--- title: "Data function - numeric-combinations" slug: "numeric-combinations-data-function" updated: 2022-07-27T14:36:28Z published: 2022-07-27T14:36:39Z canonical: "code.tag.bio/numeric-combinations-data-function" --- > ## Documentation Index > Fetch the complete documentation index at: https://code.tag.bio/llms.txt > Use this file to discover all available pages before exploring further. # numeric-combinations ### numeric-combinations **creates: *numeric collection*** A `numeric-combinations` data function produces one or more collections of numeric variables, contingent on the categorical data function provided in the criteria array. The `numerics` attribute specifies which numeric variables will be transformed in combination with the dynamic categorical combinations. | **attribute** | **type** | **description** | **options** | | --- | --- | --- | --- | | `operator` | string | how to combine the criteria and categoricals data | `all``single``pairwise` | | `use_variable_name` | boolean | numeric variable names are used as the new collection names | `true``false` | | `criteria` | array | categorical data | `collection``data_function``argument_reference` | | `numerics` | array | numeric data | `collection``data_function``argument_reference` | ``` { "data_function_type": "numeric-combinations", "operator": "all", "use_variable_name": true, "criteria": [ ... ], "numerics": [ ... ] } ``` By default, the `operator` attribute is set to `all`. The `all` operator means that each `variable` from the collections defined in the `criteria` array will be intersected with every numeric `variable`, one at a time, from all of the other collections in criteria. Dynamic collections from all-vs-all combinations of `variables` are used to transform each numeric `variable` from the `numerics` array. If the operator attribute is set to `single`, each categorical `variable` from the collections defined in the `criteria` array will be independently used to transform each numeric `variable` from the `numerics` array. If the operator attribute is set to `pairwise`, each numeric `variable` from the collections defined in the `criteria` array will be intersected with every numeric `variable`, from each of the other collections in `numerics`. Dynamic collections from all pairwise combinations of variables are used to transform each numeric `variable` from the `numerics` array.