import numpy as np
import pandas as pd

crosstab, pivot_table

titanic = pd.read_csv('https://raw.githubusercontent.com/shekhar270779/Learn_ML/main/datasets/Titanic.csv')titanic.head()

Cross Tabulation / Contingency table

pd.crosstab(titanic.Survived, titanic.Pclass)

rows = titanic.Survived
cols = titanic.Pclass
pd.crosstab(rows, cols, margins=['rows','columns'])

pd.crosstab(titanic.Survived, titanic.Pclass, normalize=True, margins=True)

Pivoting

pd.pivot_table(index='Survived', columns='Pclass', values=['Fare'], aggfunc=lambda x : np.mean(x), data=titanic)

pd.pivot_table(index='Survived', columns='Pclass', values=['Age'], aggfunc=lambda x : np.mean(x), data=titanic)

titanic.pivot_table(index='Pclass', columns='Sex', values='Age', aggfunc=lambda x : np.mean(x)).unstack()Sex     Pclass
female  1         34.611765
        2         28.722973
        3         21.750000
male    1         41.281386
        2         30.740707
        3         26.507589
dtype: float64

Challenge on pivot_table, compare avg. age based on

• Survived vs Pclass
• Survived vs Sex
• Class vs Sex

titanic.pivot_table(index='Survived', columns='Pclass', values=['Age'], aggfunc=lambda x : np.mean(x))

titanic.groupby(['Survived', 'Pclass']).agg({'Age': np.mean})