Source code for featuretools.primitives.standard.transform.cumulative.cum_min
from woodwork.column_schema import ColumnSchema
from featuretools.primitives.base import TransformPrimitive
[docs]class CumMin(TransformPrimitive):
"""Calculates the cumulative minimum.
Description:
Given a list of values, return the cumulative min
(or running min). There is no set window, so the min
at each point is calculated over all prior values.
`NaN` values will return `NaN`, but in the window of a
cumulative caluclation, they're ignored.
Examples:
>>> cum_min = CumMin()
>>> cum_min([1, 2, -3, 4, None, 5]).tolist()
[1.0, 1.0, -3.0, -3.0, nan, -3.0]
"""
name = "cum_min"
input_types = [ColumnSchema(semantic_tags={"numeric"})]
return_type = ColumnSchema(semantic_tags={"numeric"})
uses_full_dataframe = True
description_template = "the cumulative minimum of {}"
def get_function(self):
def cum_min(values):
return values.cummin()
return cum_min