Source code for featuretools.primitives.standard.aggregation.std
import numpy as np
from woodwork.column_schema import ColumnSchema
from featuretools.primitives.base.aggregation_primitive_base import AggregationPrimitive
from featuretools.utils.gen_utils import Library
[docs]class Std(AggregationPrimitive):
"""Computes the dispersion relative to the mean value, ignoring `NaN`.
Examples:
>>> std = Std()
>>> round(std([1, 2, 3, 4, 5, None]), 3)
1.414
"""
name = "std"
input_types = [ColumnSchema(semantic_tags={"numeric"})]
return_type = ColumnSchema(semantic_tags={"numeric"})
stack_on_self = False
compatibility = [Library.PANDAS, Library.DASK, Library.SPARK]
description_template = "the standard deviation of {}"
def get_function(self, agg_type=Library.PANDAS):
if agg_type in [Library.DASK, Library.SPARK]:
return "std"
return np.std