data.py

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
def plots():
       df=pd.read_csv("hotel_bookings.csv")

       df1=pd.read_csv("hotel_bookings.csv",usecols=[3,4,6,7,8,26,27])

       months = {'January':'01','February':'02','March':'03','April':'04','May':'05','June':'06','July':'07','August':'08','September':'09','October':'10','November':'11','December':'12'}

       df1['arrival_date_month'] = df['arrival_date_month'].apply(lambda x: months[x])

       df1['arrival_date_month']=pd.to_numeric(df1.arrival_date_month)
       df1['date']=pd.to_datetime((df1.arrival_date_year*10000+df1.arrival_date_month*100+df1.arrival_date_day_of_month).apply(str),format='%Y%m%d')
       df1['weekdays']=df1['date'].dt.day_name()
       df1=df1.drop(['arrival_date_year',
              'arrival_date_month',
              'arrival_date_day_of_month'],axis=1)
       weekday_names=['Monday','Tuesday','Wednesday','Thursday','Friday','Saturday','Sunday']
       weekday_counts=df1.groupby("weekdays").mean()
       weekday_counts=weekday_counts.loc[weekday_names]

       plt.figure(figsize=(8, 5))
       plt.bar(weekday_names,weekday_counts['adr'],color=(0.3, 0.7, 0.6, 0.6))
       plt.title('ADR per day',fontweight="bold", size=20)
       plt.xlabel('Days', fontsize=14)
       plt.ylabel('ADR', fontsize=14)
       plt.savefig('static/adrPerday.jpg',bbox_inches = 'tight')

       stay={'weekend':df1.stays_in_weekend_nights.mean(),'weekdays':df1.stays_in_week_nights.mean()}
       keys = stay.keys()
       values = stay.values()
       plt.figure(figsize=(4, 3))
       plt.bar(keys, values,color=(0.5, 0.4, 0.6, 0.6),width=[0.3,0.3],ec="black",align='center')
       plt.title('Stays vs week time',fontweight="bold", size=15)
       plt.xlabel('Days', fontsize=10)
       plt.ylabel('Night Stays', fontsize=10)
       plt.savefig('static/nightStayed.jpg',bbox_inches = 'tight')


       plt.figure(figsize=(12, 6))
       sns.countplot(x='hotel',hue='is_canceled', data=df,palette='Pastel1')
       plt.title("Cancelation rates in City hotel and Resort hotel",fontweight="bold", size=20)
       plt.savefig('static/cancellation.jpg',bbox_inches = 'tight')

       plt.figure(figsize=(8, 6))
       sns.countplot(x='customer_type', data=df,palette='Pastel2',ec='black')
       plt.title("Customer types",fontweight="bold", size=20)
       plt.savefig('static/customers.jpg',bbox_inches = 'tight')


       plt.figure(figsize=(12,6))
       sns.lineplot(x='arrival_date_month', y='adr',  data= df)
       plt.title("ADR wrt Month",fontweight="bold", size=20)
       plt.savefig('static/adrPerMonth.jpg',bbox_inches = 'tight')

       df['arrival_date'] = df['arrival_date_year'].astype(str) + '-' + df['arrival_date_month'] + '-' + df['arrival_date_day_of_month'].astype(str)
       df['arrival_date'] = df['arrival_date'].apply(pd.to_datetime)
       df['reservation_status_date'] = df['reservation_status_date'].apply(pd.to_datetime)

       cancelled_data = df[df['reservation_status'] == 'Canceled']
       cancelled_data['canc_to_arrival_days'] = cancelled_data['arrival_date'] - cancelled_data['reservation_status_date']
       cancelled_data['canc_to_arrival_days'] = cancelled_data['canc_to_arrival_days'].dt.days

       plt.figure(figsize=(14,6))
       sns.distplot(cancelled_data['canc_to_arrival_days'])
       plt.title("Distribution of days from cancellation to arrival",fontweight="bold", size=20)
       plt.savefig('static/cancellationsPerWeek.jpg',bbox_inches = 'tight')