Titanic

Tasks

1. Find out how many passenger died/survived.
2. How many female/male passenger died/survived?
3. How passenger died/survived for each passenger class?
4. For each passenger class, find out how many female/male passenger?
5. How many infants died/survived?
6. Oldest passenger? Female/Male? Which class?
7. Youngest passenger? Female/Male? Which class?
8. Oldest died passenger? Female/Male? Which class?
9. Oldest survived passenger? Female/Male? Which class?
10. Youngest died passenger? Female/Male? Which class?
11. Youngest survived passenger? Female/Male? Which class?
12. Find out how many passengers for each class?

import pandas as pd

# load data
data = pd.read_csv("data/Titanic.csv")

data.shape

(434, 4)

data.columns

Index(['Survived', 'Gender', 'Passenger Class', 'Age'], dtype='object')

# 1. Find out how many passenger died/survived.
data.groupby('Survived').count()

	Gender	Passenger Class	Age
Survived
died	227	227	222
survived	207	207	201

data.groupby('Survived').size()

Survived
died        227
survived    207
dtype: int64

# 2. How many female/male passenger died/survived?
data.groupby(['Survived', 'Gender']).count()

		Passenger Class	Age
Survived	Gender
died	female	58	56
	male	169	166
survived	female	122	119
	male	85	82

data.groupby(['Survived', 'Gender']).size()

Survived  Gender
died      female     58
          male      169
survived  female    122
          male       85
dtype: int64

# 3. How many passenger died/survived for each passenger class?
data.groupby(['Survived', 'Passenger Class']).size()

Survived  Passenger Class
died      1st                 52
          2nd                 61
          3rd                114
survived  1st                 84
          2nd                 62
          3rd                 61
dtype: int64

# 4. For each passenger class, find out how many female/male passenger?
data.groupby(['Passenger Class', 'Gender']).size()

Passenger Class  Gender
1st              female     53
                 male       83
2nd              female     55
                 male       68
3rd              female     72
                 male      103
dtype: int64

import numpy as np

data.shape

(423, 4)

# drop rows with empty Age
data['Age'].replace('', np.nan, inplace=True)
data.dropna(subset=['Age'], inplace=True)

# check the datatype
data.dtypes

Survived           object
Gender             object
Passenger Class    object
Age                 int32
dtype: object

# convert Age column to int
data['Age'] = data['Age'].astype(int)

# 5. How many infants died/survived?
infants = data[data['Age'] <= 2]
infants.groupby('Survived').size()

Survived
died         7
survived    15
dtype: int64

infants.groupby('Survived').count()

	Gender	Passenger Class	Age
Survived
died	7	7	7
survived	15	15	15

infants.groupby(['Survived', 'Passenger Class']).size()

Survived  Passenger Class
died      1st                1
          3rd                6
survived  1st                1
          2nd                7
          3rd                7
dtype: int64

infants.groupby(['Survived', 'Gender']).size()

Survived  Gender
died      female    3
          male      4
survived  female    7
          male      8
dtype: int64

infants.groupby(['Survived', 'Passenger Class', 'Gender']).size()

Survived  Passenger Class  Gender
died      1st              female    1
          3rd              female    2
                           male      4
survived  1st              male      1
          2nd              female    3
                           male      4
          3rd              female    4
                           male      3
dtype: int64

# 6. Oldest passenger? Female/Male? Which class?
data[data['Age'] == data['Age'].max()]

	Survived	Gender	Passenger Class	Age
384	survived	male	1st	80

# 7. Youngest passenger? Female/Male? Which class?
data[data['Age'] == data['Age'].min()]

	Survived	Gender	Passenger Class	Age
155	died	male	3rd	0
156	died	male	3rd	0
276	survived	female	2nd	0
319	survived	female	3rd	0
320	survived	female	3rd	0
350	survived	male	1st	0
386	survived	male	2nd	0
387	survived	male	2nd	0
405	survived	male	3rd	0
406	survived	male	3rd	0

# summarize
data[data['Age'] == data['Age'].min()].groupby(['Survived', 'Passenger Class', 'Gender']).size()

Survived  Passenger Class  Gender
died      3rd              male      2
survived  1st              male      1
          2nd              female    1
                           male      2
          3rd              female    2
                           male      2
dtype: int64

# 8. Oldest died passenger? Female/Male? Which class?
died_passengers = data[data['Survived'] == 'died']

died_passengers.shape

(222, 4)

died_passengers[died_passengers['Age'] == died_passengers['Age'].max()]

	Survived	Gender	Passenger Class	Age
226	died	male	3rd	74

# 9. Oldest survived passenger? Female/Male? Which class?
survived = data[data['Survived'] != 'died']
survived[survived['Age'] == survived['Age'].max()]

	Survived	Gender	Passenger Class	Age
384	survived	male	1st	80

# 10. Youngest died passenger? Female/Male? Which class?
died_passengers[died_passengers['Age'] == died_passengers['Age'].min()]

	Survived	Gender	Passenger Class	Age
155	died	male	3rd	0
156	died	male	3rd	0

# 11. Youngest survived passenger? Female/Male? Which class?
survived[survived['Age'] == survived['Age'].min()]

	Survived	Gender	Passenger Class	Age
276	survived	female	2nd	0
319	survived	female	3rd	0
320	survived	female	3rd	0
350	survived	male	1st	0
386	survived	male	2nd	0
387	survived	male	2nd	0
405	survived	male	3rd	0
406	survived	male	3rd	0

# 12. Find out how many passengers for each class?
data.groupby(['Passenger Class']).size()

Passenger Class
1st    133
2nd    119
3rd    171
dtype: int64