TD_GetFutileColumns
TD_GetFutileColumns is used with another table operator, TD_CategoricalSummary to identify columns that are ‘futile’, that is to say not useful in analytic or modelling processes. An input table and category table is provided as input to the table operator, the output is the columns that meet the criteria to be categorized as futile.
An optional syntax option enables the setting of a ThresholdValue which is the value used to determine if a column is considered futile or not. It’s a number between 0 and 1, the default is .95. The detail is in the documentation, link below.
This function is useful to support data cleansing and quality but also useful for quickly identifying columns to focus on so helps data scientists and AI practitioners quickly narrow down which features are worth working with. This function runs in database, executes on Teradata in parallel across amps and works across all columns at once. So it’s performant, example and code below.
TD_GetFutileColumns example
