numeric_edahelper.flag_outliers

Module Contents

Functions

flag_outliers(df, threshold=0.2)

Takes a dataframe and returns numeric variables (column names) containing outliers exceeding
numeric_edahelper.flag_outliers.flag_outliers(df, threshold=0.2)[source]

Takes a dataframe and returns numeric variables (column names) containing outliers exceeding user-specified threshold, as well as their respective percentage of outliers as defind by IQR

Parameters

  • df (pandas.DataFrame) – dataset used for EDA analysis

  • threshold (float (optional)) – minimum percentage threshold for the proportion of outliers, above which will have variables flagged

Returns

a dictionary containing the name of variables with high outliers and its respective percentage of outliers

Return type

dict

Examples

>>> import pandas as pd
>>> df = pd.DataFrame({'col1': [-100,-200, 1,2,3,4,5,6,7,8,9,10, 1000],
               'col2': [1,2,3,4,5,6,7,8,9,10,11,12,13],
               'col3': [-50, 1,2,3,4,5,6,7,8,9,10,11,50000]})
>>> flag_outliers(df, threshold=0.2)
{'col1': 0.23076923076923078}