import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

sns.set(style="darkgrid")

data = {
    'Country': ['USA', 'India', 'Brazil', 'Russia', 'UK', 'France', 'Spain', 'Italy'],
    'TotalCases': [34000000, 31000000, 19000000, 7000000, 5000000, 5800000, 3700000, 4200000],
    'TotalDeaths': [610000, 410000, 530000, 190000, 130000, 110000, 81000, 127000],
    'TotalRecovered': [28000000, 30000000, 17000000, 6500000, 4800000, 5000000, 3500000, 3900000],
    'Population': [331000000, 1380000000, 213000000, 146000000, 68000000, 67000000, 47000000, 60000000]
}
df = pd.DataFrame(data)
df

df.isnull().sum()

Country           0
TotalCases        0
TotalDeaths       0
TotalRecovered    0
Population        0
dtype: int64

df['ActiveCases'] = df['TotalCases'] - df['TotalRecovered'] - df['TotalDeaths']
df

df['MortalityRate'] = (df['TotalDeaths'] / df['TotalCases']) * 100
df['MortalityRate'] = df['MortalityRate'].round(2)
df

df['RecoveryRate'] = (df['TotalRecovered'] / df['TotalCases']) * 100
df['RecoveryRate'] = df['RecoveryRate'].round(2)
df

df['CasesPer100k'] = (df['TotalCases'] / df['Population']) * 100000
df['CasesPer100k'] = df['CasesPer100k'].round(2)
df

df.sort_values('TotalCases', ascending=False)

plt.figure(figsize=(10,5))
sns.barplot(x="Country", y="TotalCases", data=df)
plt.title("Total COVID-19 Cases by Country")
plt.xticks(rotation=45)
plt.show()

plt.figure(figsize=(6,6))
plt.pie(df['TotalDeaths'], labels=df['Country'], autopct='%1.1f%%', startangle=140)
plt.title("Share of Total Deaths")
plt.show()

plt.figure(figsize=(8,6))
sns.scatterplot(x="RecoveryRate", y="MortalityRate", data=df, hue="Country", s=100)
plt.title("Recovery vs Mortality Rate")
plt.show()

correlation = df.corr(numeric_only=True)
correlation

plt.figure(figsize=(10,6))
sns.heatmap(correlation, annot=True, cmap="Blues")
plt.title("Feature Correlation Matrix")
plt.show()

df.sort_values('MortalityRate', ascending=False).head(5)

totals = df[['TotalCases', 'TotalDeaths', 'TotalRecovered']].sum()
totals

TotalCases        109700000
TotalDeaths         2188000
TotalRecovered     98700000
dtype: int64

global_mortality = (totals['TotalDeaths'] / totals['TotalCases']) * 100
round(global_mortality, 2)

np.float64(1.99)

df['RiskLevel'] = df['MortalityRate'].apply(lambda x: 'High' if x > 2 else 'Moderate')
df

sns.countplot(x="RiskLevel", data=df)
plt.title("Country Risk Level Count")
plt.show()