Apurv Jain

hub-tealbox-assignment August 8, 2024 [1]: import pandas as pd import numpy as np import matplotlib.pyplot as plt %matplotlib inline import seaborn as sns import plotly.express as px import plotly.graph_objects as go import datetime from datetime import datetime import calendar [2]: df = pd.read_excel(r"E:\interviwe_assignment\Tealbox Digital.xlsx") df.head() [2]: 0 1 2 3 4 Cloud_Kitchen_Business Cloud_Kitchen_Business Cloud_Kitchen_Business Cloud_Kitchen_Business Cloud_Kitchen_Business (Ahmedabad (Ahmedabad (Ahmedabad (Ahmedabad (Ahmedabad 0 1 2 3 4 invoice_no- date payment_type 11:10:00 Online 11:14:00 Online 11:42:00 Online 11:43:00 Online 11:47:00 Online 0 1 2 3 4 total_tax- discount- 0 round_off -0.08 total 129 - … … … … … … - outlet Bodakdev) Bodakdev) Bodakdev) Bodakdev) Bodakdev) container_charge- brand Bowl_Meals Bowl_Meals Bowl_Meals North_Indian_Meals Bowl_Meals status channel Success Zomato Success Swiggy Success Swiggy Success Swiggy Success Zomato service_charge 0 0 0 0 0 \ my_amount- waived_off 0 0 0 0 0 item_name Butter Chicken Rice Meal 1 \ \ \ 1 2 3 4 0 1 2 3 4 -0.16 -0.48 0.14 -0.40 - Veg Dum Biryani Meal Hot Garlic Schezwan Chicken & Veg Fried Rice Meal Veg. Lunch Pack Thali Chicken Chilli & Egg Fried Rice Meal category_name item_price Indian Meals [M101] 189 Indian Meals [M101] 159 Chinese Meals [M101] 169 Lunch Packs Thali [MB] 169 Egg Fried Rice Meals [M101] 179 item_quantity 1 1 1 3 1 item_total- [5 rows x 21 columns] [3]: df.drop(['delivery_charge','service_charge','waived_off','round_off'], axis=1,␣ ↪inplace = True) [4]: df.info() RangeIndex: 32938 entries, 0 to 32937 Data columns (total 17 columns): # Column Non-Null Count Dtype --- ------------------- ----0 outlet 32938 non-null object 1 brand 32938 non-null object 2 invoice_no 32938 non-null int64 3 date 32938 non-null object 4 payment_type 32938 non-null object 5 status 32938 non-null object 6 channel 32938 non-null object 7 my_amount 32938 non-null int64 8 total_tax 32938 non-null float64 9 discount 32938 non-null float64 10 container_charge 32938 non-null int64 11 total 32938 non-null int64 12 item_name 32938 non-null object 13 category_name 32938 non-null object 14 item_price 32938 non-null int64 15 item_quantity 32938 non-null int64 16 item_total 32938 non-null int64 dtypes: float64(2), int64(7), object(8) memory usage: 4.3+ MB [5]: df.dropna(inplace = True) df['date'] = pd.to_datetime(df['date'], dayfirst=True) df['req_hour'] = df['date'].dt.hour 2 df.set_index('date', inplace= True) df['day_of_week'] = df.index.strftime('%A') df.head() [5]: outlet date- 11:10:00 11:14:00 11:42:00 11:43:00 11:47:00 Cloud_Kitchen_Business Cloud_Kitchen_Business Cloud_Kitchen_Business Cloud_Kitchen_Business Cloud_Kitchen_Business brand date- date- date- date- 11:10:00 11:14:00 11:42:00 11:43:00 11:47:00 11:10:00 11:14:00 11:42:00 11:43:00 11:47:00 11:10:00 11:14:00 11:42:00 11:43:00 11:47:00 11:10:00 11:14:00 11:42:00 11:43:00 11:47:00 (Ahmedabad (Ahmedabad (Ahmedabad (Ahmedabad (Ahmedabad - Bodakdev) Bodakdev) Bodakdev) Bodakdev) Bodakdev) invoice_no payment_type Bowl_Meals Bowl_Meals Bowl_Meals North_Indian_Meals Bowl_Meals - \ Online Online Online Online Online status \ Success Success Success Success Success channel my_amount total_tax discount container_charge Zomato Swiggy Swiggy Swiggy Zomato - - - - total item_name - Butter Chicken Rice Meal Veg Dum Biryani Meal Hot Garlic Schezwan Chicken & Veg Fried Rice Meal Veg. Lunch Pack Thali Chicken Chilli & Egg Fried Rice Meal category_name item_price item_quantity Indian Meals [M101] Indian Meals [M101] Chinese Meals [M101] Lunch Packs Thali [MB] Egg Fried Rice Meals [M101] - 1 1 1 3 1 item_total req_hour day_of_week date 3 \ \ \ - 11:10:00 11:14:00 11:42:00 11:43:00 11:47:00 - - Sunday Sunday Sunday Sunday Sunday Data exploration and Manipulations [6]: out = df['outlet'].str.split(" ", n=1, expand=True) df['city_outlet'] = out[1] df['city_outlet'] = df['city_outlet'].str.strip("()") df.drop('outlet', axis=1, inplace=True) df.head() [6]: brand date- date- date- date- 11:10:00 11:14:00 11:42:00 11:43:00 11:47:00 11:10:00 11:14:00 11:42:00 11:43:00 11:47:00 11:10:00 11:14:00 11:42:00 11:43:00 11:47:00 11:10:00 11:14:00 11:42:00 11:43:00 11:47:00 invoice_no payment_type Bowl_Meals Bowl_Meals Bowl_Meals North_Indian_Meals Bowl_Meals - Online Online Online Online Online status \ Success Success Success Success Success channel my_amount total_tax discount container_charge Zomato Swiggy Swiggy Swiggy Zomato - - - - total item_name - Butter Chicken Rice Meal Veg Dum Biryani Meal Hot Garlic Schezwan Chicken & Veg Fried Rice Meal Veg. Lunch Pack Thali Chicken Chilli & Egg Fried Rice Meal category_name item_price item_quantity Indian Meals [M101] Indian Meals [M101] Chinese Meals [M101] Lunch Packs Thali [MB] Egg Fried Rice Meals [M101] - 1 1 1 3 1 item_total req_hour day_of_week 4 city_outlet \ \ \ date- 11:10:00 11:14:00 11:42:00 11:43:00 11:47:00 - - Sunday Sunday Sunday Sunday Sunday Ahmedabad Ahmedabad Ahmedabad Ahmedabad Ahmedabad - Bodakdev Bodakdev Bodakdev Bodakdev Bodakdev [7]: df.payment_type.value_counts() [7]: Online 31579 Cash 847 Other [PayTM] 148 CARD 123 Other [Paytm] 60 Other [Google Pay] 50 Other [paytm] 42 Other [Googlepay] 31 Other [GPAY] 17 Other [RAZORPAY] 12 Other [gpay] 10 Other [Online] 8 Other [Gpay] 5 Other [Google Pay](ROZAY PAY) 3 Other [Phone PE] 2 Other [phone pe] 1 Name: payment_type, dtype: int64 [8]: df['payment_type'] =df['payment_type'].replace(['Other [PayTM]','Other␣ ↪[Paytm]','Other [paytm]' ,'Other [Google Pay]','Other [Googlepay]','Other [Google Pay](ROZAY PAY)'␣ ↪,'Other [GPAY]','Other [gpay]','Other [GPAY]', 'Other [RAZORPAY]' , 'Other [gpay]', 'Other [Online]', 'Other [Gpay]', 'Other␣ ↪[Phone PE]','Other [phone pe]'], 'UPI') [9]: df_categorical = df.select_dtypes(include= ['object']) [10]: df.info() DatetimeIndex: 32938 entries,-:10:00 to-:49:00 Data columns (total 18 columns): # Column Non-Null Count Dtype -------------- ------- -----0 brand 32938 non-null object 1 invoice_no 32938 non-null int64 2 payment_type 32938 non-null object 5 3 status 32938 non-null object 4 channel 32938 non-null object 5 my_amount 32938 non-null int- non-null float64 6 total_tax 7 discount 32938 non-null float64 8 container_charge 32938 non-null int64 9 total 32938 non-null int64 10 item_name 32938 non-null object 11 category_name 32938 non-null object 12 item_price 32938 non-null int64 13 item_quantity 32938 non-null int64 14 item_total 32938 non-null int64 15 req_hour 32938 non-null int64 16 day_of_week 32938 non-null object 17 city_outlet 32938 non-null object dtypes: float64(2), int64(8), object(8) memory usage: 4.8+ MB [11]: df.channel.value_counts() [11]: Swiggy 17089 Zomato 14694 App/Web 1155 Name: channel, dtype: int64 [12]: df_numerical = df.select_dtypes(include= ['float','int64']) #df_numerical.drop('invoice_no', axis=1, inplace = True) [13]: df_numerical.head() [13]: date- date- 11:10:00 11:14:00 11:42:00 11:43:00 11:47:00 invoice_no my_amount total_tax discount - - - - \ container_charge total item_price item_quantity - - - 1 1 1 3 1 11:10:00 11:14:00 11:42:00 11:43:00 11:47:00 item_total req_hour 6 \ date- 11:10:00 11:14:00 11:42:00 11:43:00 11:47:00 - - [130]: plt.figure(figsize=(15,10)) for measure in df_numerical: plt.figure(figsize=(10,8)) sns.distplot(df.loc[df["channel"] == 'Swiggy', measure].dropna()) sns.distplot(df.loc[df["channel"] == 'Zomato', measure].dropna()) sns.distplot(df.loc[df["channel"] == 'App/Web', measure].dropna()) plt.legend(["Swiggy", "Zomato", "App/Web"]) plt.show() C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) 7 C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) 8 C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) 9 C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) 10 C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) 11 C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) 12 C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) 13 C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) 14 C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) 15 C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) 16 [15]: df['Timeslot']=pd.cut(df['req_hour'],bins=[0,5,10,15,19,22,24],labels=["Early␣ ↪Morning","Morning","Afternoon","Evening","Night","Late Night"]) [16]: df.info() DatetimeIndex: 32938 entries,-:10:00 to-:49:00 Data columns (total 19 columns): # Column Non-Null Count Dtype --- ------------------- ----32938 non-null object 0 brand 1 invoice_no 32938 non-null int64 2 payment_type 32938 non-null object 3 status 32938 non-null object 4 channel 32938 non-null object 5 my_amount 32938 non-null int64 6 total_tax 32938 non-null float64 7 discount 32938 non-null float64 8 container_charge 32938 non-null int64 9 total 32938 non-null int64 17 10 item_name 32938 non-null object 11 category_name 32938 non-null object 12 item_price 32938 non-null int64 13 item_quantity 32938 non-null int64 14 item_total 32938 non-null int64 15 req_hour 32938 non-null int64 16 day_of_week 32938 non-null object 17 city_outlet 32938 non-null object 18 Timeslot 30366 non-null category dtypes: category(1), float64(2), int64(8), object(8) memory usage: 4.8+ MB [17]: plt.figure(figsize=(5,8)) fig1=sns.countplot(hue='status',x='channel',data = df) fig1.set_title("Status of Payment according to channel") ncount=len(df) for p in fig1.patches: x=p.get_bbox().get_points()[:,0] y=p.get_bbox().get_points()[1,1] fig1.annotate((y),(x.mean(),y),ha='center',va='bottom') plt.show() 18 [18]: plt.figure(figsize=(14, 6)) fig1=sns.countplot(hue ='payment_type',x='channel',data = df) fig1.set_title("Status of Payment mode according to channel") ncount=len(df) for p in fig1.patches: x=p.get_bbox().get_points()[:,0] y=p.get_bbox().get_points()[1,1] 19 fig1.annotate((y),(x.mean(),y),ha='center',va='bottom') plt.show() [19]: percent = round(df.channel.value_counts()/len(df.index)*100,1) sub=['Swiggy','Zomato','Others'] plt.axis("equal") plt.pie(percent , labels=sub,radius=1.6,autopct='%1.2f%%',explode=[0.01,0.01,0. ↪01],startangle=90,shadow=True,counterclock=False,pctdistance=0.6) plt.show() 20 [20]: df.channel.value_counts() [20]: Swiggy 17089 Zomato 14694 App/Web 1155 Name: channel, dtype: int64 [21]: plt.figure(figsize=(10,6)) fig = sns.countplot(hue="channel",x = 'brand', data=df) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.3, \ str(round((i.get_height()),0)), fontsize=14, color='black', rotation=0) plt.xticks(rotation = 90, fontsize = 14) plt.title('No of Orders of diff brands of food for diff channels', fontsize =␣ ↪12) plt.xlabel('Channel of Opn', fontsize = 12) plt.show() 21 [22]: plt.figure(figsize=(15,8)) fig = sns.countplot(hue="channel",x = 'city_outlet', data=df) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.3, \ str(round((i.get_height()),0)), fontsize=15, color='black', rotation=0) plt.xticks(rotation = 90, fontsize = 14) plt.title('No of Orders of diff outlets for diff channels', fontsize = 12) plt.xlabel('Channel of Opn', fontsize = 12) plt.show() 22 [23]: plt.figure(figsize=(15,8)) fig = sns.countplot(hue="channel",x = 'day_of_week', data=df) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.3, \ str(round((i.get_height()),0)), fontsize=15, color='black', rotation=0) plt.title('No of Orders as per diff Channel for days of week', fontsize = 12) plt.xlabel('Channel of Opn', fontsize = 12) plt.show() 23 [24]: plt.figure(figsize=(16,8)) fig = sns.countplot(x="day_of_week",hue = 'Timeslot', data=df) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.3, \ str(round((i.get_height()),0)), fontsize=12, color='black', rotation=0) plt.xticks(rotation = 0, fontsize = 15) plt.title('No of Orders per day per as per diff timeslots', fontsize = 15) plt.xlabel('Channel of Opn', fontsize = 12) plt.show() 24 [25]: plt.figure(figsize=(15,8)) fig = sns.countplot(hue="channel",x = 'Timeslot', data=df) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.3, \ str(round((i.get_height()),0)), fontsize=12, color='black', rotation=0) plt.xticks(rotation = 0, fontsize = 15) plt.title('No of Orders as per diff Channel for diff Timeslots', fontsize = 15) plt.xlabel('Channel of Opn', fontsize = 12) plt.show() 25 [26]: # factor plot of hour and day w.r.t. status plt.figure(figsize=(18,6)) fig = sns.catplot(x = "req_hour", hue = "channel", row = "day_of_week", data =␣ ↪df, kind = "count") plt.xticks(rotation=90, fontsize=12) plt.show() 26 27 [27]: df.info() DatetimeIndex: 32938 entries,-:10:00 to-:49:00 Data columns (total 19 columns): # Column Non-Null Count Dtype --- ------------------- ----0 brand 32938 non-null object 1 invoice_no 32938 non-null int64 2 payment_type 32938 non-null object 3 status 32938 non-null object 4 channel 32938 non-null object 5 my_amount 32938 non-null int64 6 total_tax 32938 non-null float64 7 discount 32938 non-null float64 8 container_charge 32938 non-null int64 9 total 32938 non-null int64 10 item_name 32938 non-null object 11 category_name 32938 non-null object 12 item_price 32938 non-null int64 13 item_quantity 32938 non-null int64 14 item_total 32938 non-null int64 15 req_hour 32938 non-null int64 16 day_of_week 32938 non-null object 17 city_outlet 32938 non-null object 18 Timeslot 30366 non-null category dtypes: category(1), float64(2), int64(8), object(8) memory usage: 4.8+ MB [28]: plt.figure(figsize=(16,8)) fig = sns.barplot(y = 'discount', hue ='brand',x ='channel', data = df,␣ ↪estimator=np.mean) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.4, \ str(round((i.get_height()),0)), fontsize=14, color='black', rotation=0) plt.xticks(rotation=90, fontsize=12) plt.xlabel('city_outlet') plt.ylabel('avg price of orders') plt.title("Avg discount offred by each channel according to diff brands of food␣ ↪",fontsize= 15) 28 plt.xticks(fontsize=12) plt.legend(loc=2,shadow=True,fontsize='medium') plt.show() [29]: plt.figure(figsize=(16,8)) fig = sns.barplot(y = 'discount', hue ='Timeslot',x ='channel', data = df,␣ ↪estimator=np.mean) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.4, \ str(round((i.get_height()),0)), fontsize=14, color='black', rotation=0) plt.xticks(rotation=90, fontsize=12) plt.xlabel('city_outlet') plt.ylabel('avg price of orders') plt.title("Avg discount offred by each channel according to diff Timeslots␣ ↪",fontsize= 15) plt.xticks(fontsize=12) plt.legend(loc=2,shadow=True,fontsize='medium') plt.show() 29 [30]: plt.figure(figsize=(16,8)) fig = sns.barplot(y = 'total', hue ='brand',x ='Timeslot', data= df,␣ ↪estimator=np.mean) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.4, \ str(round((i.get_height()),0)), fontsize=14, color='black', rotation=0) plt.xticks(rotation=90, fontsize=12) plt.xlabel('city_outlet') plt.ylabel('avg price of orders') plt.title("Avg Orders Price for each brand according to TimeSlot",fontsize= 15) plt.xticks(fontsize=12) plt.legend(loc=2,shadow=True,fontsize='medium') plt.show() posx and posy should be finite values posx and posy should be finite values 30 [31]: plt.figure(figsize=(16,8)) fig = sns.barplot(y = 'total', hue ='brand',x ='city_outlet', data= df,␣ ↪estimator=np.mean) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.4, \ str(round((i.get_height()),0)), fontsize=14, color='black', rotation=0) plt.xticks(rotation=90, fontsize=12) plt.xlabel('city_outlet') plt.ylabel('avg price of orders') plt.title("Avg Orders Price from each Outlet vs for each brand",fontsize= 15) plt.xticks(fontsize=12) plt.legend(loc=2,shadow=True,fontsize='medium') plt.show() posx posx posx posx posx posx and and and and and and posy posy posy posy posy posy should should should should should should be be be be be be finite finite finite finite finite finite values values values values values values 31 posx and posy should be finite values posx and posy should be finite values [32]: plt.figure(figsize=(12,6)) fig = sns.barplot(y = 'total', hue ='channel',x ='city_outlet', data= df,␣ ↪estimator=np.mean) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.35, \ str(round((i.get_height()),0)), fontsize=14, color='black', rotation=45) plt.xticks(rotation=90, fontsize=12) plt.xlabel('city_outlet') plt.ylabel('avg price of orders') plt.title("Avg Orders Price from each Outlet vs Channel of␣ ↪Operations",fontsize= 15) plt.xticks(fontsize=12) plt.legend(loc=2,shadow=True,fontsize='medium') plt.show() 32 [33]: plt.figure(figsize=(12,6)) fig = sns.barplot(y = 'total', hue ='channel',x ='brand', data= df,␣ ↪estimator=np.mean) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.35, \ str(round((i.get_height()),0)), fontsize=14, color='black', rotation=45) plt.xticks(rotation=90, fontsize=12) plt.xlabel('city_outlet') plt.ylabel('avg price of orders') plt.title("Avg Orders Price for each brand vs Channel of Operations",fontsize=␣ ↪15) plt.xticks(fontsize=12) plt.legend(loc=2,shadow=True,fontsize='medium') plt.show() 33 [34]: plt.figure(figsize=(12,6)) fig = sns.barplot(y = 'total', hue ='channel',x ='day_of_week', data= df,␣ ↪estimator=np.mean) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.3, \ str(round((i.get_height()),0)), fontsize=12, color='black', rotation=0) plt.xlabel('channel') plt.ylabel('avg price of orders') plt.title("Avg Orders Price Each day of week vs Channel of␣ ↪Operations",fontsize= 15) plt.xticks(fontsize=15) plt.legend(loc=2,shadow=True,fontsize='large') plt.show() 34 [35]: plt.figure(figsize=(12,6)) fig = sns.barplot(y = 'item_quantity', hue ='channel',x ='Timeslot', data= df,␣ ↪estimator=np.mean) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.01 , \ str(round((i.get_height()),0)), fontsize=15, color='black', rotation=0) plt.xlabel('channel') plt.ylabel('quantity of orders') plt.title("Avg Orders Size Each Hour vs Channel of Operations",fontsize= 15) plt.xticks(fontsize=15) plt.legend(loc=2,shadow=True,fontsize='large') plt.show() 35 [36]: plt.figure(figsize=(5,5)) fig = sns.barplot(y = 'total', x ='channel',data= df, estimator=np.mean) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.3, \ str(round((i.get_height()), 2)), fontsize=15, color='black', rotation=0) plt.xlabel('channel') plt.ylabel('avg price of orders') plt.title("Avg Price of Orders vs Channel",fontsize= 15) plt.xticks(fontsize=15) plt.legend(loc=2,shadow=True,fontsize='large') plt.show() No handles with labels found to put in legend. 36 [37]: plt.figure(figsize=(18,8)) fig = sns.barplot(x = 'req_hour', y ='invoice_no',data= df, estimator=len) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.3, \ str(round((i.get_height()), 2)), fontsize=15, color='black', rotation=0) plt.xlabel('hour of day') plt.ylabel('no of orders') plt.title("No of Orders",fontsize= 25) plt.xticks(fontsize=15) plt.legend(loc=2,shadow=True,fontsize='x-large') plt.show() No handles with labels found to put in legend. 37 [38]: df_swiggy = df[df['channel']=='Swiggy'] df_zomato = df[df['channel']=='Zomato'] df_others = df[df['channel']=='App/Web'] [39]: df.info() DatetimeIndex: 32938 entries,-:10:00 to-:49:00 Data columns (total 19 columns): # Column Non-Null Count Dtype --- ------------------- ----0 brand 32938 non-null object 1 invoice_no 32938 non-null int64 2 payment_type 32938 non-null object 3 status 32938 non-null object 4 channel 32938 non-null object 5 my_amount 32938 non-null int64 6 total_tax 32938 non-null float64 7 discount 32938 non-null float64 8 container_charge 32938 non-null int64 9 total 32938 non-null int64 10 item_name 32938 non-null object 11 category_name 32938 non-null object 12 item_price 32938 non-null int64 13 item_quantity 32938 non-null int64 14 item_total 32938 non-null int64 15 req_hour 32938 non-null int- non-null object 16 day_of_week 17 city_outlet 32938 non-null object 38 18 Timeslot 30366 non-null category dtypes: category(1), float64(2), int64(8), object(8) memory usage: 4.8+ MB [40]: Time_pivot_df = df. ↪pivot_table(values=['total'],index=['day_of_week','Timeslot'],columns=['channel'], aggfunc=np.mean) Time_pivot_df [40]: channel day_of_week Timeslot Friday Early Morning Morning Afternoon Evening Night Late Night Early Morning Monday Morning Afternoon Evening Night Late Night Saturday Early Morning Morning Afternoon Evening Night Late Night Sunday Early Morning Morning Afternoon Evening Night Late Night Thursday Early Morning Morning Afternoon Evening Night Late Night Tuesday Early Morning Morning Afternoon Evening Night total App/Web Swiggy Zomato NaN NaN- NaN- NaN- NaN- - - 39 Wednesday Late Night Early Morning Morning Afternoon Evening Night Late Night - NaN- - - [41]: outlet_pivot_df = df. ↪pivot_table(values=['total'],index=['city_outlet','brand'],columns=['channel'], aggfunc=np.mean) outlet_pivot_df [41]: channel city_outlet Ahmedabad - Bodakdev brand Bowl_Meals Fast_Food North_Indian_Meals Thali_Meals Ahmedabad - Chandkheda Bowl_Meals Fast_Food North_Indian_Meals Thali_Meals Baroda - Alkapuri Bowl_Meals Fast_Food North_Indian_Meals Thali_Meals Indore- Vijaynagar Bowl_Meals Fast_Food North_Indian_Meals Jaipur - Vaishali Nagar Bowl_Meals Fast_Food North_Indian_Meals Maharashtra- Nagpur Bowl_Meals Fast_Food North_Indian_Meals Pune -Kharadi Bowl_Meals Fast_Food North_Indian_Meals Thali_Meals Surat - Vesu Bowl_Meals Fast_Food North_Indian_Meals total App/Web Swiggy Zomato - - NaN- NaN- - NaN- NaN- [42]: rels = Time_pivot_df.to_excel('Tealbox_pivot_df_time.xlsx') 40 [43]: pels = outlet_pivot_df.to_excel('Tealbox_outlet.xlsx') [44]: df.payment_type.value_counts() [44]: Online 31579 Cash 847 UPI 389 123 CARD Name: payment_type, dtype: int64 [45]: df_swiggy.info() DatetimeIndex: 17089 entries,-:14:00 to-:49:00 Data columns (total 19 columns): # Column Non-Null Count Dtype --- ------------------- ----0 brand 17089 non-null object 1 invoice_no 17089 non-null int64 2 payment_type 17089 non-null object 3 status 17089 non-null object 4 channel 17089 non-null object 5 my_amount 17089 non-null int64 6 total_tax 17089 non-null float64 7 discount 17089 non-null float64 8 container_charge 17089 non-null int64 9 total 17089 non-null int64 10 item_name 17089 non-null object 11 category_name 17089 non-null object 12 item_price 17089 non-null int64 13 item_quantity 17089 non-null int64 14 item_total 17089 non-null int64 15 req_hour 17089 non-null int64 16 day_of_week 17089 non-null object 17089 non-null object 17 city_outlet 18 Timeslot 16377 non-null category dtypes: category(1), float64(2), int64(8), object(8) memory usage: 2.5+ MB [46]: df.channel.value_counts() [46]: Swiggy 17089 Zomato 14694 App/Web 1155 Name: channel, dtype: int64 41 [47]: plt.figure(figsize=(8,6)) sizes = df.channel.value_counts() # Setting labels for items in Chart labels = ['Swiggy', 'Zomato', 'Others'] # colors colors = ['#FF0000', '#0000FF', '#ADFF2F'] # explosion explode = (0.0, 0.0,0.0) # Pie Chart plt.pie(sizes, colors=colors, labels=labels, autopct='%1.1f%%', pctdistance=0.85, explode=explode) # draw circle centre_circle = plt.Circle((0, 0), 0.70, fc='white') fig = plt.gcf() # Adding Circle in Pie chart fig.gca().add_artist(centre_circle) # Adding Title of chart plt.title('Favourite brand of food for swiggy customers') # Add Legends #plt.legend(labels, loc="lower left") # Displaying Chart plt.show() 42 [48]: plt.figure(figsize=(8,6)) sizes = df_swiggy.brand.value_counts() # Setting labels for items in Chart labels = ['Bowl_Meals', 'North_Indian_Meals', 'Fast_Food', 'Thali_Meals'] # colors colors = ['#FF0000', '#0000FF', '#FFFF00', '#ADFF2F'] # explosion explode = (0.0, 0.0, 0.0, 0.0) # Pie Chart plt.pie(sizes, colors=colors, labels=labels, autopct='%1.1f%%', pctdistance=0.85, explode=explode) # draw circle 43 centre_circle = plt.Circle((0, 0), 0.70, fc='white') fig = plt.gcf() # Adding Circle in Pie chart fig.gca().add_artist(centre_circle) # Adding Title of chart plt.title('Favourite brand of food for swiggy customers') # Add Legends #plt.legend(labels, loc="lower left") # Displaying Chart plt.show() [49]: plt.figure(figsize=(8,6)) sizes = df_zomato.brand.value_counts() 44 # Setting labels for items in Chart #labels = ['Bowl_Meals', 'North_Indian_Meals', 'Fast_Food', 'Thali_Meals'] # colors colors = ['#FF0000', '#0000FF', '#FFFF00', '#ADFF2F'] # explosion explode = (0.0, 0.0, 0.0, 0.0) # Pie Chart plt.pie(sizes, colors=colors, labels=labels, autopct='%1.1f%%', pctdistance=0.85, explode=explode) # draw circle centre_circle = plt.Circle((0, 0), 0.70, fc='white') fig = plt.gcf() # Adding Circle in Pie chart fig.gca().add_artist(centre_circle) # Adding Title of chart plt.title('Favourite brand of food for zomato customers') # Add Legends #plt.legend(labels, loc="lower left") # Displaying Chart plt.show() 45 [50]: #now plotting for requests per hour according to status plt.figure(figsize=(18,6),dpi=80,facecolor='y',edgecolor='k') fig = sns. ↪barplot(x='city_outlet',y='total',hue='brand',data=df_swiggy,estimator = np. ↪mean) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.2, \ str(round((i.get_height()), 2)), fontsize=14, color='black', rotation=0) plt.title("Mean Order value for each Brand from each Outlet for Swiggy",␣ ↪fontsize= 14) plt.ylabel("Price") plt.xticks(rotation = 90, fontsize = 14) plt.show() posx and posy should be finite values 46 posx posx posx posx posx posx posx posx posx posx posx and and and and and and and and and and and posy posy posy posy posy posy posy posy posy posy posy should should should should should should should should should should should be be be be be be be be be be be finite finite finite finite finite finite finite finite finite finite finite values values values values values values values values values values values [51]: #plots of frquency of request plt.figure(figsize=(12,6),dpi=80,facecolor='y',edgecolor='k') fig = sns. ↪barplot(x='brand',y='total',hue='channel',data=df[(df['city_outlet']=='Baroda␣ ↪- Alkapuri')],estimator=np.mean) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.2, \ str(round((i.get_height()), 0)), fontsize=10, color='black', rotation=0) plt.title("'Mean Order value for each Brand each channel for Baroda -␣ ↪Alkapuri'") 47 plt.ylabel("Price") [51]: Text(0, 0.5, 'Price') [52]: plt.figure(figsize=(12,6),dpi=80,facecolor='y',edgecolor='k') fig = sns. ↪barplot(x='brand',y='total',hue='channel',data=df[(df['city_outlet']=='Indore-␣ ↪Vijaynagar')],estimator=np.mean) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.2, \ str(round((i.get_height()), 0)), fontsize=10, color='black', rotation=0) plt.title("'Mean Order value for each Brand each channel for Indore-␣ ↪Vijaynagar'") plt.ylabel("Price") [52]: Text(0, 0.5, 'Price') 48 [53]: plt.figure(figsize=(12,6),dpi=80,facecolor='y',edgecolor='k') fig = sns. ↪barplot(x='brand',y='total',hue='channel',data=df[(df['city_outlet']=='Ahmedabad␣ ↪- Chandkheda')],estimator=np.mean) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.2, \ str(round((i.get_height()), 0)), fontsize=10, color='black', rotation=0) plt.title("'Mean Order value for each Brand each channel for Ahmedabad -␣ ↪Chandkheda'") plt.ylabel("Price") [53]: Text(0, 0.5, 'Price') posx posx posx posx and and and and posy posy posy posy should should should should be be be be finite finite finite finite values values values values 49 [54]: plt.figure(figsize=(12,6),dpi=80,facecolor='y',edgecolor='k') fig = sns. ↪barplot(x='brand',y='total',hue='channel',data=df[(df['city_outlet']=='Pune␣ ↪-Kharadi')],estimator=np.mean) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.2, \ str(round((i.get_height()), 0)), fontsize=10, color='black', rotation=0) plt.title("'Mean Order value for each Brand each channel for Pune -Kharadi'") plt.ylabel("Price") [54]: Text(0, 0.5, 'Price') posx posx posx posx and and and and posy posy posy posy should should should should be be be be finite finite finite finite values values values values 50 [55]: #plots of frquency of request plt.figure(figsize=(12,6),dpi=80,facecolor='y',edgecolor='k') fig = sns. ↪barplot(x='brand',y='total',hue='channel',data=df[(df['city_outlet']=='Ahmedabad␣ ↪- Bodakdev')],estimator=np.mean) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.2, \ str(round((i.get_height()), 0)), fontsize=10, color='black', rotation=0) plt.title("'Mean Order value for each Brand each channel for Ahmedabad -␣ ↪Bodakdev'") plt.ylabel("Price") [55]: Text(0, 0.5, 'Price') 51 [56]: #plots of frquency of request plt.figure(figsize=(12,6),dpi=80,facecolor='y',edgecolor='k') fig = sns. ↪barplot(x='brand',y='total',hue='channel',data=df[(df['city_outlet']=='Jaipur␣ ↪- Vaishali Nagar')],estimator=np.mean) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.2, \ str(round((i.get_height()), 0)), fontsize=10, color='black', rotation=0) plt.title("'Mean Order value for each Brand each channel for Jaipur - Vaishali␣ ↪Nagar'") plt.ylabel("Price") [56]: Text(0, 0.5, 'Price') 52 [57]: #plots of frquency of request plt.figure(figsize=(12,6),dpi=80,facecolor='y',edgecolor='k') fig = sns. ↪barplot(x='brand',y='total',hue='channel',data=df[(df['city_outlet']=='Surat␣ ↪- Vesu')],estimator=np.mean) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.2, \ str(round((i.get_height()), 0)), fontsize=10, color='black', rotation=0) plt.title("'Mean Order value for each Brand each channel for Surat - Vesu'") plt.ylabel("Price") [57]: Text(0, 0.5, 'Price') 53 [58]: #plots of frquency of request plt.figure(figsize=(12,6),dpi=80,facecolor='y',edgecolor='k') fig = sns. ↪barplot(x='brand',y='total',hue='channel',data=df[(df['city_outlet']=='Maharashtra-␣ ↪Nagpur')],estimator=np.mean) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.2, \ str(round((i.get_height()), 0)), fontsize=10, color='black', rotation=0) plt.title("'Mean Order value for each Brand each channel for Maharashtra-␣ ↪Nagpur'") plt.ylabel("Price") [58]: Text(0, 0.5, 'Price') 54 [59]: df_swiggy.info() DatetimeIndex: 17089 entries,-:14:00 to-:49:00 Data columns (total 19 columns): # Column Non-Null Count Dtype --- ------------------- ----0 brand 17089 non-null object 1 invoice_no 17089 non-null int64 2 payment_type 17089 non-null object 3 status 17089 non-null object 4 channel 17089 non-null object 5 my_amount 17089 non-null int64 6 total_tax 17089 non-null float64 7 discount 17089 non-null float64 8 container_charge 17089 non-null int64 9 total 17089 non-null int64 10 item_name 17089 non-null object 11 category_name 17089 non-null object 12 item_price 17089 non-null int64 13 item_quantity 17089 non-null int64 14 item_total 17089 non-null int64 15 req_hour 17089 non-null int64 16 day_of_week 17089 non-null object 17 city_outlet 17089 non-null object 18 Timeslot 16377 non-null category dtypes: category(1), float64(2), int64(8), object(8) memory usage: 2.5+ MB 55 [60]: #plots of frquency of request plt.figure(figsize=(18,8), facecolor='y', edgecolor='r') fig = sns. ↪barplot(hue='brand',y='total',x='city_outlet',data=df_swiggy,estimator=np. ↪mean) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.2, \ str(round((i.get_height()), 0)), fontsize=14, color='black', rotation=0) plt.xticks(rotation = 90, fontsize = 14) plt.title("'Mean Order value for each Brand each outlet for diff outlets for␣ ↪swiggy'", fontsize = 15) plt.ylabel("Price") plt.show() posx posx posx posx posx posx posx posx posx posx posx posx and and and and and and and and and and and and posy posy posy posy posy posy posy posy posy posy posy posy should should should should should should should should should should should should be be be be be be be be be be be be finite finite finite finite finite finite finite finite finite finite finite finite values values values values values values values values values values values values 56 [61]: #plots of frquency of request plt.figure(figsize=(18,8), facecolor='y', edgecolor='r') fig = sns. ↪barplot(hue='brand',y='total',x='city_outlet',data=df_zomato,estimator=np. ↪mean) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.2, \ str(round((i.get_height()), 0)), fontsize=14, color='black', rotation=0) plt.xticks(rotation = 90, fontsize = 14) plt.title("'Mean Order value for each Brand each outlet for diff outlets for␣ ↪Zomato'", fontsize = 15) plt.ylabel("Price") plt.show() posx posx posx posx posx and and and and and posy posy posy posy posy should should should should should be be be be be finite finite finite finite finite values values values values values 57 posx posx posx posx posx posx posx and and and and and and and posy posy posy posy posy posy posy should should should should should should should be be be be be be be finite finite finite finite finite finite finite values values values values values values values [63]: df.brand.value_counts() [63]: Bowl_Meals 13594 North_Indian_Meals 12828 Fast_Food 5261 Thali_Meals 1255 Name: brand, dtype: int64 [69]: #plots of frquency of request plt.figure(figsize=(18,8), facecolor='y', edgecolor='r') fig = sns.barplot(hue='brand',y='total',x='channel',data=df,estimator=np.mean) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.2, \ 58 str(round((i.get_height()), 0)), fontsize=14, color='black', rotation=0) plt.title("'Mean Order value for each Brand each outlet for diff channels'",␣ ↪fontsize = 15) plt.ylabel("Price") plt.show() [74]: df.info() DatetimeIndex: 32938 entries,-:10:00 to-:49:00 Data columns (total 19 columns): # Column Non-Null Count Dtype --- ------------------- ----0 brand 32938 non-null object 1 invoice_no 32938 non-null int64 2 payment_type 32938 non-null object 3 status 32938 non-null object 4 channel 32938 non-null object 5 my_amount 32938 non-null int64 6 total_tax 32938 non-null float64 7 discount 32938 non-null float64 8 container_charge 32938 non-null int64 9 total 32938 non-null int64 10 item_name 32938 non-null object 11 category_name 32938 non-null object 12 item_price 32938 non-null int64 13 item_quantity 32938 non-null int64 14 item_total 32938 non-null int64 59 15 req_hour 32938 non-null int64 16 day_of_week 32938 non-null object 17 city_outlet 32938 non-null object 30366 non-null category 18 Timeslot dtypes: category(1), float64(2), int64(8), object(8) memory usage: 4.8+ MB [88]: #plots of frquency of request plt.figure(figsize=(8,8),dpi=80,facecolor='cyan',edgecolor='k') fig = sns.countplot(x='channel', hue = 'brand',data=df) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.2, \ str(round((i.get_height()), 0)), fontsize=10, color='black', rotation=0) plt.title("'No of Order for diff brand of food for diff channels'") plt.legend(loc=2,shadow=True,fontsize='medium') plt.ylabel("Price") [88]: Text(0, 0.5, 'Price') 60 [93]: df.brand.value_counts() [93]: Bowl_Meals 13594 North_Indian_Meals 12828 Fast_Food 5261 Thali_Meals 1255 Name: brand, dtype: int64 [96]: No handles with labels found to put in legend. [96]: Text(0, 0.5, 'Price') 61 [99]: #plots of frquency of request plt.figure(figsize=(12,10),dpi=80,facecolor='cyan',edgecolor='k') fig = sns.barplot(x =␣ ↪'category_name',y='total',data=df_zomato[(df_zomato['brand']=='North_Indian_Meals')],estimat ↪mean) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.2, \ 62 str(round((i.get_height()), 0)), fontsize=10, color='black', rotation=0) plt.title("'Mean Order value for diff category in Bowl meals for Zomato'") plt.legend(loc=2,shadow=True,fontsize='medium') plt.xticks(rotation= 90, fontsize =10) plt.ylabel("Price") No handles with labels found to put in legend. [99]: Text(0, 0.5, 'Price') 63 [103]: #plots of frquency of request plt.figure(figsize=(12,6),dpi=80,facecolor='cyan',edgecolor='k') fig = sns.barplot(x = 'channel',hue = 'Timeslot',y='total',data=df,estimator=np. ↪mean) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.2, \ str(round((i.get_height()), 0)), fontsize=10, color='black', rotation=0) plt.title("'Mean Order value for diff Timeslots for all the channels'") plt.legend(loc=2,shadow=True,fontsize='medium') plt.xticks( fontsize =10) plt.ylabel("Price") [103]: Text(0, 0.5, 'Price') [100]: df.info() DatetimeIndex: 32938 entries,-:10:00 to-:49:00 Data columns (total 19 columns): # Column Non-Null Count Dtype --- ------------------- ----0 brand 32938 non-null object 1 invoice_no 32938 non-null int64 2 payment_type 32938 non-null object 64 3 status 32938 non-null object 4 channel 32938 non-null object 5 my_amount 32938 non-null int- non-null float64 6 total_tax 7 discount 32938 non-null float64 8 container_charge 32938 non-null int64 9 total 32938 non-null int64 10 item_name 32938 non-null object 11 category_name 32938 non-null object 12 item_price 32938 non-null int64 13 item_quantity 32938 non-null int64 14 item_total 32938 non-null int64 15 req_hour 32938 non-null int64 16 day_of_week 32938 non-null object 17 city_outlet 32938 non-null object 18 Timeslot 30366 non-null category dtypes: category(1), float64(2), int64(8), object(8) memory usage: 4.8+ MB [104]: df_zomato.brand.value_counts() [104]: North_Indian_Meals 6236 Bowl_Meals 5274 Fast_Food 2736 Thali_Meals 448 Name: brand, dtype: int64 [115]: percent = round(df_zomato.brand.value_counts()) sub=['North_Indian_Meals','Bowl_Meals','Fast_Food','Thali_Meals'] plt.axis("equal") plt.pie(percent,labels= sub ,radius=1.6,autopct='%1.2f%%',explode=[0.01,0.01,0. ↪01,0.01],startangle=90,shadow=True,counterclock=False,pctdistance=0.6) plt.show() 65 [136]: #plots of frquency of request plt.figure(figsize=(10,6),dpi=80,facecolor='cyan',edgecolor='k') fig = sns.countplot(x =␣ ↪'category_name',data=df_zomato[(df_zomato['brand']=='North_Indian_Meals')]) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.2, \ str(round((i.get_height()), 0)), fontsize=10, color='black', rotation=0) plt.title("'No of Order for diff category in North_Indian_Meals for Zomato'") plt.legend(loc=2,shadow=True,fontsize='medium') plt.xticks(rotation= 90, fontsize =10) plt.ylabel("Price") No handles with labels found to put in legend. [136]: Text(0, 0.5, 'Price') 66 [119]: percent = round(df_swiggy.brand.value_counts()) sub=['Bowl_Meals', 'North_Indian_Meals','Fast_Food','Thali_Meals'] plt.axis("equal") plt.pie(percent,labels= sub ,radius=1.6,autopct='%1.2f%%',explode=[0.01,0.01,0. ↪01,0.01],startangle=90,shadow=True,counterclock=False,pctdistance=0.6) plt.show() 67 [137]: #plots of frquency of request plt.figure(figsize=(10,6),dpi=80,facecolor='cyan',edgecolor='k') fig = sns.countplot(x =␣ ↪'category_name',data=df_swiggy[(df_swiggy['brand']=='Bowl_Meals')]) for i in fig .patches: # get_x pulls left or right; get_height pushes up or down fig.text(i.get_x()+.01, i.get_height()+0.2, \ str(round((i.get_height()), 0)), fontsize=10, color='black', rotation=0) plt.title("'No of Order for diff category in Bowl_Meals for Swiggy'") plt.legend(loc=2,shadow=True,fontsize='medium') plt.xticks(rotation= 90, fontsize =10) plt.ylabel("Price") No handles with labels found to put in legend. [137]: Text(0, 0.5, 'Price') 68 [ ]: df_swiggy.discount.describe() [134]: plt.figure(figsize=(6,5)) ax = sns.distplot(df_swiggy['discount'], hist=True) plt.show() C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) 69 [135]: plt.figure(figsize=(6,5)) ax = sns.distplot(df_zomato['discount'], hist=True) plt.show() C:\Users\Lenovo\anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms). warnings.warn(msg, FutureWarning) 70 [ ]: 71