datareactor.sieve package

Classes

Sieve

The Sieve class filters out derived columns that don’t make sense.

class datareactor.sieve.Sieve[source]

Bases: object

The Sieve class filters out derived columns that don’t make sense.

Methods

filter(dataset, columns)

The filter function takes in a dataset and a list of derived columns; it returns a subset of the derived columns after removing any derived columns which don’t make sense - i.e.

filter(dataset, columns)[source]

The filter function takes in a dataset and a list of derived columns; it returns a subset of the derived columns after removing any derived columns which don’t make sense - i.e. all constant values, redundant with another column, etc.

Parameters
  • dataset (Dataset) – The dataset.

  • columns (list of DerivedColumn) – The derived columns.

Returns

A list of derived columns.