#import required libraries for dataframes and visualizations import pandas as pd import matplotlib.pyplot as plt import seaborn as sns %matplotlib inline #import csv file df = pd.read_csv('2019 Winter Data Science Intern Challenge Data Set - Sheet1.csv') #take a look at the dataframe df.info() #statistical data of the dataframe df.describe()

#checking for order_amount with outliers sns.boxplot(y='order_amount',data=df, showfliers=True)

#checking for order_amount without outliers sns.boxplot(y='order_amount',data=df, showfliers=False, color='#95BF47')

#checking for total_items with outliers sns.boxplot(y='total_items',data=df, showfliers=True)

#checking for order_amount without outliers sns.boxplot(y='total_items',data=df, showfliers=False, color='#95BF47')

#a closer look on the extreme outliers input within the data set where total items is greater than 100 df[df['total_items']>100].head()

#a closer look on the extreme outliers input within the data set where order amount is greater than $10000 but excluding those we saw in the previous dataframe df[(df['order_amount']>10000) &(df['total_items']<2000)].head()

#creating a new dataframe that excludes extreme outliers for further analysis new_df = df[(df['total_items']<100) & (df['order_amount']<10000)] #checking the new dataframe new_df.info()

#checking for order_amount with outliers sns.boxplot(y='order_amount',data=new_df, showfliers=True, color='#95BF47')

#checking for total_items with outliers sns.boxplot(y='total_items',data=new_df, showfliers=True, color='#95BF47')

new_df.describe()