Covid 19 - India State Wise analysis

 

Covid EDA on Indian Covid Dataset

In [2]:

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
sns.set_style('darkgrid')
import warnings
warnings.filterwarnings("ignore")

India reported 34,703 cases, 553 deaths in last 24 hours. As cases and fatalities dip, the recovery rate has risen to 97.17%. The country's active caseload has declined to 4,64,357.

Don't be like them. Stay Home Stay Safe. Now let's start our EDA

In [3]:

data= pd.read_csv("../input/latest-covid19-india-statewise-data/Latest Covid-19 India Status.csv")

In [4]:

data.head()

Out[4]:

	State/UTs	Total Cases	Active	Discharged	Deaths	Active Ratio	Discharge Ratio	Death Ratio
0	Maharashtra	6061404	119558	5819901	121945	1.97	96.02	2.01
1	Kerala	2924165	101343	2809587	13235	3.47	96.08	0.45
2	Karnataka	2843810	76528	2732242	35040	2.69	96.08	1.23
3	Tamil Nadu	2479696	38191	2408886	32619	1.54	97.14	1.32
4	Andhra Pradesh	1889513	38338	1838469	12706	2.03	97.30	0.67

In [5]:

data.isnull().sum()

Out[5]:

State/UTs          0
Total Cases        0
Active             0
Discharged         0
Deaths             0
Active Ratio       0
Discharge Ratio    0
Death Ratio        0
dtype: int64

In [6]:

data.describe()

Out[6]:

	Total Cases	Active	Discharged	Deaths	Active Ratio	Discharge Ratio	Death Ratio
count	3.600000e+01	36.000000	3.600000e+01	36.000000	36.000000	36.000000	36.000000
mean	8.447676e+05	14534.916667	8.191366e+05	11096.083333	2.902500	95.844444	1.252778
std	1.209487e+06	28312.184103	1.164892e+06	21092.496707	4.069466	3.977242	0.567247
min	7.467000e+03	31.000000	7.308000e+03	4.000000	0.070000	79.000000	0.040000
25%	5.863075e+04	1422.500000	5.664575e+04	776.500000	0.407500	96.065000	0.927500
50%	4.270470e+05	2779.000000	4.085465e+05	4431.000000	1.425000	97.200000	1.315000
75%	9.629365e+05	7856.000000	9.502918e+05	12838.250000	3.252500	98.185000	1.660000
max	6.061404e+06	119558.000000	5.819901e+06	121945.000000	20.550000	99.540000	2.700000

Some stats gathered from the dataset

In [7]:

plt.figure(figsize=(6,6))
sns.heatmap(data.corr(),annot=True,cmap='summer')
plt.title("Heatmap of the dataset")

Out[7]:

Text(0.5, 1.0, 'Heatmap of the dataset')

High correlation between Active and Discharged, Active and Deaths, and Total cases and Active and Deaths

In [8]:

sns.distplot(data.Active,color='Blue')
plt.title("Active cases in India",fontsize=15)

Out[8]:

Text(0.5, 1.0, 'Active cases in India')

In [9]:

sns.distplot(data.Deaths,color='Red')
plt.title("Death cases in India",fontsize=15)

Out[9]:

Text(0.5, 1.0, 'Death cases in India')

In [10]:

sns.distplot(data['Discharged'],color='Yellow')
plt.title("Discharged cases in India",fontsize=15)

Out[10]:

Text(0.5, 1.0, 'Discharged cases in India')

The dataset is a bit right skewed

In [11]:

plt.figure(figsize=(10,10))
sns.barplot(x='State/UTs',y='Total Cases',palette='CMRmap',data=data)
plt.title("The total cases as per State/UTs are",fontsize=15)
plt.xticks(rotation=90)

Out[11]:

(array([ 0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16,
        17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33,
        34, 35]),
 [Text(0, 0, 'Maharashtra'),
  Text(1, 0, 'Kerala'),
  Text(2, 0, 'Karnataka'),
  Text(3, 0, 'Tamil Nadu'),
  Text(4, 0, 'Andhra Pradesh'),
  Text(5, 0, 'Uttar Pradesh'),
  Text(6, 0, 'West Bengal'),
  Text(7, 0, 'Delhi'),
  Text(8, 0, 'Chhattisgarh'),
  Text(9, 0, 'Rajasthan'),
  Text(10, 0, 'Odisha'),
  Text(11, 0, 'Gujarat'),
  Text(12, 0, 'Madhya Pradesh'),
  Text(13, 0, 'Haryana'),
  Text(14, 0, 'Bihar'),
  Text(15, 0, 'Telengana'),
  Text(16, 0, 'Punjab'),
  Text(17, 0, 'Assam'),
  Text(18, 0, 'Jharkhand'),
  Text(19, 0, 'Uttarakhand'),
  Text(20, 0, 'Jammu and Kashmir'),
  Text(21, 0, 'Himachal Pradesh'),
  Text(22, 0, 'Goa'),
  Text(23, 0, 'Puducherry'),
  Text(24, 0, 'Manipur'),
  Text(25, 0, 'Tripura'),
  Text(26, 0, 'Chandigarh'),
  Text(27, 0, 'Meghalaya'),
  Text(28, 0, 'Arunachal Pradesh'),
  Text(29, 0, 'Nagaland'),
  Text(30, 0, 'Sikkim'),
  Text(31, 0, 'Mizoram'),
  Text(32, 0, 'Ladakh'),
  Text(33, 0, 'Dadra and Nagar Haveli and Daman and Diu'),
  Text(34, 0, 'Lakshadweep'),
  Text(35, 0, 'Andaman and Nicobar')])

Maharashtra has the highest number of total cases in India, Kerala and Karnataka rank second and third although there is not much difference between their total counts.

In [12]:

plt.figure(figsize=(7,7))
labels = data.index
plt.pie(x='Active',data=data[:5],labels='State/UTs',startangle=90,autopct='%.1f%%')
plt.title("Active Cases in Top 5 states in India", fontsize = 24) 
plt.tight_layout() 
plt.show()

32% of the active cases are in Maharshtra alone, 27.1% and 20.5% are in Kerela and Karnataka

In [13]:

plt.figure(figsize=(7,7))
labels = data.index
plt.pie(x='Total Cases',data=data[:5],labels='State/UTs',startangle=90,autopct='%.1f%%')
plt.title("Total Cases in Top 5 states in India", fontsize = 24) 
plt.tight_layout() 
plt.show()

Maharashtra has the highest count with 37.4% of total cases

In [14]:

plt.figure(figsize=(7,7))
labels = data.index
plt.pie(x='Discharged',data=data[:5],labels='State/UTs',startangle=90,autopct='%.1f%%')
plt.title("People Discharged in Top 5 states in India", fontsize = 24) 
plt.tight_layout() 
plt.show()

It's good to see the most affected state with Top Discharge Rate

In [15]:

plt.figure(figsize=(7,7))
labels = data.index
plt.pie(x='Deaths',data=data[:5],labels='State/UTs',startangle=90,autopct='%.1f%%')
plt.title("Deaths in Top 5 states in India", fontsize = 24) 
plt.tight_layout() 
plt.show()

56.6% of the total deaths have happened in Maharashtra followed by 16.3% in Karanataka

In [16]:

sns.scatterplot(x='Active Ratio',y='Death Ratio',data=data[:10],palette='Spectral',legend='brief',hue='State/UTs')
plt.title("Top 10 Active Ratio to Deaths Ratio in India",fontsize=15)

Out[16]:

Text(0.5, 1.0, 'Top 10 Active Ratio to Deaths Ratio in India')

Maharashtra has the highest Active to Death Ratio in the country. Delhi standing at second and Kerela with the lowest ratio

In [17]:

sort_data= data.sort_values(by='Active Ratio',ascending=False)
sns.barplot(x='State/UTs',y='Active Ratio',data=data[:10],palette='copper',hue='Death Ratio')
plt.xticks(rotation=90)
plt.title("Top 10 Active Ratio to Deaths Ratio in India",fontsize=15)

Out[17]:

Text(0.5, 1.0, 'Top 10 Active Ratio to Deaths Ratio in India')

---

---

In [18]:

sns.lineplot(y='Active Ratio',data=data[:10],x='State/UTs')
plt.xticks(rotation=90)
plt.title("Line Plot for Active Ratio",fontsize=15)

Out[18]:

Text(0.5, 1.0, 'Line Plot for Active Ratio')

Kerela has the highest Active Ratio in the country

In [19]:

sns.barplot(x='State/UTs',y='Death Ratio',data=data[:10],hue='Death Ratio')
plt.xticks(rotation=90)
plt.title("Top 10 Death Ratio in India",fontsize=15)

Out[19]:

Text(0.5, 1.0, 'Top 10 Death Ratio in India')

Death ratio in Maharashtra and Delhi are the highest

In [20]:

sns.lineplot(y='Death Ratio',data=data[:10],x='State/UTs')
plt.xticks(rotation=90)
plt.title(" Lineplot for Death Ratio",fontsize=15)

Out[20]:

Text(0.5, 1.0, ' Lineplot for Death Ratio')

In [21]:

sns.scatterplot(x='Discharge Ratio',y='Active Ratio',data=data[:10],palette='twilight',hue='State/UTs')
plt.title("Discharge Ratio for top 10 states",fontsize=15)

Out[21]:

Text(0.5, 1.0, 'Discharge Ratio for top 10 states')

People in Kerela and Karnataka have been discharged the most from Hospitals

In [22]:

sns.barplot(x='State/UTs',y='Death Ratio',data=data[:10],hue='Discharge Ratio')
plt.xticks(rotation=90)
plt.title("Top 10 Discharge Ratio in India",fontsize=15)

Out[22]:

Text(0.5, 1.0, 'Top 10 Discharge Ratio in India')

In [23]:

sns.lineplot(y='Discharge Ratio',data=data[:10],x='State/UTs')
plt.xticks(rotation=90)
plt.title(" Lineplot for Discharge Ratio",fontsize=15)

Out[23]:

Text(0.5, 1.0, ' Lineplot for Discharge Ratio')

Rajasthan has the highest Discharge Ratio in the country. Good to see people coming back to home from hospitals

In [24]:

data['Recovered']=data['Total Cases']-(data['Active']+data['Deaths'])

In [25]:

data.head()

Out[25]:

	State/UTs	Total Cases	Active	Discharged	Deaths	Active Ratio	Discharge Ratio	Death Ratio	Recovered
0	Maharashtra	6061404	119558	5819901	121945	1.97	96.02	2.01	5819901
1	Kerala	2924165	101343	2809587	13235	3.47	96.08	0.45	2809587
2	Karnataka	2843810	76528	2732242	35040	2.69	96.08	1.23	2732242
3	Tamil Nadu	2479696	38191	2408886	32619	1.54	97.14	1.32	2408886
4	Andhra Pradesh	1889513	38338	1838469	12706	2.03	97.30	0.67	1838469

Recovered is a custom column for the difference between total cases and sum of active and death cases

In [26]:

plt.figure(figsize=(10,10))
sns.barplot(x='State/UTs',y='Recovered',palette='CMRmap',data=data)
plt.title("The Recovered cases as per State/UTs are",fontsize=15)
plt.xticks(rotation=90)

Out[26]:

(array([ 0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16,
        17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33,
        34, 35]),
 [Text(0, 0, 'Maharashtra'),
  Text(1, 0, 'Kerala'),
  Text(2, 0, 'Karnataka'),
  Text(3, 0, 'Tamil Nadu'),
  Text(4, 0, 'Andhra Pradesh'),
  Text(5, 0, 'Uttar Pradesh'),
  Text(6, 0, 'West Bengal'),
  Text(7, 0, 'Delhi'),
  Text(8, 0, 'Chhattisgarh'),
  Text(9, 0, 'Rajasthan'),
  Text(10, 0, 'Odisha'),
  Text(11, 0, 'Gujarat'),
  Text(12, 0, 'Madhya Pradesh'),
  Text(13, 0, 'Haryana'),
  Text(14, 0, 'Bihar'),
  Text(15, 0, 'Telengana'),
  Text(16, 0, 'Punjab'),
  Text(17, 0, 'Assam'),
  Text(18, 0, 'Jharkhand'),
  Text(19, 0, 'Uttarakhand'),
  Text(20, 0, 'Jammu and Kashmir'),
  Text(21, 0, 'Himachal Pradesh'),
  Text(22, 0, 'Goa'),
  Text(23, 0, 'Puducherry'),
  Text(24, 0, 'Manipur'),
  Text(25, 0, 'Tripura'),
  Text(26, 0, 'Chandigarh'),
  Text(27, 0, 'Meghalaya'),
  Text(28, 0, 'Arunachal Pradesh'),
  Text(29, 0, 'Nagaland'),
  Text(30, 0, 'Sikkim'),
  Text(31, 0, 'Mizoram'),
  Text(32, 0, 'Ladakh'),
  Text(33, 0, 'Dadra and Nagar Haveli and Daman and Diu'),
  Text(34, 0, 'Lakshadweep'),
  Text(35, 0, 'Andaman and Nicobar')])

In [27]:

sns.jointplot(x='Recovered',y='State/UTs',data=data)
plt.xticks(rotation=90)

Out[27]:

(array([0., 1., 2.]), [])

In [28]:

sns.barplot(x='State/UTs',y='Recovered',data=data[:10])
plt.xticks(rotation=90)
plt.title("Top 10 most affected States",fontsize=15)

Out[28]:

Text(0.5, 1.0, 'Top 10 most affected States')

In [29]:

plt.figure(figsize=(10,5))
sns.pointplot(x='State/UTs',y='Recovered',data=data[:10],color='Red')
plt.xticks(rotation=90)
plt.title("recovered in line Plot",fontsize=15)

Out[29]:

Text(0.5, 1.0, 'recovered in line Plot')

In [30]:

plt.figure(figsize=(7,7))
labels = data.index
plt.pie(x='Recovered',data=data[:10],labels='State/UTs',startangle=90,autopct='%.1f%%')
plt.title("Recovery in Top 5 states in India", fontsize = 24) 
plt.tight_layout() 
plt.show()

Summary

1. Maharashtra has been affected the most in India. 2. Kerela stands the second followed by Karnataka 3. Rajashthan has the best discharge ratio 4. Discharge to Active Ratio is highest in Kerela followed by Karnataka. 5. Death ratio is highest in Maharashtra and Chattisgarh. 6. Most Active Ratio is in Kerela and least is in Delhi and Uttar Pradesh 7. Recovery is also highest in Maharashtra, Kerala and Karnataka stands at 2nd and 3rd respectively

In [ ]: