Skip to content

Latest commit

 

History

History
324 lines (211 loc) · 6.85 KB

EDA project of Covid-19 in INDIA.md

File metadata and controls

324 lines (211 loc) · 6.85 KB
Raw
"""importing libraries"""

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import datetime as dt
## Importing Data. 
df_india = pd.read_csv("C:/Users/kamle/Downloads/archive/covid_19_india.csv")
df_india
## Getting familiar with Data
"Checking the range of data"

df_india.shape
"Getting information about Data type and non-null values"

df_india.info()
"Getting numeric column detials "

df_india.describe()
"Getting information of null values in Dataset"

df_india.isna().sum()
###  Note: There is no null value's in dataset
"finding unique values from 'State/UnionTerritory' column"

df_india['State/UnionTerritory'].unique(),df_india['State/UnionTerritory'].nunique()
"Correcting spelling mistakes or impurities"

state_correction_dict = {
    'Bihar****':'Bihar',
    'Dadra and Nagar Haveli':'Dadra and Nagar Haveli and Daman and Diu',
    'Madhya Pradesh***':'Madhya Pradesh',
    'Maharashtra***':'Maharashtra',
    'Karanataka':'Karnataka'
}

def state_correction(state):
    try:
        return state_correction_dict[state]
    except:
        return state
    
df_india['State/UnionTerritory'] = df_india['State/UnionTerritory'].apply(state_correction)
df_india['State/UnionTerritory'].nunique()
### Note: Here we have corrected spelling mistakes in columns.
"Changing the format of date"

df_india['Date'] = pd.to_datetime(df_india['Date'])
df_india['Date'] = df_india['Date'].dt.strftime('%d-%m-%Y')
df_india['Date']
"Removing unwanted columns from dataset using 'drop'."

df_india.drop(['Time','ConfirmedIndianNational','ConfirmedForeignNational'],axis = 1,inplace = True)
df_india
"Getting only Numeric columns"

num = df_india.select_dtypes(exclude = object)
num
"Getting only categorical data"

obj = df_india.select_dtypes(include = object)
obj
## Data Manipulation 
"Identifying active cases , We counted the values by using values in confirmed, cured, deaths column"

df_india['Active'] = df_india['Confirmed'] - df_india['Cured'] - df_india['Deaths']
df_india
### Note: We can now check the active cases in each state
"using pivot function to find cured , deaths , confirmed cases State wise"

statewise = pd.pivot_table(df_india,values=['Cured','Deaths','Confirmed'],index='State/UnionTerritory',aggfunc='max',margins=True)
statewise
"Top 10 states with Active cases"

df_top_10 = df_india.nlargest(10,['Active'])

df_top_10 = df_india.groupby(['State/UnionTerritory'])['Active'].max().sort_values(ascending=False).reset_index()
df_top = df_top_10.nlargest(10,['Active'])
df_top
"Top 10 states with deaths cases"

df_deaths_10 = df_india.nlargest(10,['Deaths'])

df_deaths_10 = df_india.groupby(['State/UnionTerritory'])['Deaths'].max().sort_values(ascending=False).reset_index()
df_deaths = df_deaths_10.nlargest(10,['Deaths'])

df_deaths
"Finding recovery rate and deathrate"

statewise['Recoveryrate'] = statewise['Cured']*100/statewise['Confirmed']
statewise['Deathrate'] = statewise['Deaths']*100/statewise['Confirmed']
statewise
"Correlation amongs the columns"

statewise.corr()
## Data visualization
"""Barplot for Confirmed , Deaths , Cured , Active"""
fig = plt.figure(figsize=(20,10))

confirm= df_india['Confirmed'].sum()
cured = df_india['Cured'].sum()
deaths= df_india['Deaths'].sum()
active= df_india['Active'].sum()

print('Total Confirmed cases =',confirm)
print('Total Cured cases =',cured)
print('Total Active cases =',active)
print('Total Death cases =',deaths)

barplot = sns.barplot(x=['Confirmed','Cured','Deaths','active'],y=[confirm,cured,deaths,active])
barplot.set_yticklabels(labels=(barplot.get_yticks()*1).astype(int))

plt.show()
"Piechart for 'Confirmed','Cured',Deaths & 'Active'"

fig = plt.figure(figsize=(17,10))
df_values = [df_india['Confirmed'].sum(),df_india['Cured'].sum(),df_india['Deaths'].sum(),df_india['Active'].sum()]
df_keys = [confirm,cured,deaths,active]


plt.pie(df_keys,labels = df_keys, explode = (0.02,0.02,0.1,0.02), autopct = '%.0f%%')
plt.legend(['Confirmed','Cured','Deaths','Active'])
"Pie Chart Of 10 Top states with Active Cases"

fig = plt.figure(figsize=(17,10))
df_top.groupby(["State/UnionTerritory"]).sum()["Active"].plot(kind='pie',rot=90,explode=(0.05,0.02,0.03,0.04,0.04,0.05,0.1,0.04,0.09,0.04),autopct='%1.0f%%')
plt.title('Top 10 states with Active cases',size=20)
plt.show()
  
"Bar Plot Of Top 10 Active Cases"

fig = plt.figure(figsize=(20,10))
sns.barplot(data = df_top.iloc[:10],y='Active',x='State/UnionTerritory')
plt.title('Top 10 states with Most Active cases', size=20)
plt.show()
## Note : As per above visual's it is clear that Maharashtra has maximum number of Active cases wheras Chhattisgarh has the least number of Active cases.
"""Pie chart of top 10 states with death cases"""

fig = plt.figure(figsize=(17,10))
df_deaths.groupby(["State/UnionTerritory"]).sum()["Deaths"].plot(kind='pie',rot=90,explode=(0.05,0.02,0.03,0.04,0.04,0.05,0.1,0.04,0.09,0.04),autopct='%1.0f%%')
plt.title("Pie chart of top 10 states with death cases",size = 20)
"Bar graph with top 10 states with most Death cases"


df_deaths = df_india.groupby('State/UnionTerritory').max()[['Deaths','Date']].sort_values(by='Deaths',ascending=False).reset_index()
fig = plt.figure(figsize=(20,10))
plot1 = sns.barplot(data = df_deaths.iloc[:10],y='Deaths',x='State/UnionTerritory')
plt.title('Top 10 states with most death cases', size=20)
plt.xlabel('states')
plt.ylabel('Deaths')
plt.show()
## Note : As per above visual's it is clear that Maharashtra has most number of death cases and Chhattisgarh has least number of death cases.
" Top 5 Most affected states"

fig = plt.figure(figsize=(15,10))
plot = sns.lineplot(data = df_india[df_india['State/UnionTerritory'].isin(['Maharashtra','Karnataka','Tamil Nadu','Delhi','Uttar Pradesh'])],x='Date',y='Active',hue = 'State/UnionTerritory',size='State/UnionTerritory')
plt.title('5 most affected states',size=20)
plt.show()
## correlation Heatmap
"Correlation Heatmap"

fig = plt.figure(figsize=(15,10))
sns.heatmap(df_india.corr(),cmap="Blues")
plt.title('Correlation Heatmap',size=20)
plt.show()
"Fatality ratio of contaminated states"

df_india['Fatality_ratio']= df_india['Deaths']/df_india['Confirmed']
a4_dims = (15,7)
fig,ax = plt.subplots(figsize=a4_dims)
sns.pointplot(data = df_india,x='State/UnionTerritory',y='Fatality_ratio',ax=ax,color='Green')
plt.xticks(rotation=90)
plt.title('Fatality ratio of contaminated states',size=20)
plt.show()