Question

Basically I’m trying to split the data into numerical and categorical variables and then look at them separately to identify outliers that are three standard deviations away from the mean of that variable.

so I have the code but im not sure about it. please give me code for the requirements.

Code:

import pandas as pd

import numpy as np

import matplotlib.pyplot as plt

df = pd.read_csv('training_data.csv')

num_only = df._get_numeric_data()

outliers=[]

def detect_outlier(num_only):

   

    threshold=3

    mean_1 = mean(num_only)

    std_1 = std(num_only)

   

    

    for y in num_only:

        zscore = float(np.abs((y - mean_1)/std_1))

        if zscore > threshold:

            outliers.append(y)

            return outliers

outlier_datapoints = detect_outlier(num_only)

print(outlier_datapoints)

Answer 1

Code to identify outliers:-

import numpy as np

import matplotlib.pyplot as plt

seed(1)

anomalies = []

data = np.random.randn(50000) * 20 + 20

def find_anomalies(random_data):

  random_data_std = std(random_data)

random_data_mean = mean(random_data)

anomaly_cut_off = random_data_std * 3

lower_limit = random_data_mean - anomaly_cut_off

  upper_limit = random_data_mean + anomaly_cut_off

  print(lower_limit)

  for outlier in random_data:

  if outlier > upper_limit or outlier < lower_limit:

  anomalies.append(outlier)

  return anomalies

find_anomalies(data)