Last active
August 31, 2023 18:28
-
-
Save Vuong-Chu/59524d82bc66de160913f28934e2a815 to your computer and use it in GitHub Desktop.
This function is to remove outliers in columns of a dataframe and ignore missing values that may be processed in following steps.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# Define function to detect outliers for numerical variables | |
import pandas as pd | |
def clean_outliers(data, types = "IQR", threshold = 3.0): | |
''' | |
This function will cleanse outliers only | |
and leave missing values alone. | |
Parameters: | |
data (DataFrame): Raw data that need to detect and clean the outliers. | |
types (str): Declare the method to detect outliers ("IQR" - interquartile range or "Z" - Z-score) | |
threshold (floar or int): Declare the threshold when detect the outliers with Z-score. | |
Returns: | |
result (DataFrame): Cleaned data. | |
''' | |
def detect_discrete_outliers(data, types, threshold): | |
data.dropna(inplace = True) | |
if types == "IQR": | |
Q1 = data.quantile(0.25) | |
Q3 = data.quantile(0.75) | |
IQR = Q3 - Q1 | |
mask1 = Q1 - 1.5 * IQR | |
mask2 = Q3 + 1.5 * IQR | |
outliers = (data < mask1) | (data > mask2) | |
elif types == "Z": | |
mean = data.mean() | |
std = data.std() | |
z_score = (data - mean)/std | |
outliers = abs(z_score) > threshold | |
else: | |
raise Warning("Only 2 types: IQR or Z") | |
return outliers | |
df = data.copy() | |
list_of_outliers = [False]*df.shape[0] | |
for x in df: | |
list_of_outliers |= detect_discrete_outliers(df[x], types, 3) | |
result = df[-list_of_outliers] | |
return result |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment