import pandas as pd
import numpy as np

%pylab inline

figsize(7, 4)

df = pd.read_csv('titanic.csv')

df.head(5)

df['Survived']

len(df)

df['Sex'].value_counts()

df['Survived'].head(10)

df['Sex'].value_counts().plot(kind='bar')

df_survived = df[ df['Survived'] == 1 ]

df_survived.head(5)

df0 = df[ df['Age'] < 10 ]

len([1, 4, 5, 6, 6])

(df_survived['Pclass'].value_counts() / df['Pclass'].value_counts().astype(float)).plot(kind='bar')

(df_survived['Sex'].value_counts() / df['Sex'].value_counts().astype(float)).plot(kind='bar')

df['Age'].value_counts().plot(kind='bar')

df['Age'].hist(bins=30)

df['Age'].dropna().plot(kind='kde')
df_survived['Age'].dropna().plot(kind='kde')

df[ df['Sex'] == 'male']['Age'].dropna().plot(kind='kde')
df_survived[df_survived['Sex'] == 'male']['Age'].dropna().plot(kind='kde')