value that specifies the highest degree of allcombinations of features in the range of[2, 4]. The default value is2.

Output

ML.FEATURE_CROSS returns aSTRUCT<STRING> value that identifies allcombinations of the crossed categorical features with a degree no larger thanthedegree value, except for 1-degree items (the original features) andself-crossing items. The field names in the output struct are concatenationsof the original feature names.

Example

The following example crosses three features:

SELECTML.FEATURE_CROSS(STRUCT('a'ASf1,'b'ASf2,'c'ASf3))ASoutput;

The output looks similar to the following:

+---------------------------------------------+|                   output                    |+---------------------------------------------+| {"f1_f2":"a_b","f1_f3":"a_c","f2_f3":"b_c"} |+---------------------------------------------+

What's next

• For information about feature preprocessing, seeFeature preprocessing overview.