{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"---\n",
"title : \"🚢 Analyse statistique et visualisation : le jeu de données du Titanic\"\n",
"---"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Nous allons effectuer une analyse exploratoire approfondie (EDA) de l'ensemble de données relatives au naufrage du Titanic. Notre objectif est d'étudier les données en détail et de voir comment même des observations simples peuvent révéler des informations précieuses.\n",
"\n",
"Le naufrage du Titanic est l'un des plus célèbres de l'histoire. Le **15 avril 1912**, lors de son voyage inaugural, le RMS Titanic, un paquebot transatlantique britannique, a sombré après être entré en collision avec un iceberg, **tuant 1502 passagers et membres d'équipage sur 2224**. (Selon [Wikipedia](https://fr.wikipedia.org/wiki/Titanic) il y avait 1316 passagers et 889 membres d'équipage, ce qui donne 2205 personnes) L'une des raisons pour lesquelles le naufrage a causé tant de morts est qu'il n'y avait pas assez de canots de sauvetage pour les passagers et l'équipage. On a toujours entendu dire que certains groupes de personnes étaient plus susceptibles de survivre que d'autres, comme les femmes, les enfants et les classes supérieures. Nous allons analyser les données pour vérifier ces affirmations.\n",
"\n",
"Dans cet exemple, nous allons récupérer un jeu de données (liste de personnes, caractéristiques, survivants ou non…) et analyser quelles catégories de personnes ont survécu. Nous utiliserons les bibliothèques **pandas**, **Matplotlib**, **Seaborn** et **Plotly** pour explorer et visualiser ces données.\n",
"\n",
""
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Importation des données"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:31.408063Z",
"iopub.status.busy": "2026-10-30T13:10:31.407293Z",
"iopub.status.idle": "2026-10-30T13:10:32.489209Z",
"shell.execute_reply": "2026-10-30T13:10:32.488714Z"
}
},
"outputs": [],
"source": [
"# Import data analysis libraries\n",
"# ==============================\n",
"import numpy as np\n",
"import pandas as pd \n",
"pd.set_option(\"display.max_columns\", None)\n",
"\n",
"\n",
"# Import visualization libraries and set favourite style\n",
"# ======================================================\n",
"import matplotlib.pyplot as plt\n",
"plt.rcParams['font.size'] = 12\n",
"plt.rcParams['figure.figsize'] = (10, 6) # set default size of plots\n",
"# plt.style.use('ggplot') # set ggplot style\n",
"\n",
"import seaborn as sns\n",
"sns.set_theme(style = \"ticks\", # \"ticks\", \"whitegrid\", \"darkgrid\", \"white\", \"dark\"\n",
" palette = \"pastel\", # None, Set2, husl, deep, muted, bright, pastel, dark, colorblind\n",
" rc = {\"axes.spines.right\": False, \"axes.spines.top\": False})\n",
"\n",
"\n",
"import plotly.express as px\n",
"from plotly.subplots import make_subplots\n",
"\n",
"# Import Notebook display library\n",
"# ==============================\n",
"from IPython.display import display, Markdown\n",
"\n",
"\n",
"# Pour la génération automatique de graphiques interactifs dans le polycopié\n",
"# ==============================\n",
"import plotly.io as pio\n",
"import os\n",
"if os.getenv(\"QUARTO_OUTPUT_FORMAT\") is not None:\n",
" # Export Quarto HTML\n",
" pio.renderers.default = \"iframe_connected\"\n",
"else:\n",
" # Notebook interactif\n",
" pio.renderers.default = \"notebook_connected\""
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"L'une des forces de pandas est l'importation et l'exportation des données. Ce package possède un ensemble de fonctions très large pour charger des données en mémoire et les exporter dans divers formats.\n",
"\n",
"Pandas dédie un sous-répertoire entier du package à l'importation et à l'exportation vers des formats de données exploitables avec d'autres outils. On peut citer les formats csv, txt, Excel®, SAS®, SQL, HDF5 entre autres. Suivant le format, les outils seront différents, mais les principes restent les mêmes. Ainsi, nous allons considérer uniquement le format `csv`.\n",
"\n",
"Un fichier **CSV** (.csv) est un fichier de données tabulaires. Le sigle CSV signifie *Comma Separated Values* qui se traduit par « valeurs séparées par des virgules ». L'avantage de ce type de fichier est qu'il s'agit d'un fichier texte qui ne conserve que les données du tableau (pas de mise en page) et peut être lu par n'importe quel tableur. Les données d'une même ligne sont souvent séparées par des points-virgules ou des virgules."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Pour importer les données stockées dans un fichier csv dont les éléments sont séparés par des virgules, on utilisera :\n",
"```python\n",
"df = pd.read_csv('nom_du_fichier.csv', sep=',')\n",
"```"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Dans l'exemple suivant nous allons importer et afficher un fichier csv contenant des données sur le Titanic."
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.491406Z",
"iopub.status.busy": "2026-10-30T13:10:32.491147Z",
"iopub.status.idle": "2026-10-30T13:10:32.497275Z",
"shell.execute_reply": "2026-10-30T13:10:32.496821Z"
}
},
"outputs": [],
"source": [
"df = pd.read_csv('titanic.csv')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Étude globale de l'ensemble de données"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"
\n",
"\n",
"Dans cette partie nous allons utiliser les méthodes suivantes :\n",
"```python\n",
".shape\n",
".info()\n",
".head() .tail()\n",
".columns\n",
".isnull() .dropna() .fillna()\n",
".unique() .nunique() .value_counts()\n",
".describe() .min(), .max(), .mean(), .median(), .std()\n",
"```\n",
"\n",
"
"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"La première étape de tout projet d'analyse de données consiste à examiner les données. Nous devons voir combien d'observations (lignes) et combien d'entités (colonnes) sont contenues, ce que signifient ces colonnes, et ainsi de suite. Cela nous aidera à nous familiariser avec l'ensemble de données, et pourrait même nous aider à évaluer quelles informations sont importantes et lesquelles ne le sont pas."
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.499225Z",
"iopub.status.busy": "2026-10-30T13:10:32.499050Z",
"iopub.status.idle": "2026-10-30T13:10:32.505800Z",
"shell.execute_reply": "2026-10-30T13:10:32.505401Z"
}
},
"outputs": [
{
"data": {
"text/plain": [
"(1309, 7)"
]
},
"execution_count": 3,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.shape # tailles du dataset : 1309 lignes et 7 colonnes\n",
"# df.columns"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Un moyen rapide de vérifier le contenu consiste à appeler les 5 premières lignes à l'aide de la méthode `.head()` sans spécifier l'argument entre parenthèses. Si nous voulons vérifier les 20 premières lignes, nous mettons 20 comme argument."
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.541499Z",
"iopub.status.busy": "2026-10-30T13:10:32.541321Z",
"iopub.status.idle": "2026-10-30T13:10:32.550965Z",
"shell.execute_reply": "2026-10-30T13:10:32.550566Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
classe
\n",
"
survivant
\n",
"
nom
\n",
"
sexe
\n",
"
age
\n",
"
prix
\n",
"
port_depart
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
1
\n",
"
1
\n",
"
Allen, Miss. Elisabeth Walton
\n",
"
F
\n",
"
29.0000
\n",
"
211.3375
\n",
"
S
\n",
"
\n",
"
\n",
"
1
\n",
"
1
\n",
"
1
\n",
"
Allison, Master. Hudson Trevor
\n",
"
H
\n",
"
0.9167
\n",
"
151.5500
\n",
"
S
\n",
"
\n",
"
\n",
"
2
\n",
"
1
\n",
"
0
\n",
"
Allison, Miss. Helen Loraine
\n",
"
F
\n",
"
2.0000
\n",
"
151.5500
\n",
"
S
\n",
"
\n",
"
\n",
"
3
\n",
"
1
\n",
"
0
\n",
"
Allison, Mr. Hudson Joshua Creighton
\n",
"
H
\n",
"
30.0000
\n",
"
151.5500
\n",
"
S
\n",
"
\n",
"
\n",
"
4
\n",
"
1
\n",
"
0
\n",
"
Allison, Mrs. Hudson J C (Bessie Waldo Daniels)
\n",
"
F
\n",
"
25.0000
\n",
"
151.5500
\n",
"
S
\n",
"
\n",
"
\n",
"
5
\n",
"
1
\n",
"
1
\n",
"
Anderson, Mr. Harry
\n",
"
H
\n",
"
48.0000
\n",
"
26.5500
\n",
"
S
\n",
"
\n",
"
\n",
"
6
\n",
"
1
\n",
"
1
\n",
"
Andrews, Miss. Kornelia Theodosia
\n",
"
F
\n",
"
63.0000
\n",
"
77.9583
\n",
"
S
\n",
"
\n",
"
\n",
"
7
\n",
"
1
\n",
"
0
\n",
"
Andrews, Mr. Thomas Jr
\n",
"
H
\n",
"
39.0000
\n",
"
0.0000
\n",
"
S
\n",
"
\n",
"
\n",
"
8
\n",
"
1
\n",
"
1
\n",
"
Appleton, Mrs. Edward Dale (Charlotte Lamson)
\n",
"
F
\n",
"
53.0000
\n",
"
51.4792
\n",
"
S
\n",
"
\n",
"
\n",
"
9
\n",
"
1
\n",
"
0
\n",
"
Artagaveytia, Mr. Ramon
\n",
"
H
\n",
"
71.0000
\n",
"
49.5042
\n",
"
C
\n",
"
\n",
"
\n",
"
10
\n",
"
1
\n",
"
0
\n",
"
Astor, Col. John Jacob
\n",
"
H
\n",
"
47.0000
\n",
"
227.5250
\n",
"
C
\n",
"
\n",
"
\n",
"
11
\n",
"
1
\n",
"
1
\n",
"
Astor, Mrs. John Jacob (Madeleine Talmadge Force)
\n",
"
F
\n",
"
18.0000
\n",
"
227.5250
\n",
"
C
\n",
"
\n",
"
\n",
"
12
\n",
"
1
\n",
"
1
\n",
"
Aubart, Mme. Leontine Pauline
\n",
"
F
\n",
"
24.0000
\n",
"
69.3000
\n",
"
C
\n",
"
\n",
"
\n",
"
13
\n",
"
1
\n",
"
1
\n",
"
Barber, Miss. Ellen \"Nellie\"
\n",
"
F
\n",
"
26.0000
\n",
"
78.8500
\n",
"
S
\n",
"
\n",
"
\n",
"
14
\n",
"
1
\n",
"
1
\n",
"
Barkworth, Mr. Algernon Henry Wilson
\n",
"
H
\n",
"
80.0000
\n",
"
30.0000
\n",
"
S
\n",
"
\n",
"
\n",
"
15
\n",
"
1
\n",
"
0
\n",
"
Baumann, Mr. John D
\n",
"
H
\n",
"
NaN
\n",
"
25.9250
\n",
"
S
\n",
"
\n",
"
\n",
"
16
\n",
"
1
\n",
"
0
\n",
"
Baxter, Mr. Quigg Edmond
\n",
"
H
\n",
"
24.0000
\n",
"
247.5208
\n",
"
C
\n",
"
\n",
"
\n",
"
17
\n",
"
1
\n",
"
1
\n",
"
Baxter, Mrs. James (Helene DeLaudeniere Chaput)
\n",
"
F
\n",
"
50.0000
\n",
"
247.5208
\n",
"
C
\n",
"
\n",
"
\n",
"
18
\n",
"
1
\n",
"
1
\n",
"
Bazzani, Miss. Albina
\n",
"
F
\n",
"
32.0000
\n",
"
76.2917
\n",
"
C
\n",
"
\n",
"
\n",
"
19
\n",
"
1
\n",
"
0
\n",
"
Beattie, Mr. Thomson
\n",
"
H
\n",
"
36.0000
\n",
"
75.2417
\n",
"
C
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" classe survivant nom sexe \\\n",
"0 1 1 Allen, Miss. Elisabeth Walton F \n",
"1 1 1 Allison, Master. Hudson Trevor H \n",
"2 1 0 Allison, Miss. Helen Loraine F \n",
"3 1 0 Allison, Mr. Hudson Joshua Creighton H \n",
"4 1 0 Allison, Mrs. Hudson J C (Bessie Waldo Daniels) F \n",
"5 1 1 Anderson, Mr. Harry H \n",
"6 1 1 Andrews, Miss. Kornelia Theodosia F \n",
"7 1 0 Andrews, Mr. Thomas Jr H \n",
"8 1 1 Appleton, Mrs. Edward Dale (Charlotte Lamson) F \n",
"9 1 0 Artagaveytia, Mr. Ramon H \n",
"10 1 0 Astor, Col. John Jacob H \n",
"11 1 1 Astor, Mrs. John Jacob (Madeleine Talmadge Force) F \n",
"12 1 1 Aubart, Mme. Leontine Pauline F \n",
"13 1 1 Barber, Miss. Ellen \"Nellie\" F \n",
"14 1 1 Barkworth, Mr. Algernon Henry Wilson H \n",
"15 1 0 Baumann, Mr. John D H \n",
"16 1 0 Baxter, Mr. Quigg Edmond H \n",
"17 1 1 Baxter, Mrs. James (Helene DeLaudeniere Chaput) F \n",
"18 1 1 Bazzani, Miss. Albina F \n",
"19 1 0 Beattie, Mr. Thomson H \n",
"\n",
" age prix port_depart \n",
"0 29.0000 211.3375 S \n",
"1 0.9167 151.5500 S \n",
"2 2.0000 151.5500 S \n",
"3 30.0000 151.5500 S \n",
"4 25.0000 151.5500 S \n",
"5 48.0000 26.5500 S \n",
"6 63.0000 77.9583 S \n",
"7 39.0000 0.0000 S \n",
"8 53.0000 51.4792 S \n",
"9 71.0000 49.5042 C \n",
"10 47.0000 227.5250 C \n",
"11 18.0000 227.5250 C \n",
"12 24.0000 69.3000 C \n",
"13 26.0000 78.8500 S \n",
"14 80.0000 30.0000 S \n",
"15 NaN 25.9250 S \n",
"16 24.0000 247.5208 C \n",
"17 50.0000 247.5208 C \n",
"18 32.0000 76.2917 C \n",
"19 36.0000 75.2417 C "
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# df\n",
"df.head(20)\n",
"# df.tail(10)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On compte 7 caractéristiques (colonnes) décrivant chaque personne à bord du Titanic. Notre caractéristique cible (également appelée variable dépendante) est la colonne `survivant`, qui vaut 1 si la personne a survécu et 0 sinon. Dans ce cas, il est facile de déduire que cette colonne ne contient que des valeurs numériques 0 et 1. Cependant, pour d'autres caractéristiques (ou variables indépendantes), il peut ne pas être possible de déduire au premier coup d'œil le type de données contenues.\n",
"\n",
"Afin de générer rapidement un tableau des types de données contenus dans chaque colonne, nous utilisons la méthode `.info()`. "
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.552732Z",
"iopub.status.busy": "2026-10-30T13:10:32.552579Z",
"iopub.status.idle": "2026-10-30T13:10:32.559922Z",
"shell.execute_reply": "2026-10-30T13:10:32.559518Z"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n",
"RangeIndex: 1309 entries, 0 to 1308\n",
"Data columns (total 7 columns):\n",
" # Column Non-Null Count Dtype \n",
"--- ------ -------------- ----- \n",
" 0 classe 1309 non-null int64 \n",
" 1 survivant 1309 non-null int64 \n",
" 2 nom 1309 non-null object \n",
" 3 sexe 1309 non-null object \n",
" 4 age 1046 non-null float64\n",
" 5 prix 1308 non-null float64\n",
" 6 port_depart 1307 non-null object \n",
"dtypes: float64(2), int64(2), object(3)\n",
"memory usage: 71.7+ KB\n"
]
}
],
"source": [
"# Info about data frame dimensions, column types, and file size\n",
"\n",
"df.info()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Données manquantes ?"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Combien de valeurs manquantes dans chaque colonne ?"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.561868Z",
"iopub.status.busy": "2026-10-30T13:10:32.561677Z",
"iopub.status.idle": "2026-10-30T13:10:32.567478Z",
"shell.execute_reply": "2026-10-30T13:10:32.566979Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Données manquantes
\n",
"
\n",
" \n",
" \n",
"
\n",
"
age
\n",
"
263
\n",
"
\n",
"
\n",
"
port_depart
\n",
"
2
\n",
"
\n",
"
\n",
"
prix
\n",
"
1
\n",
"
\n",
"
\n",
"
classe
\n",
"
0
\n",
"
\n",
"
\n",
"
survivant
\n",
"
0
\n",
"
\n",
"
\n",
"
nom
\n",
"
0
\n",
"
\n",
"
\n",
"
sexe
\n",
"
0
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Données manquantes\n",
"age 263\n",
"port_depart 2\n",
"prix 1\n",
"classe 0\n",
"survivant 0\n",
"nom 0\n",
"sexe 0"
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.isnull().sum().sort_values(ascending=False).to_frame('Données manquantes')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Il est utile d'utiliser la méthode `.isnull()` combinée avec `.sum()` ou `.mean()` pour connaître l'ampleur des valeurs manquantes dans un jeu de données."
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.569274Z",
"iopub.status.busy": "2026-10-30T13:10:32.569081Z",
"iopub.status.idle": "2026-10-30T13:10:32.575631Z",
"shell.execute_reply": "2026-10-30T13:10:32.575191Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
% données manquantes
\n",
"
\n",
" \n",
" \n",
"
\n",
"
age
\n",
"
20.09
\n",
"
\n",
"
\n",
"
port_depart
\n",
"
0.15
\n",
"
\n",
"
\n",
"
prix
\n",
"
0.08
\n",
"
\n",
"
\n",
"
classe
\n",
"
0.00
\n",
"
\n",
"
\n",
"
survivant
\n",
"
0.00
\n",
"
\n",
"
\n",
"
nom
\n",
"
0.00
\n",
"
\n",
"
\n",
"
sexe
\n",
"
0.00
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" % données manquantes\n",
"age 20.09\n",
"port_depart 0.15\n",
"prix 0.08\n",
"classe 0.00\n",
"survivant 0.00\n",
"nom 0.00\n",
"sexe 0.00"
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"(df.isnull().mean().sort_values(ascending = False) * 100).round(2).sort_values(ascending=False).to_frame('% données manquantes')\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Nous pouvons voir que la colonne `port_depart` contient 2 valeurs manquantes (indiquées par `NaN`). Qui sont les passagers pour lesquels on n'a pas ces informations ?"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.577356Z",
"iopub.status.busy": "2026-10-30T13:10:32.577200Z",
"iopub.status.idle": "2026-10-30T13:10:32.584342Z",
"shell.execute_reply": "2026-10-30T13:10:32.583925Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
classe
\n",
"
survivant
\n",
"
nom
\n",
"
sexe
\n",
"
age
\n",
"
prix
\n",
"
port_depart
\n",
"
\n",
" \n",
" \n",
"
\n",
"
168
\n",
"
1
\n",
"
1
\n",
"
Icard, Miss. Amelie
\n",
"
F
\n",
"
38.0
\n",
"
80.0
\n",
"
NaN
\n",
"
\n",
"
\n",
"
284
\n",
"
1
\n",
"
1
\n",
"
Stone, Mrs. George Nelson (Martha Evelyn)
\n",
"
F
\n",
"
62.0
\n",
"
80.0
\n",
"
NaN
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" classe survivant nom sexe age \\\n",
"168 1 1 Icard, Miss. Amelie F 38.0 \n",
"284 1 1 Stone, Mrs. George Nelson (Martha Evelyn) F 62.0 \n",
"\n",
" prix port_depart \n",
"168 80.0 NaN \n",
"284 80.0 NaN "
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# mask = df['age'].isnull()\n",
"# df.loc[mask] # il y a 263 valeurs manquantes pour l'age, car 1309-1046=263\n",
"\n",
"mask = df['port_depart'].isnull()\n",
"df.loc[mask]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"De même, nous pouvons voir que dans la colonne `age` il manque 263 données (nous n'allons pas les afficher). Peut-être que ces données n'ont pas été collectées ou ont été perdues. Que faire ? Nous pourrions soit supprimer ces lignes, soit remplacer les valeurs manquantes par la moyenne de l'âge des passagers par exemple. Les choix relatifs au traitement des valeurs manquantes ne sont pas des choix méthodologiques neutres. Pandas donne les outils techniques pour faire ceci mais la question de la légitimité de ces choix et de leur pertinence est propre à chaque jeu de données.\n",
"\n",
"Pour le moment nous n'allons pas modifier les données."
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.586121Z",
"iopub.status.busy": "2026-10-30T13:10:32.585920Z",
"iopub.status.idle": "2026-10-30T13:10:32.588254Z",
"shell.execute_reply": "2026-10-30T13:10:32.587831Z"
}
},
"outputs": [],
"source": [
"# Option 1 : éliminer les lignes avec des valeurs manquantes\n",
"# df = df.dropna(axis=0)\n",
"# df.shape # (1046, 7)\n",
"\n",
"# Option 2 : remplacer par la moyenne (Attention : on modifie la réalité)\n",
"# df.fillna(df['age'].mean(), inplace=True)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Bilan sur l'ensemble de données"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Conclusion** : nous savons maintenant que ce jeu de données comporte 1309 lignes et 7 colonnes. 4 colonnes sont de type numérique (flottants et entiers), 3 colonnes sont non numériques :\n",
"\n",
"- `classe` : valeurs possibles 1, 2 ou 3\n",
"- `survivant` : 0 (décédé), 1 (a survécu)\n",
"- `nom` : nom, prénom et titre\n",
"- `sexe` : H (homme), F (femme)\n",
"- `age` : en années\n",
"- `prix` : prix du ticket\n",
"- `port_depart` : trois ports d'embarquement, C = Cherbourg (France), S = Southampton (Angleterre), Q = Queenstown (Irlande)\n",
"\n",
"\n",
"\n",
"\n",
""
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Statistiques descriptives"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Pour les colonnes numériques telles que `age`, `prix`, etc., nous aimerions connaître leurs valeurs moyennes, maximales et minimales pour voir si les données sont raisonnablement distribuées ou s'il y a des anomalies. Pour afficher un tableau de statistiques pour chaque colonne numérique, nous utilisons la méthode `.describe()`."
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.589813Z",
"iopub.status.busy": "2026-10-30T13:10:32.589666Z",
"iopub.status.idle": "2026-10-30T13:10:32.609060Z",
"shell.execute_reply": "2026-10-30T13:10:32.608633Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
"
],
"text/plain": [
" count unique top freq\n",
"nom 1309 1307 Connolly, Miss. Kate 2\n",
"sexe 1309 2 H 843\n",
"port_depart 1307 3 S 914"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# df.describe(include='all')\n",
"display(df.describe().T)\n",
"display(df.describe(include='object').T)\n",
"# df.describe(include=\"all\")\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On note que la colonne `prix` a une valeur minimale de 0, ce qui semble étrange. Cela pourrait signifier que certaines personnes ont voyagé gratuitement."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Analyse univariée (chaque colonne individuellement)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Commençons par l'analyse univariée, qui consiste à explorer une seule variable à la fois. Cela nous aide à comprendre la distribution, les tendances et les caractéristiques sous-jacentes de cette variable. En traçant et en analysant les courbes d'une caractéristique, nous pouvons obtenir des informations précieuses sur les données avant de passer à des relations plus complexes."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"
\n",
"\n",
"Dans cette partie nous allons utiliser les méthodes suivantes :\n",
"\n",
"```python\n",
".drop() .replace()\n",
".value_counts()\n",
".agg()\n",
".cut() .map()\n",
".plot(kind='bar') .plot(kind='hist')\n",
"sns.countplot() sns.catplot() sns.histplot()\n",
"px.histogram() px.pie()\n",
"```\n",
"\n",
"
"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Tout d'abord nous allons examiner la colonne `nom` pour retrouver des informations intéressantes sur les passagers (en vérifiant les informations indiquées sur le site https://titanicfacts.net/titanic-passengers/). Ensuite, nous analyserons les autres colonnes une par une. Nous nous intéresserons particulièrement aux colonnes `classe`, `sexe`, `survivant`, `age` et verrons ensuite comment ces caractéristiques influencent les chances de survie des passagers. "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Colonne `nom`"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Plus jeune et plus vieux passagers"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Une autre valeur qui attire l'attention est l'âge minimum, soit 0,1667 ans, avec un éventuel bébé de 9 semaines à bord. De plus amples informations sur https://titanicfacts.net/titanic-passengers/ indiquent qu'il s'agit de Elizabeth Gladys “Milvina” Dean, une fille âgée de 2 mois et 13 jours au moment du naufrage (qui est décédée en 2009 à l'âge de 97 ans). Vérifions si cette information est correcte :"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.610893Z",
"iopub.status.busy": "2026-10-30T13:10:32.610736Z",
"iopub.status.idle": "2026-10-30T13:10:32.613094Z",
"shell.execute_reply": "2026-10-30T13:10:32.612683Z"
}
},
"outputs": [],
"source": [
"# mask_idx = df['age'].idxmin() # index de la personne la plus jeune\n",
"# df.iloc[mask_idx]"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.614782Z",
"iopub.status.busy": "2026-10-30T13:10:32.614633Z",
"iopub.status.idle": "2026-10-30T13:10:32.621436Z",
"shell.execute_reply": "2026-10-30T13:10:32.621024Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
classe
\n",
"
survivant
\n",
"
nom
\n",
"
sexe
\n",
"
age
\n",
"
prix
\n",
"
port_depart
\n",
"
\n",
" \n",
" \n",
"
\n",
"
763
\n",
"
3
\n",
"
1
\n",
"
Dean, Miss. Elizabeth Gladys \"Millvina\"
\n",
"
F
\n",
"
0.1667
\n",
"
20.575
\n",
"
S
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" classe survivant nom sexe age \\\n",
"763 3 1 Dean, Miss. Elizabeth Gladys \"Millvina\" F 0.1667 \n",
"\n",
" prix port_depart \n",
"763 20.575 S "
]
},
"execution_count": 12,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"mask = df['age'] == df['age'].min()\n",
"df.loc[mask]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Et le plus jeune passager garçon ? Toujours selon le site précédent, il s'agit de Master Gilbert Sigvard Emanuel Danbom, âgé de 4 mois et 29 jours, qui n'a malheureusement pas survécu. Vérifions si cette information est correcte :"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.623101Z",
"iopub.status.busy": "2026-10-30T13:10:32.622908Z",
"iopub.status.idle": "2026-10-30T13:10:32.630724Z",
"shell.execute_reply": "2026-10-30T13:10:32.630334Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
classe
\n",
"
survivant
\n",
"
nom
\n",
"
sexe
\n",
"
age
\n",
"
prix
\n",
"
port_depart
\n",
"
\n",
" \n",
" \n",
"
\n",
"
747
\n",
"
3
\n",
"
0
\n",
"
Danbom, Master. Gilbert Sigvard Emanuel
\n",
"
H
\n",
"
0.3333
\n",
"
14.4
\n",
"
S
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" classe survivant nom sexe age \\\n",
"747 3 0 Danbom, Master. Gilbert Sigvard Emanuel H 0.3333 \n",
"\n",
" prix port_depart \n",
"747 14.4 S "
]
},
"execution_count": 13,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Filtrer les individus dont sex == \"H\"\n",
"filtered_df = df.loc[df['sexe'] == \"H\"]\n",
"# Trouver l'âge minimum dans cette sous-table\n",
"min_age = filtered_df['age'].min()\n",
"# Sélectionner les lignes correspondant à cet âge minimum\n",
"youngest = filtered_df.loc[filtered_df['age'] == min_age]\n",
"# Afficher les résultats\n",
"youngest"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Et le passager le plus âgé ?"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.632386Z",
"iopub.status.busy": "2026-10-30T13:10:32.632232Z",
"iopub.status.idle": "2026-10-30T13:10:32.639152Z",
"shell.execute_reply": "2026-10-30T13:10:32.638731Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
classe
\n",
"
survivant
\n",
"
nom
\n",
"
sexe
\n",
"
age
\n",
"
prix
\n",
"
port_depart
\n",
"
\n",
" \n",
" \n",
"
\n",
"
14
\n",
"
1
\n",
"
1
\n",
"
Barkworth, Mr. Algernon Henry Wilson
\n",
"
H
\n",
"
80.0
\n",
"
30.0
\n",
"
S
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" classe survivant nom sexe age prix \\\n",
"14 1 1 Barkworth, Mr. Algernon Henry Wilson H 80.0 30.0 \n",
"\n",
" port_depart \n",
"14 S "
]
},
"execution_count": 14,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.loc[df['age']==df['age'].max()]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
" Selon le site précédent, il s'agit de Mr Johan Svensson, âgé de 74 ans et 10 mois, qui n'a pas survécu. Cependant, notre jeu de données ne contient pas cette information :"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.640799Z",
"iopub.status.busy": "2026-10-30T13:10:32.640647Z",
"iopub.status.idle": "2026-10-30T13:10:32.648982Z",
"shell.execute_reply": "2026-10-30T13:10:32.648558Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
classe
\n",
"
survivant
\n",
"
nom
\n",
"
sexe
\n",
"
age
\n",
"
prix
\n",
"
port_depart
\n",
"
\n",
" \n",
" \n",
"
\n",
"
979
\n",
"
3
\n",
"
0
\n",
"
Lundahl, Mr. Johan Svensson
\n",
"
H
\n",
"
51.0
\n",
"
7.0542
\n",
"
S
\n",
"
\n",
"
\n",
"
1235
\n",
"
3
\n",
"
0
\n",
"
Svensson, Mr. Johan
\n",
"
H
\n",
"
74.0
\n",
"
7.7750
\n",
"
S
\n",
"
\n",
"
\n",
"
1236
\n",
"
3
\n",
"
1
\n",
"
Svensson, Mr. Johan Cervin
\n",
"
H
\n",
"
14.0
\n",
"
9.2250
\n",
"
S
\n",
"
\n",
"
\n",
"
1237
\n",
"
3
\n",
"
0
\n",
"
Svensson, Mr. Olof
\n",
"
H
\n",
"
24.0
\n",
"
7.7958
\n",
"
S
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" classe survivant nom sexe age prix \\\n",
"979 3 0 Lundahl, Mr. Johan Svensson H 51.0 7.0542 \n",
"1235 3 0 Svensson, Mr. Johan H 74.0 7.7750 \n",
"1236 3 1 Svensson, Mr. Johan Cervin H 14.0 9.2250 \n",
"1237 3 0 Svensson, Mr. Olof H 24.0 7.7958 \n",
"\n",
" port_depart \n",
"979 S \n",
"1235 S \n",
"1236 S \n",
"1237 S "
]
},
"execution_count": 15,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df[df['nom'].str.contains('Svensson', case=False, na=False)]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Et la femme la plus âgée ?"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.650865Z",
"iopub.status.busy": "2026-10-30T13:10:32.650708Z",
"iopub.status.idle": "2026-10-30T13:10:32.658210Z",
"shell.execute_reply": "2026-10-30T13:10:32.657793Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
classe
\n",
"
survivant
\n",
"
nom
\n",
"
sexe
\n",
"
age
\n",
"
prix
\n",
"
port_depart
\n",
"
\n",
" \n",
" \n",
"
\n",
"
61
\n",
"
1
\n",
"
1
\n",
"
Cavendish, Mrs. Tyrell William (Julia Florence...
\n",
"
F
\n",
"
76.0
\n",
"
78.85
\n",
"
S
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" classe survivant nom sexe \\\n",
"61 1 1 Cavendish, Mrs. Tyrell William (Julia Florence... F \n",
"\n",
" age prix port_depart \n",
"61 76.0 78.85 S "
]
},
"execution_count": 16,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Filtrer les individus dont sex == \"F\"\n",
"filtered_df = df.loc[df['sexe'] == \"F\"]\n",
"# Trouver l'âge maximum dans cette sous-table\n",
"max_age = filtered_df['age'].max()\n",
"# Sélectionner les lignes correspondant à cet âge maximum\n",
"oldest = filtered_df.loc[filtered_df['age'] == max_age]\n",
"# Afficher les résultats\n",
"oldest"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Selon le site il s'agit de Mrs Mary Eliza Compton, âgée de 64 ans et 8 mois, qui a survécu. À nouveau, notre jeu de données ne contient pas cette information :"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.659832Z",
"iopub.status.busy": "2026-10-30T13:10:32.659682Z",
"iopub.status.idle": "2026-10-30T13:10:32.667639Z",
"shell.execute_reply": "2026-10-30T13:10:32.667248Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
classe
\n",
"
survivant
\n",
"
nom
\n",
"
sexe
\n",
"
age
\n",
"
prix
\n",
"
port_depart
\n",
"
\n",
" \n",
" \n",
"
\n",
"
76
\n",
"
1
\n",
"
1
\n",
"
Compton, Miss. Sara Rebecca
\n",
"
F
\n",
"
39.0
\n",
"
83.1583
\n",
"
C
\n",
"
\n",
"
\n",
"
77
\n",
"
1
\n",
"
0
\n",
"
Compton, Mr. Alexander Taylor Jr
\n",
"
H
\n",
"
37.0
\n",
"
83.1583
\n",
"
C
\n",
"
\n",
"
\n",
"
78
\n",
"
1
\n",
"
1
\n",
"
Compton, Mrs. Alexander Taylor (Mary Eliza Ing...
\n",
"
F
\n",
"
64.0
\n",
"
83.1583
\n",
"
C
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" classe survivant nom sexe \\\n",
"76 1 1 Compton, Miss. Sara Rebecca F \n",
"77 1 0 Compton, Mr. Alexander Taylor Jr H \n",
"78 1 1 Compton, Mrs. Alexander Taylor (Mary Eliza Ing... F \n",
"\n",
" age prix port_depart \n",
"76 39.0 83.1583 C \n",
"77 37.0 83.1583 C \n",
"78 64.0 83.1583 C "
]
},
"execution_count": 17,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df[df['nom'].str.contains('Compton', case=False, na=False)]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Plus jeune et plus vieux passagers à avoir survécu"
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.669323Z",
"iopub.status.busy": "2026-10-30T13:10:32.669171Z",
"iopub.status.idle": "2026-10-30T13:10:32.682613Z",
"shell.execute_reply": "2026-10-30T13:10:32.682222Z"
}
},
"outputs": [
{
"data": {
"text/markdown": [
"**Plus jeune survivant :**"
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
classe
\n",
"
survivant
\n",
"
nom
\n",
"
sexe
\n",
"
age
\n",
"
prix
\n",
"
port_depart
\n",
"
\n",
" \n",
" \n",
"
\n",
"
763
\n",
"
3
\n",
"
1
\n",
"
Dean, Miss. Elizabeth Gladys \"Millvina\"
\n",
"
F
\n",
"
0.1667
\n",
"
20.575
\n",
"
S
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" classe survivant nom sexe age \\\n",
"763 3 1 Dean, Miss. Elizabeth Gladys \"Millvina\" F 0.1667 \n",
"\n",
" prix port_depart \n",
"763 20.575 S "
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/markdown": [
"**Plus vieux survivant :**"
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
classe
\n",
"
survivant
\n",
"
nom
\n",
"
sexe
\n",
"
age
\n",
"
prix
\n",
"
port_depart
\n",
"
\n",
" \n",
" \n",
"
\n",
"
14
\n",
"
1
\n",
"
1
\n",
"
Barkworth, Mr. Algernon Henry Wilson
\n",
"
H
\n",
"
80.0
\n",
"
30.0
\n",
"
S
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" classe survivant nom sexe age prix \\\n",
"14 1 1 Barkworth, Mr. Algernon Henry Wilson H 80.0 30.0 \n",
"\n",
" port_depart \n",
"14 S "
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# parmi les survivants, qui est le plus jeune et qui est le plus âgé ? \n",
"survivants = df[df['survivant'] == 1]\n",
"# plus jeune\n",
"min_age_survivant = survivants['age'].min()\n",
"plus_jeune_survivant = survivants[survivants['age'] == min_age_survivant]\n",
"display(Markdown(\"**Plus jeune survivant :**\"))\n",
"display(plus_jeune_survivant)\n",
"# plus âgé\n",
"max_age_survivant = survivants['age'].max()\n",
"plus_vieux_survivant = survivants[survivants['age'] == max_age_survivant]\n",
"display(Markdown(\"**Plus vieux survivant :**\"))\n",
"display(plus_vieux_survivant)\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Plus jeune et plus vieux passagers à être décédé"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.684398Z",
"iopub.status.busy": "2026-10-30T13:10:32.684205Z",
"iopub.status.idle": "2026-10-30T13:10:32.697500Z",
"shell.execute_reply": "2026-10-30T13:10:32.697107Z"
}
},
"outputs": [
{
"data": {
"text/markdown": [
"**Plus jeune décédé :**"
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"
"
],
"text/plain": [
" classe survivant nom sexe age prix port_depart\n",
"1235 3 0 Svensson, Mr. Johan H 74.0 7.775 S"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# parmi les decede, qui est le plus jeune et qui est le plus âgé ? \n",
"decede = df[df['survivant'] == 0]\n",
"# plus jeune\n",
"min_age_decede = decede['age'].min()\n",
"plus_jeune_decede = decede[decede['age'] == min_age_decede]\n",
"display(Markdown(\"**Plus jeune décédé :**\"))\n",
"display(plus_jeune_decede)\n",
"# plus âgé\n",
"max_age_decede = decede['age'].max()\n",
"plus_vieux_decede = decede[decede['age'] == max_age_decede]\n",
"display(Markdown(\"**Plus vieux décédé :**\"))\n",
"display(plus_vieux_decede)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Le capitaine du Titanic ?"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Si on cherche dans notre jeu de données le titre capitaine, on trouve le passager suivant :"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.699245Z",
"iopub.status.busy": "2026-10-30T13:10:32.699088Z",
"iopub.status.idle": "2026-10-30T13:10:32.706725Z",
"shell.execute_reply": "2026-10-30T13:10:32.706327Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
classe
\n",
"
survivant
\n",
"
nom
\n",
"
sexe
\n",
"
age
\n",
"
prix
\n",
"
port_depart
\n",
"
\n",
" \n",
" \n",
"
\n",
"
81
\n",
"
1
\n",
"
0
\n",
"
Crosby, Capt. Edward Gifford
\n",
"
H
\n",
"
70.0
\n",
"
71.0
\n",
"
S
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" classe survivant nom sexe age prix \\\n",
"81 1 0 Crosby, Capt. Edward Gifford H 70.0 71.0 \n",
"\n",
" port_depart \n",
"81 S "
]
},
"execution_count": 20,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df[df['nom'].str.contains('Capt', case=False, na=False)]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Ce n'est pas le capitaine du Titanic qui était Edward John Smith (1850-1912) mais il n'apparaît pas dans notre jeu de données."
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.708470Z",
"iopub.status.busy": "2026-10-30T13:10:32.708317Z",
"iopub.status.idle": "2026-10-30T13:10:32.716303Z",
"shell.execute_reply": "2026-10-30T13:10:32.715873Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
classe
\n",
"
survivant
\n",
"
nom
\n",
"
sexe
\n",
"
age
\n",
"
prix
\n",
"
port_depart
\n",
"
\n",
" \n",
" \n",
"
\n",
"
267
\n",
"
1
\n",
"
0
\n",
"
Smith, Mr. James Clinch
\n",
"
H
\n",
"
56.0
\n",
"
30.6958
\n",
"
C
\n",
"
\n",
"
\n",
"
268
\n",
"
1
\n",
"
0
\n",
"
Smith, Mr. Lucien Philip
\n",
"
H
\n",
"
24.0
\n",
"
60.0000
\n",
"
S
\n",
"
\n",
"
\n",
"
269
\n",
"
1
\n",
"
0
\n",
"
Smith, Mr. Richard William
\n",
"
H
\n",
"
NaN
\n",
"
26.0000
\n",
"
S
\n",
"
\n",
"
\n",
"
270
\n",
"
1
\n",
"
1
\n",
"
Smith, Mrs. Lucien Philip (Mary Eloise Hughes)
\n",
"
F
\n",
"
18.0
\n",
"
60.0000
\n",
"
S
\n",
"
\n",
"
\n",
"
564
\n",
"
2
\n",
"
1
\n",
"
Smith, Miss. Marion Elsie
\n",
"
F
\n",
"
40.0
\n",
"
13.0000
\n",
"
S
\n",
"
\n",
"
\n",
"
1215
\n",
"
3
\n",
"
0
\n",
"
Smith, Mr. Thomas
\n",
"
H
\n",
"
NaN
\n",
"
7.7500
\n",
"
Q
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" classe survivant nom sexe \\\n",
"267 1 0 Smith, Mr. James Clinch H \n",
"268 1 0 Smith, Mr. Lucien Philip H \n",
"269 1 0 Smith, Mr. Richard William H \n",
"270 1 1 Smith, Mrs. Lucien Philip (Mary Eloise Hughes) F \n",
"564 2 1 Smith, Miss. Marion Elsie F \n",
"1215 3 0 Smith, Mr. Thomas H \n",
"\n",
" age prix port_depart \n",
"267 56.0 30.6958 C \n",
"268 24.0 60.0000 S \n",
"269 NaN 26.0000 S \n",
"270 18.0 60.0000 S \n",
"564 40.0 13.0000 S \n",
"1215 NaN 7.7500 Q "
]
},
"execution_count": 21,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df[df['nom'].str.contains('Smith', case=True, na=False)]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Curiosités"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Sur Wikipedia, à la page [https://fr.m.wikipedia.org/wiki/Violet_Constance_Jessop](https://fr.m.wikipedia.org/wiki/Violet_Constance_Jessop)\n",
"il est indiqué que **Violet Constance Jessop** (2 octobre 1887 - 5 mai 1971), une hôtesse et infirmière britannique de la White Star Line, a survécu aux trois naufrages des navires de la classe Olympic. En effet, elle sert comme hôtesse à bord de l’*Olympic* quand il heurte le croiseur HMS Hawke le 20 septembre 1911. Elle se trouve également à bord du *Titanic* lorsqu’il fait naufrage le 15 avril 1912, et sert en tant qu’infirmière volontaire à bord du *Britannic* lorsqu’il fait naufrage en mer Égée le 21 novembre 1916.\n",
"\n",
"Vérifions si elle apparaît dans notre jeu de données :\n"
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.718195Z",
"iopub.status.busy": "2026-10-30T13:10:32.717998Z",
"iopub.status.idle": "2026-10-30T13:10:32.725916Z",
"shell.execute_reply": "2026-10-30T13:10:32.725518Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
classe
\n",
"
survivant
\n",
"
nom
\n",
"
sexe
\n",
"
age
\n",
"
prix
\n",
"
port_depart
\n",
"
\n",
" \n",
" \n",
"
\n",
"
501
\n",
"
2
\n",
"
1
\n",
"
Mellinger, Miss. Madeleine Violet
\n",
"
F
\n",
"
13.0
\n",
"
19.5
\n",
"
S
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" classe survivant nom sexe age prix \\\n",
"501 2 1 Mellinger, Miss. Madeleine Violet F 13.0 19.5 \n",
"\n",
" port_depart \n",
"501 S "
]
},
"execution_count": 22,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df[df['nom'].str.contains('Violet', case=False, na=False)]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"La seule Violet dans notre jeu de données est une passagère de 13 ans, qui a embarqué à Southampton en 2ème classe. Elle a survécu au naufrage du Titanic mais n'est probablement pas la même personne que Violet Jessop."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Et **Molly (Margaret Tobin) Brown** (1867-1932) ? Elle est connue pour être l'une des rescapées du naufrage du Titanic. Sauvée à bord du canot n° 6, dans lequel elle déplore le comportement du quartier-maître Robert Hichens, elle participe ensuite à la création du Comité des Survivants. Un siècle après le naufrage, Margaret Brown est presque exclusivement surnommée l'« Insubmersible Molly Brown », bien qu'elle n'ait jamais été appelée ainsi de son vivant. Il s'agit d'une invention du cinéma hollywoodien, qui s'empare de son histoire pour en faire un mythe, parfois très éloigné de la vérité, notamment dans la comédie musicale La Reine du Colorado (1964) et le drame Titanic (1997)."
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.727707Z",
"iopub.status.busy": "2026-10-30T13:10:32.727553Z",
"iopub.status.idle": "2026-10-30T13:10:32.735638Z",
"shell.execute_reply": "2026-10-30T13:10:32.735140Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
classe
\n",
"
survivant
\n",
"
nom
\n",
"
sexe
\n",
"
age
\n",
"
prix
\n",
"
port_depart
\n",
"
\n",
" \n",
" \n",
"
\n",
"
41
\n",
"
1
\n",
"
1
\n",
"
Brown, Mrs. James Joseph (Margaret Tobin)
\n",
"
F
\n",
"
44.0
\n",
"
27.7208
\n",
"
C
\n",
"
\n",
"
\n",
"
1249
\n",
"
3
\n",
"
0
\n",
"
Tobin, Mr. Roger
\n",
"
H
\n",
"
NaN
\n",
"
7.7500
\n",
"
Q
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" classe survivant nom sexe age \\\n",
"41 1 1 Brown, Mrs. James Joseph (Margaret Tobin) F 44.0 \n",
"1249 3 0 Tobin, Mr. Roger H NaN \n",
"\n",
" prix port_depart \n",
"41 27.7208 C \n",
"1249 7.7500 Q "
]
},
"execution_count": 23,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df[df['nom'].str.contains('Tobin', case=False, na=False)]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Doublons ?"
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.737401Z",
"iopub.status.busy": "2026-10-30T13:10:32.737212Z",
"iopub.status.idle": "2026-10-30T13:10:32.741137Z",
"shell.execute_reply": "2026-10-30T13:10:32.740707Z"
}
},
"outputs": [
{
"data": {
"text/plain": [
"(1307, 1309)"
]
},
"execution_count": 24,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df['nom'].nunique(), df['nom'].count() # 1307 noms différents pour 1309 passagers"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On constate que la colonne `nom` contient 1307 valeurs uniques pour 1309 lignes, ce qui indique la présence de doublons. Affichons les lignes concernées pour examiner ces cas particuliers :"
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.742892Z",
"iopub.status.busy": "2026-10-30T13:10:32.742741Z",
"iopub.status.idle": "2026-10-30T13:10:32.749965Z",
"shell.execute_reply": "2026-10-30T13:10:32.749584Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
classe
\n",
"
survivant
\n",
"
nom
\n",
"
sexe
\n",
"
age
\n",
"
prix
\n",
"
port_depart
\n",
"
\n",
" \n",
" \n",
"
\n",
"
725
\n",
"
3
\n",
"
1
\n",
"
Connolly, Miss. Kate
\n",
"
F
\n",
"
22.0
\n",
"
7.7500
\n",
"
Q
\n",
"
\n",
"
\n",
"
726
\n",
"
3
\n",
"
0
\n",
"
Connolly, Miss. Kate
\n",
"
F
\n",
"
30.0
\n",
"
7.6292
\n",
"
Q
\n",
"
\n",
"
\n",
"
924
\n",
"
3
\n",
"
0
\n",
"
Kelly, Mr. James
\n",
"
H
\n",
"
34.5
\n",
"
7.8292
\n",
"
Q
\n",
"
\n",
"
\n",
"
925
\n",
"
3
\n",
"
0
\n",
"
Kelly, Mr. James
\n",
"
H
\n",
"
44.0
\n",
"
8.0500
\n",
"
S
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" classe survivant nom sexe age prix port_depart\n",
"725 3 1 Connolly, Miss. Kate F 22.0 7.7500 Q\n",
"726 3 0 Connolly, Miss. Kate F 30.0 7.6292 Q\n",
"924 3 0 Kelly, Mr. James H 34.5 7.8292 Q\n",
"925 3 0 Kelly, Mr. James H 44.0 8.0500 S"
]
},
"execution_count": 25,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# keep=False signifie que chaque valeur de la colonne qui apparaît plus d'une fois sera marquée comme True\n",
"# Les valeurs uniques (apparaissant une seule fois) seront marquées comme False\n",
"mask = df['nom'].duplicated(keep=False)\n",
"df[mask]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Suppression de la colonne `nom`"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Dans notre jeu de données, la colonne `nom` ne semble plus être pertinente pour l'analys qui va suivre. Nous pouvons la supprimer en utilisant la méthode `.drop()`. En effet, certaines colonnes peuvent être négligées si elles ne sont pas pertinentes pour l'analyse et inutile de les garder dans le DataFrame. \n"
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.751808Z",
"iopub.status.busy": "2026-10-30T13:10:32.751627Z",
"iopub.status.idle": "2026-10-30T13:10:32.754708Z",
"shell.execute_reply": "2026-10-30T13:10:32.754317Z"
}
},
"outputs": [],
"source": [
"# errors=\"ignore\" pour éviter une erreur si la colonne n'existe pas\n",
"df.drop('nom', axis='columns', inplace=True, errors=\"ignore\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Préambule: valeurs uniques dans chaque colonne"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Pour chaque colonne, nous voulons connaître **combien de valeurs différentes** elle contient, **lesquelles** sont ces valeurs et **combien de fois chacune apparaît**.\n",
"\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Pour le nombre de valeurs différentes par colonne, nous avons :"
]
},
{
"cell_type": "code",
"execution_count": 27,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.756498Z",
"iopub.status.busy": "2026-10-30T13:10:32.756268Z",
"iopub.status.idle": "2026-10-30T13:10:32.761149Z",
"shell.execute_reply": "2026-10-30T13:10:32.760664Z"
}
},
"outputs": [
{
"data": {
"text/plain": [
"classe 3\n",
"survivant 2\n",
"sexe 2\n",
"age 98\n",
"prix 281\n",
"port_depart 3\n",
"dtype: int64"
]
},
"execution_count": 27,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.nunique()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On note que la colonne `classe` ne contient que 3 valeurs différentes, la colonne `survivant` ne contient que 2 valeurs différentes, la colonne `sexe` ne contient que 2 valeurs différentes, la colonne `port_depart` ne contient que 3 valeurs différentes. Pour ces colonnes ayant moins de 5 valeurs différentes, nous pouvons lister quelles sont ces valeurs différentes en utilisant la méthode `.unique()` :"
]
},
{
"cell_type": "code",
"execution_count": 28,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.762971Z",
"iopub.status.busy": "2026-10-30T13:10:32.762821Z",
"iopub.status.idle": "2026-10-30T13:10:32.767632Z",
"shell.execute_reply": "2026-10-30T13:10:32.767240Z"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"classe [1 2 3]\n",
"survivant [1 0]\n",
"sexe ['F' 'H']\n",
"port_depart ['S' 'C' nan 'Q']\n"
]
}
],
"source": [
"filtre = [df[col].nunique() < 5 for col in df.columns]\n",
"\n",
"for col in df[df.columns[filtre]]:\n",
" print(col, df[col].unique())"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Pour une colonne donnée, nous pouvons lister non seulement les valeurs différentes en utilisant la méthode `.unique()`, mais aussi compter le nombre d'occurrences de chaque valeurs différentes en utilisant la méthode `.value_counts()`. "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Pour analyser plus en profondeur une colonne catégorielle, il est utile d'afficher à la fois :\n",
"- Le **nombre absolu** d'occurrences de chaque catégorie\n",
"- Le **pourcentage** que représente chaque catégorie par rapport au total\n",
"- Le traitement des **valeurs manquantes** (si présentes)\n",
"\n",
"Comme ces informations sont intéressantes à afficher pour toute colonne catégorielle ou numérique discrète, nous allons créer une fonction réutilisable nommée `afficher_compte_et_pourcentage()` qui :\n",
"1. Calcule le nombre d'occurrences de chaque valeur unique\n",
"2. Calcule le pourcentage correspondant\n",
"3. Gère l'affichage des valeurs manquantes de manière explicite\n",
"4. Retourne un DataFrame synthétique facile à lire\n",
"\n",
"Cette fonction sera ensuite appelée pour différentes colonnes en changeant simplement le nom de la colonne passée en paramètre."
]
},
{
"cell_type": "code",
"execution_count": 29,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.769382Z",
"iopub.status.busy": "2026-10-30T13:10:32.769231Z",
"iopub.status.idle": "2026-10-30T13:10:32.772301Z",
"shell.execute_reply": "2026-10-30T13:10:32.771860Z"
}
},
"outputs": [],
"source": [
"def afficher_compte_et_pourcentage(df, colonne):\n",
" compte = df[colonne].value_counts(dropna=False)\n",
" pourcentage = df[colonne].value_counts(normalize=True, dropna=False) * 100\n",
" resultat = pd.DataFrame({'Nb occurrences': compte, 'Pourcentage (%)': round(pourcentage, 2)})\n",
" # Remplacer 'nan' dans l'index par 'Valeur manquante'\n",
" resultat.index = resultat.index.map(lambda x: 'Valeur manquante' if pd.isna(x) else str(x))\n",
" return resultat"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Colonne `classe`"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Combien de passagers étaient dans chaque classe ?\n",
"\n",
"Dans la colonne `classe` seulement trois catégories sont possibles : 1, 2 ou 3. La méthode `.value_counts()` permet de compter le nombre d'occurrences pour chaque catégorie unique dans une colonne. "
]
},
{
"cell_type": "code",
"execution_count": 30,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.774064Z",
"iopub.status.busy": "2026-10-30T13:10:32.773884Z",
"iopub.status.idle": "2026-10-30T13:10:32.791092Z",
"shell.execute_reply": "2026-10-30T13:10:32.790657Z"
}
},
"outputs": [
{
"data": {
"text/markdown": [
"**Résumé statistique de la colonne 'classe'**"
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"
"
],
"text/plain": [
" Proportions\n",
"classe \n",
"3 0.541635\n",
"1 0.246753\n",
"2 0.211612"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Résumé statistique de la colonne 'classe'\n",
"# ================================================================\n",
"\n",
"colonne = df['classe']\n",
"\n",
"display(Markdown(\"**Résumé statistique de la colonne 'classe'**\"))\n",
"display(colonne.describe().to_frame())\n",
"\n",
"display(Markdown(\"**Nombre de valeurs uniques**\"))\n",
"display(colonne.nunique())\n",
"\n",
"display(Markdown(\"**Valeurs uniques**\"))\n",
"display(colonne.unique())\n",
"\n",
"display(Markdown(\"**Combien d'occurrences de chaque valeur unique**\"))\n",
"display(colonne.value_counts().to_frame('Occurrences'))\n",
"\n",
"display(Markdown(\"**Proportions des valeurs uniques**\"))\n",
"display(colonne.value_counts(normalize=True).to_frame('Proportions')) # proportions"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Selon le site [Titanic Facts](https://titanicfacts.net/titanic-passengers/) il est indiqué qu'il y avait 2222 personnes à bord du Titanic (passagers et équipage) dont 1327 de passagers effectivement à bord et répartis comme suit : 324 en première classe, 284 en deuxième classe et 709 en troisième classe. \n",
"Il y avait 49% de places passagers inutilisées.\n",
"Nos données semblent cohérentes avec ces informations."
]
},
{
"cell_type": "code",
"execution_count": 31,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.792837Z",
"iopub.status.busy": "2026-10-30T13:10:32.792681Z",
"iopub.status.idle": "2026-10-30T13:10:32.798741Z",
"shell.execute_reply": "2026-10-30T13:10:32.798345Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Nb occurrences
\n",
"
Pourcentage (%)
\n",
"
\n",
"
\n",
"
classe
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
3
\n",
"
709
\n",
"
54.16
\n",
"
\n",
"
\n",
"
1
\n",
"
323
\n",
"
24.68
\n",
"
\n",
"
\n",
"
2
\n",
"
277
\n",
"
21.16
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Nb occurrences Pourcentage (%)\n",
"classe \n",
"3 709 54.16\n",
"1 323 24.68\n",
"2 277 21.16"
]
},
"execution_count": 31,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"afficher_compte_et_pourcentage(df, 'classe')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Ce tableau montre que la majorité des passagers (plus précisément 54 %) étaient de 3e classe. Fait intéressant, il y avait plus de passagers en première classe (323 passagers) qu'en seconde classe (277 passagers), même si la différence n'est pas très grande."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Pour avoir une idée visuelle du nombre de passagers par classe, on peut afficher ces informations \n",
"\n",
"- soit avec la fonction `.plot()` de pandas en spécifiant le type de graphique `bar` ou `pie`, \n",
"- soit avec la fonction `countplot()` de Seaborn \n",
"- soit avec les fonctions `histogram()` ou `pie()` de Plotly."
]
},
{
"cell_type": "code",
"execution_count": 32,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:32.800520Z",
"iopub.status.busy": "2026-10-30T13:10:32.800333Z",
"iopub.status.idle": "2026-10-30T13:10:33.306007Z",
"shell.execute_reply": "2026-10-30T13:10:33.305281Z"
}
},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Combien de fois chaque valeur unique apparaît dans la colonne ?\n",
"# ==================================================================\n",
"\n",
"colonne = 'classe'\n",
"\n",
"# Avec Pandas\n",
"# ------------------------------\n",
"\n",
"# Avec la méthode value_counts() de la série, on peut obtenir le compte de chaque valeur unique dans la colonne.\n",
"df[colonne].value_counts().sort_index().plot(kind='bar'); # histogramme\n",
"plt.title(f'Pandas : Histogramme de la colonne {colonne}')\n",
"plt.show()\n",
"\n",
"\n",
"# Avec seaborn\n",
"# ------------------------------\n",
"ax = sns.countplot(data=df, x=colonne); # histogramme\n",
"plt.title(f'Seaborn : Histogramme de la colonne {colonne}')\n",
"plt.show()\n",
"# Ajouter les annotations directement sur les barres\n",
"# for p in ax.patches: # Les rectangles/barres sont dans ax.patches\n",
"# ax.annotate(f'{int(p.get_height())}', # Le texte est la hauteur de la barre\n",
"# (p.get_x() + p.get_width() / 2., p.get_height()), # Position : centre de la barre\n",
"# ha='center', va='bottom', fontsize=10, color='black') # Alignement et style\n",
"\n",
"\n",
"# Avec plotly.express\n",
"# ------------------------------\n",
"px.histogram(df, x=colonne, text_auto=True, title=f'Plotly : Histogramme de la colonne {colonne}').update_layout(bargap=0.2).show()\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Et en pourcentage ?"
]
},
{
"cell_type": "code",
"execution_count": 33,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:33.308845Z",
"iopub.status.busy": "2026-10-30T13:10:33.308602Z",
"iopub.status.idle": "2026-10-30T13:10:33.667767Z",
"shell.execute_reply": "2026-10-30T13:10:33.667363Z"
}
},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Avec Pandas et dans un camembert\n",
"# ------------------------------\n",
"df['sexe'].value_counts().plot(kind='pie', autopct='%1.2f%%')\n",
"plt.title('Pandas : Pourcentage par sexe')\n",
"plt.show()\n",
"\n",
"# Avec Seaborn\n",
"# ------------------------------\n",
"# 'catplot()': Figure-level interface for drawing categorical plots onto a FacetGrid.\n",
"# sns.catplot(x='sexe', data=df, kind='count');\n",
"sns.catplot(x='sexe', data=df, kind='count', stat='percent')\n",
"plt.title('Seaborn : Pourcentage par sexe')\n",
"plt.show()\n",
"\n",
"# Avec plotly.express\n",
"# ------------------------------\n",
"# px.pie(df, names='sexe').show()\n",
"# px.histogram(df, x='sexe', text_auto=True).show()\n",
"px.pie(df, \n",
" names='sexe', \n",
" color='sexe', \n",
" title='Plotly : Pourcentage par sexe',\n",
" ).update_traces( texttemplate=\"%{label} : %{value} (%{percent})\" ,\n",
" textposition=\"outside\", # texte à l’extérieur du secteur\n",
" pull=[0.05]*len(df['sexe'].unique())# léger décalage pour chaque secteur\n",
" ).show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Colonne `survivant`"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Selon le site [Titanic Facts](https://titanicfacts.net/titanic-passengers/), 706 personnes ont survécu, dont 492 passagers et 214 membres d'équipage. Vérifions si nos données confirment cette information :"
]
},
{
"cell_type": "code",
"execution_count": 37,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:33.937445Z",
"iopub.status.busy": "2026-10-30T13:10:33.937234Z",
"iopub.status.idle": "2026-10-30T13:10:33.943430Z",
"shell.execute_reply": "2026-10-30T13:10:33.942891Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Nb occurrences
\n",
"
Pourcentage (%)
\n",
"
\n",
"
\n",
"
survivant
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
809
\n",
"
61.8
\n",
"
\n",
"
\n",
"
1
\n",
"
500
\n",
"
38.2
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Nb occurrences Pourcentage (%)\n",
"survivant \n",
"0 809 61.8\n",
"1 500 38.2"
]
},
"execution_count": 37,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Compte et le pourcentage de chaque catégorie\n",
"\n",
"afficher_compte_et_pourcentage(df, 'survivant')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Seulement environ 38 % des personnes ont survécu. Existe-t-il un moyen de savoir quels types de personnes ont eu le plus de chances de survivre ? Y a-t-il des caractéristiques particulières partagées par les survivants ? Pour répondre à ces questions, nous examinerons plus tard les autres colonnes et leurs relations avec la colonne survivant.\n"
]
},
{
"cell_type": "code",
"execution_count": 38,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:33.945255Z",
"iopub.status.busy": "2026-10-30T13:10:33.945103Z",
"iopub.status.idle": "2026-10-30T13:10:33.947453Z",
"shell.execute_reply": "2026-10-30T13:10:33.947016Z"
}
},
"outputs": [],
"source": [
"# Barres : matplotlib ou seaborn\n",
"\n",
"# df['survivant'].value_counts().plot(kind='bar');\n",
"# sns.countplot(data=df, x='survivant');\n",
"\n",
"# df['survivant'].value_counts(normalize=True).plot(kind='bar');\n",
"# sns.countplot(data=df, x='survivant', stat='percent');\n",
"# sns.countplot(data=df, x='survivant', stat='proportion');"
]
},
{
"cell_type": "code",
"execution_count": 39,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:33.949214Z",
"iopub.status.busy": "2026-10-30T13:10:33.949022Z",
"iopub.status.idle": "2026-10-30T13:10:34.047706Z",
"shell.execute_reply": "2026-10-30T13:10:34.047216Z"
}
},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Pourcentage dans un camembert\n",
"# ================================================================\n",
"\n",
"# Avec Pandas\n",
"# ------------------------------\n",
"df['survivant'].value_counts().plot(kind='pie', autopct='%1.2f%%')\n",
"plt.title('Pandas : Pourcentage de survivants')\n",
"plt.show()\n",
"\n",
"\n",
"# Avec plotly.express\n",
"# ------------------------------\n",
"\n",
"# px.pie(df, names='survivant')\n",
"\n",
"# Mieux : chaque secteur affiche directement le nom (1 ou 0), le nombre d’occurrences et le pourcentage.\n",
"px.pie(df, \n",
" names='survivant', \n",
" color='survivant', \n",
" color_discrete_sequence=px.colors.qualitative.Pastel,\n",
" title='Plotly : Pourcentage de survivants'\n",
" ).update_traces( texttemplate=\"%{label} : %{value} (%{percent})\" ,\n",
" textposition=\"outside\", # texte à l’extérieur du secteur\n",
" pull=[0.05]*len(df['survivant'].unique())# léger décalage pour chaque secteur)\n",
" ).show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Colonne `age`"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Analysons d'abord les données numériques pour la colonne de l'âge. On peut choisir de n'afficher que certaines informations (par exemple, la moyenne et l'écart-type) en utilisant la méthode `.agg()` :"
]
},
{
"cell_type": "code",
"execution_count": 40,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:34.049415Z",
"iopub.status.busy": "2026-10-30T13:10:34.049257Z",
"iopub.status.idle": "2026-10-30T13:10:34.060500Z",
"shell.execute_reply": "2026-10-30T13:10:34.060011Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
count
\n",
"
missing_count
\n",
"
missing_percent
\n",
"
mean
\n",
"
median
\n",
"
std
\n",
"
min
\n",
"
max
\n",
"
nunique
\n",
"
\n",
" \n",
" \n",
"
\n",
"
age
\n",
"
1046.0
\n",
"
263.0
\n",
"
20.091673
\n",
"
29.881135
\n",
"
28.0
\n",
"
14.4135
\n",
"
0.1667
\n",
"
80.0
\n",
"
98.0
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" count missing_count missing_percent mean median std \\\n",
"age 1046.0 263.0 20.091673 29.881135 28.0 14.4135 \n",
"\n",
" min max nunique \n",
"age 0.1667 80.0 98.0 "
]
},
"execution_count": 40,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Statistiques descriptives pour une variable numérique\n",
"# ==================================================================\n",
"\n",
"# Méthode 1 : la méthode describe()\n",
"# ----------------------------------\n",
"# df[['age']].describe()\n",
"\n",
"# Méthode 2 : en choisissant les statistiques souhaitées\n",
"# ------------------------------------------------------\n",
"# df[['age']].agg(['count','mean','median','std','min','max','nunique'])\n",
"\n",
"# Méthode 3 : encore mieux, en définissant une fonction pour le nombre de valeurs manquantes et l'utilisant dans agg()\n",
"# ----------------------------------------------------------------------------------------------------------------------\n",
"\n",
"missing_percent = lambda x: x.isna().mean() * 100\n",
"missing_percent.__name__ = 'missing_percent' # pour un joli nom dans le résultat\n",
"\n",
"def missing(x):\n",
" return x.isna().sum()\n",
"\n",
"df[['age']].agg({\n",
" 'age': ['count', missing, missing_percent, 'mean', 'median', 'std', 'min', 'max', 'nunique']\n",
"}).rename(index={'missing': 'missing_count'}).T\n",
"\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Comme on peut le voir, l'âge moyen des passagers est d'environ 29,9 ans, avec un écart-type de 14,5 ans. Cependant, il manque 263 valeurs dans cette colonne, ce qui représente environ 20 % des données. Nous pourrions soit supprimer ces lignes, soit remplacer les valeurs manquantes par la moyenne de l'âge des passagers par exemple ou simplement ne pas les prendre en compte dans notre analyse. Pour le moment nous n'allons pas modifier les données."
]
},
{
"cell_type": "code",
"execution_count": 41,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:34.062272Z",
"iopub.status.busy": "2026-10-30T13:10:34.062102Z",
"iopub.status.idle": "2026-10-30T13:10:34.156645Z",
"shell.execute_reply": "2026-10-30T13:10:34.156121Z"
}
},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Boxplot pour la variable 'age' : \n",
"# quartiles, médiane, valeurs extrêmes et outliers\n",
"# ==================================================================\n",
"sns.boxplot(x=df[\"age\"])\n",
"plt.title(\"age Boxplot\")\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On peut afficher l'histogramme de la distribution de l'âge des passagers en utilisant le type de graphique `hist` de la fonction `.plot()` ou utiliser la fonction `histplot()` de Seaborn ou `histogram()` de Plotly Express. Dans les trois cas on peut décider de regrouper les âges par intervalles de 10 ans en utilisant l'argument `bins`.\n",
"\n",
"On peut aussi ajouter une courbe de densité (KDE = Kernel Density Estimation) : c’est une courbe lissée qui représente la distribution d’un ensemble de données.\n",
"Elle sert à visualiser la forme générale de la distribution, sans dépendre des bins de l’histogramme. "
]
},
{
"cell_type": "code",
"execution_count": 42,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:34.158420Z",
"iopub.status.busy": "2026-10-30T13:10:34.158267Z",
"iopub.status.idle": "2026-10-30T13:10:34.960210Z",
"shell.execute_reply": "2026-10-30T13:10:34.959745Z"
}
},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Pandas\n",
"# ------------------------------\n",
"# Rque : Pandas n’a pas de kde=True dans plot(kind='hist'), il a une fonction KDE séparée. \n",
"# Utiliser density=True pour que l’histogramme et le KDE soient sur la même échelle \n",
"# Dans ce cas en ordonnée on a la densité de probabilité (somme des barres = 1) et non plus le nombre d’occurrences\n",
"df['age'].plot(kind='hist', bins=range(0, 90), edgecolor='white', title=\"Histogramme avec plot, bins=range(0, 90)\") #, density=True)\n",
"# df['age'].plot(kind='kde')\n",
"plt.show()\n",
"\n",
"df['age'].plot(kind='hist', bins=range(0, 90, 10), edgecolor='white', title=\"Histogramme avec plot, bins=range(0, 90, 10)\") #, density=True)\n",
"# df['age'].plot(kind='kde')\n",
"plt.show()\n",
"\n",
"\n",
"# Seaborn\n",
"# ------------------------------\n",
"# Rque : seaborn a kde=True dans histplot() et le rescale automatiquement\n",
"sns.histplot(data=df, x='age', bins=range(0, 90), kde=True);\n",
"plt.title(\"Histogramme avec histplot de Seaborn, bins=range(0, 90)\");\n",
"plt.show()\n",
"\n",
"sns.histplot(df['age'], bins=range(0, 90, 10), edgecolor='white', kde=True);\n",
"plt.title(\"Histogramme avec histplot de Seaborn, bins=range(0, 90, 10)\")\n",
"plt.show()\n",
"\n",
"\n",
"# Avec plotly.express\n",
"# ------------------------------\n",
"px.histogram(df, x='age', nbins=90, title='Histogramme des âges').show()\n",
"\n",
"# NB Avec plotly.express, la gestion des bins est différente de matplotlib ou seaborn : \n",
"# il faut préciser xbins pour définir les bornes des classes.\n",
"px.histogram(\n",
" df, \n",
" x='age', \n",
" title=\"Histogramme Plotly : bins de 10 ans\"\n",
").update_traces( xbins=dict( start=0, end=90, size=10 ) ).update_layout(bargap=0.2).show() "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"La répartition par âge est globalement en forme de cloche, avec un pic autour de la tranche d'âge 20-30 ans. Il est intéressant de noter qu'il y a un pic local vers la gauche dans la tranche d'âge 0-5 ans, ce qui indique la présence de nombreux nourrissons et enfants en bas âge à bord. \n",
"D'un point de vue historique, cela est logique, car à cette époque, les voyages concernaient souvent de jeunes adultes qui commençaient une nouvelle vie, parfois accompagnés d'enfants. Pour l'analyse de survie, cela est important, car les enfants étaient souvent prioritaires lors des opérations de sauvetage."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Nouvelle Colonne `age_group`"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On peut regrouper les données en découpant un ensemble de valeurs selon des intervalles, comme on l'a vu sur les graphiques précédents. Cela semble pertinent pour l'âge. On va donc ajouter une colonne `age_group` à notre DataFrame qui contiendra l'âge regroupé par intervalles de 10 ans. Pour cela,\n",
"\n",
"1. on peut utiliser la méthode `pd.cut()` en spécifiant les valeurs à découper et les étiquettes des intervalles,\n",
"\n",
"2. on peut passer à `pd.map()` une fonction que nous aurons définie pour regrouper les âges par intervalles de 10 ans."
]
},
{
"cell_type": "code",
"execution_count": 43,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:34.962305Z",
"iopub.status.busy": "2026-10-30T13:10:34.962118Z",
"iopub.status.idle": "2026-10-30T13:10:34.974219Z",
"shell.execute_reply": "2026-10-30T13:10:34.973775Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
classe
\n",
"
survivant
\n",
"
sexe
\n",
"
age
\n",
"
prix
\n",
"
port_depart
\n",
"
age_group
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
1
\n",
"
1
\n",
"
F
\n",
"
29.0000
\n",
"
211.3375
\n",
"
S
\n",
"
20-30
\n",
"
\n",
"
\n",
"
1
\n",
"
1
\n",
"
1
\n",
"
H
\n",
"
0.9167
\n",
"
151.5500
\n",
"
S
\n",
"
0-10
\n",
"
\n",
"
\n",
"
2
\n",
"
1
\n",
"
0
\n",
"
F
\n",
"
2.0000
\n",
"
151.5500
\n",
"
S
\n",
"
0-10
\n",
"
\n",
"
\n",
"
3
\n",
"
1
\n",
"
0
\n",
"
H
\n",
"
30.0000
\n",
"
151.5500
\n",
"
S
\n",
"
20-30
\n",
"
\n",
"
\n",
"
4
\n",
"
1
\n",
"
0
\n",
"
F
\n",
"
25.0000
\n",
"
151.5500
\n",
"
S
\n",
"
20-30
\n",
"
\n",
"
\n",
"
...
\n",
"
...
\n",
"
...
\n",
"
...
\n",
"
...
\n",
"
...
\n",
"
...
\n",
"
...
\n",
"
\n",
"
\n",
"
1304
\n",
"
3
\n",
"
0
\n",
"
F
\n",
"
14.5000
\n",
"
14.4542
\n",
"
C
\n",
"
10-20
\n",
"
\n",
"
\n",
"
1305
\n",
"
3
\n",
"
0
\n",
"
F
\n",
"
NaN
\n",
"
14.4542
\n",
"
C
\n",
"
Unknown
\n",
"
\n",
"
\n",
"
1306
\n",
"
3
\n",
"
0
\n",
"
H
\n",
"
26.5000
\n",
"
7.2250
\n",
"
C
\n",
"
20-30
\n",
"
\n",
"
\n",
"
1307
\n",
"
3
\n",
"
0
\n",
"
H
\n",
"
27.0000
\n",
"
7.2250
\n",
"
C
\n",
"
20-30
\n",
"
\n",
"
\n",
"
1308
\n",
"
3
\n",
"
0
\n",
"
H
\n",
"
29.0000
\n",
"
7.8750
\n",
"
S
\n",
"
20-30
\n",
"
\n",
" \n",
"
\n",
"
1309 rows × 7 columns
\n",
"
"
],
"text/plain": [
" classe survivant sexe age prix port_depart age_group\n",
"0 1 1 F 29.0000 211.3375 S 20-30\n",
"1 1 1 H 0.9167 151.5500 S 0-10\n",
"2 1 0 F 2.0000 151.5500 S 0-10\n",
"3 1 0 H 30.0000 151.5500 S 20-30\n",
"4 1 0 F 25.0000 151.5500 S 20-30\n",
"... ... ... ... ... ... ... ...\n",
"1304 3 0 F 14.5000 14.4542 C 10-20\n",
"1305 3 0 F NaN 14.4542 C Unknown\n",
"1306 3 0 H 26.5000 7.2250 C 20-30\n",
"1307 3 0 H 27.0000 7.2250 C 20-30\n",
"1308 3 0 H 29.0000 7.8750 S 20-30\n",
"\n",
"[1309 rows x 7 columns]"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Première méthode\n",
"# ------------------------------\n",
"\n",
"df['age_group'] = pd.cut(df['age'], bins=range(0, 90, 10), labels=[f'{i}-{i+10}' for i in range(0, 80, 10)])\n",
"df['age_group'] = df['age_group'].astype(str).replace('nan', 'Unknown')\n",
"\n",
"# Deuxième méthode\n",
"# ------------------------------\n",
"# df['age_group'] = df['age'].map(lambda x: f'{int(x/10)*10}-{int(x/10)*10+10}' if not np.isnan(x) else 'Unknown')\n",
"\n",
"display(df)\n"
]
},
{
"cell_type": "code",
"execution_count": 44,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:34.976105Z",
"iopub.status.busy": "2026-10-30T13:10:34.975900Z",
"iopub.status.idle": "2026-10-30T13:10:34.982600Z",
"shell.execute_reply": "2026-10-30T13:10:34.982195Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Avec Pandas et dans un camembert\n",
"# ------------------------------\n",
"df['age_group'].value_counts().plot(kind='pie', \n",
" autopct='%1.2f%%', \n",
" #explode=[0.1]*len(df['age_group'].unique())\n",
" )\n",
"plt.title(\"Pandas : Pourcentage par groupe d'âge\")\n",
"plt.show()\n",
"\n",
"# Avec plotly.express\n",
"# ------------------------------\n",
"\n",
"# px.pie(df, names='age_group')\n",
"\n",
"# Mieux \n",
"px.pie(df, \n",
" names='age_group', \n",
" color='age_group', \n",
" title=\"Plotly : Pourcentage par groupe d'âge\",\n",
" color_discrete_sequence=px.colors.qualitative.Pastel\n",
" ).update_traces( texttemplate=\"%{label} : %{value} (%{percent})\" ,\n",
" textposition=\"outside\", # texte à l’extérieur du secteur\n",
" pull=[0.05]*len(df['age_group'].unique())) # léger décalage pour chaque secteur)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Liens entre deux ou plusieurs colonnes"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Jusqu'à présent, nous n'avons exploré que les statistiques descriptives, mais les véritables informations apparaissent lorsque nous relions ces caractéristiques à la variable cible : `survivant`. \n",
"Les passagers plus jeunes ont-ils un taux de survie plus élevé ? Les passagers payant plus cher, souvent ceux de première classe, avaient-ils un meilleur accès aux canots de sauvetage ? Ce sont les questions que nous allons aborder ensuite en superposant les résultats de survie à nos distributions. Intuitivement, on pourrait s'attendre à ce que les enfants et les passagers de classe supérieure bénéficient d'un avantage en termes de survie."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"
\n",
"\n",
"Dans cette partie nous allons utiliser les méthodes suivantes :\n",
"\n",
"```python\n",
".crosstab()\n",
".plot(kind='bar') .plot(kind='hist') .value_counts().plot(kind='pie')\n",
"sns.countplot() sns.histplot() sns.catplot()\n",
"px.histogram() px.pie() \n",
"```\n",
"\n",
"
"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Analyse bivariée, vue d'ensemble"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Pour une vue d'ensemble, nous pouvons utiliser la fonction `pairplot()` de Seaborn : elle crée une matrice de graphiques où chaque variable numérique du dataset est comparée à toutes les autres. Chaque ligne et colonne correspond à une variable. Les graphes diagonaux montrent généralement la distribution de chaque variable (histogramme ou KDE). Les graphes hors diagonale montrent la relation entre chaque paire de variables (scatter plot par défaut).\n",
"\n",
"Avec ce pairplot, on peut visualiser d’un coup d’œil les patterns les plus importants comme la survie selon le sexe et la classe ou encore l’âge des survivants comparé à celui des non-survivants. "
]
},
{
"cell_type": "code",
"execution_count": 47,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:35.432317Z",
"iopub.status.busy": "2026-10-30T13:10:35.432157Z",
"iopub.status.idle": "2026-10-30T13:10:35.435901Z",
"shell.execute_reply": "2026-10-30T13:10:35.435485Z"
}
},
"outputs": [],
"source": [
"df['sexe_num'] = df['sexe'].astype('category').cat.codes\n",
"# df['port_depart_num'] = df['port_depart'].astype('category').cat.codes"
]
},
{
"cell_type": "code",
"execution_count": 48,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:35.437662Z",
"iopub.status.busy": "2026-10-30T13:10:35.437476Z",
"iopub.status.idle": "2026-10-30T13:10:38.433718Z",
"shell.execute_reply": "2026-10-30T13:10:38.433303Z"
}
},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Sélection des variables numériques et catégorielles codées\n",
"# ==========================================================\n",
"pairplot_vars = ['age', 'prix', 'classe', 'sexe_num']\n",
"\n",
"\n",
"\n",
"# Avec Seaborn\n",
"# ------------------------------\n",
"\n",
"g = sns.pairplot(\n",
" data=df[pairplot_vars + ['survivant']], # inclure la variable cible\n",
" hue='survivant', # couleur selon survie\n",
" palette={0: 'red', 1: 'green'}, # couleurs personnalisées\n",
" diag_kind='kde', # KDE pour la diagonale\n",
" markers=['o', 's'], # forme différente selon la classe\n",
" plot_kws={'alpha':0.6, 's':50} # transparence et taille des points\n",
")\n",
"# Ajuster les légendes\n",
"# for t, l in zip(g._legend.texts, [\"Décédé\", \"Survécu\"]):\n",
"# t.set_text(l)\n",
"plt.title('Avec Seaborn')\n",
"plt.show()\n",
"\n",
"\n",
"# Avec plotly.express\n",
"# ------------------------------\n",
"# fig = px.scatter_matrix(\n",
"# df,\n",
"# dimensions=pairplot_vars, # variables à afficher\n",
"# color=df['survivant'].map({0: 'Décédé', 1: 'Survécu'}),\n",
"# color_discrete_map={'Décédé': 'red', 'Survécu': 'green'},\n",
"# symbol='survivant', # forme différente selon la classe\n",
"# symbol_map={0: 'circle', 1: 'square'},\n",
"# opacity=0.6, # transparence\n",
"# height=800, # hauteur de la figure\n",
"# title='Avec Plotly'\n",
"# )\n",
"# fig.update_traces(\n",
"# marker=dict(size=5), # taille des points\n",
"# diagonal_visible=False # masquer la diagonale (pas de KDE natif))\n",
"# )\n",
"# fig.update_layout(width=1000, height=1000)\n",
"# fig.show()\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"\n",
"\n",
"Caveat\n",
"\n",
"\n",
"**Le paradoxe de Simpson : la difficulté de comparer des pourcentages** \n",
"\n",
"L'étude des relations entre 3 ou plus caractéristiques (par exemple **sexe**, **classe** et **survie**), peut être délicat. En effet, il existe un phénomène statistique contre-intuitif appelé **paradoxe de Simpson**. \n",
"Ce paradoxe se produit lorsqu’une tendance observée dans **chaque sous-groupe** disparaît ou **s’inverse** lorsque l’on regroupe les données. \n",
"Ainsi, les pourcentages globaux peuvent raconter une histoire différente (et parfois totalement fausse) de celle révélée par les sous-groupes.\n",
"\n",
"**Exemple illustratif : l’inversion des pourcentages**\n",
"\n",
"Considérons deux médicaments (A et B) testés sur des hommes et des femmes. Les tableaux suivants indiquent le nombre de personnes traitées et le nombre de guérisons.\n",
"
\n",
"
\n",
"
\n",
"
Médicament A
\n",
"
Médicament B
\n",
"
\n",
"
\n",
"
Genre
\n",
"
Total
\n",
"
Guéris
\n",
"
Total
\n",
"
Guéris
\n",
"
\n",
"
\n",
"
Hommes
\n",
"
5
\n",
"
1
\n",
"
10
\n",
"
3
\n",
"
\n",
"
\n",
"
Femmes
\n",
"
9
\n",
"
4
\n",
"
2
\n",
"
1
\n",
"
\n",
"
\n",
"\n",
"\n",
"*Taux de guérison*\n",
"\n",
"| Groupe | Médicament A | Médicament B |\n",
"|---------|--------------|--------------|\n",
"| Hommes | $1/5=20\\%$ | $3/10=30\\%$ |\n",
"| Femmes | $4/9=44\\%$ | $1/2=50\\%$ |\n",
"| Total | $5/14=36\\%$ | $4/12=33\\%$ |\n",
"\n",
"\n",
"Conclusion :\n",
"\n",
"- chez les *hommes*, le médicament B guérit davantage que A ;\n",
"- chez les *femmes*, le médicament B guérit encore davantage que A ; \n",
"- mais **une fois les deux groupes combinés**, le médicament A apparaît meilleur que B.\n",
"\n",
"Le classement s’inverse au niveau global : c’est le **paradoxe de Simpson**. Le renversement provient de la **répartition très inégale des tailles de groupes**. Mathématiquement, le paradoxe vient du fait que les inégalités \n",
"\n",
"$$\n",
"\\frac{a}{b}<\\frac{A}{B}\n",
"\\quad\\text{et}\\quad\n",
"\\frac{c}{d}<\\frac{C}{D}\n",
"$$\n",
"\n",
"n’impliquent pas que \n",
"\n",
"$$\n",
"\\frac{a+c}{b+d}<\\frac{A+C}{B+D}.\n",
"$$\n",
"\n",
"\n",
"\n",
"Dans la suite, il faudra faire très attention à comparer uniquement les pourcentages globaux de survivants par catégorie d'une autre variable si les distributions des sous-groupes sont très différentes.\n",
"\n",
""
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Classe / Survie"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Observons la colonne `classe`. \n",
"Il y a trois catégories dans cette colonne : les passagers de 1re classe, de 2e classe et de 3e classe représentés respectivement par les chiffres 1, 2 et 3. \n",
"Est-ce que les passagers de 1re classe ont eu la priorité pour monter à bord du navire et peut-être la priorité pour être sauvés dans les canots de sauvetage ? Autrement-dit, est-ce que les passagers de 1re classe ont un taux de survie plus élevé par rapport à ceux de 2e ou 3e classe ? Vérifions si c'est le cas."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"La méthode `.crosstab()` permet de créer un **tableau de contingence** pour voir combien de personnes de chaque classe ont survécu ou non. "
]
},
{
"cell_type": "code",
"execution_count": 49,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:38.436470Z",
"iopub.status.busy": "2026-10-30T13:10:38.436300Z",
"iopub.status.idle": "2026-10-30T13:10:38.456975Z",
"shell.execute_reply": "2026-10-30T13:10:38.456535Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
survivant
\n",
"
0
\n",
"
1
\n",
"
Total
\n",
"
\n",
"
\n",
"
classe
\n",
"
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
1
\n",
"
123
\n",
"
200
\n",
"
323
\n",
"
\n",
"
\n",
"
2
\n",
"
158
\n",
"
119
\n",
"
277
\n",
"
\n",
"
\n",
"
3
\n",
"
528
\n",
"
181
\n",
"
709
\n",
"
\n",
"
\n",
"
Total
\n",
"
809
\n",
"
500
\n",
"
1309
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
"survivant 0 1 Total\n",
"classe \n",
"1 123 200 323\n",
"2 158 119 277\n",
"3 528 181 709\n",
"Total 809 500 1309"
]
},
"execution_count": 49,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"pd.crosstab(df['classe'], df['survivant'], margins=True, margins_name='Total')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Même résultat avec `.pivot_table()` : la fonction `size` compte le nombre d'occurrences dans chaque groupe et `fill_value=0` remplace les NaN par 0 (utile si certaines combinaisons de valeurs n'existent pas). La fonction `pivot_table()` ne propose pas de manière directe pour ajouter des marges comme `crosstab()`, il faut donc additionner les marges séparément."
]
},
{
"cell_type": "code",
"execution_count": 50,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:38.458777Z",
"iopub.status.busy": "2026-10-30T13:10:38.458621Z",
"iopub.status.idle": "2026-10-30T13:10:38.468271Z",
"shell.execute_reply": "2026-10-30T13:10:38.467838Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
survivant
\n",
"
0
\n",
"
1
\n",
"
Total
\n",
"
\n",
"
\n",
"
classe
\n",
"
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
1
\n",
"
123
\n",
"
200
\n",
"
323
\n",
"
\n",
"
\n",
"
2
\n",
"
158
\n",
"
119
\n",
"
277
\n",
"
\n",
"
\n",
"
3
\n",
"
528
\n",
"
181
\n",
"
709
\n",
"
\n",
"
\n",
"
Total
\n",
"
809
\n",
"
500
\n",
"
1309
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
"survivant 0 1 Total\n",
"classe \n",
"1 123 200 323\n",
"2 158 119 277\n",
"3 528 181 709\n",
"Total 809 500 1309"
]
},
"execution_count": 50,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"pt = df.pivot_table(index='classe', columns='survivant', aggfunc='size', fill_value=0)\n",
"pt['Total'] = pt.sum(axis=1) # Total par ligne\n",
"pt.loc['Total'] = pt.sum(axis=0) # Total par colonne\n",
"pt"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On peut également afficher le **pourcentage de survie pour chaque classe**, mais il faut choisir entre 3 options :\n",
"\n",
"- le pourcentage de survivants dans chaque classe \n",
"- la répartition de chaque classe parmi les survivants/non-survivants\n",
"- le pourcentage de couple (classe, survie) parmi le total\n"
]
},
{
"cell_type": "code",
"execution_count": 51,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:38.470240Z",
"iopub.status.busy": "2026-10-30T13:10:38.470033Z",
"iopub.status.idle": "2026-10-30T13:10:38.528859Z",
"shell.execute_reply": "2026-10-30T13:10:38.528450Z"
}
},
"outputs": [
{
"data": {
"text/markdown": [
"**Pourcentage par ligne : combien de survivants parmi chaque classe**"
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
survivant
\n",
"
0
\n",
"
1
\n",
"
\n",
"
\n",
"
classe
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
1
\n",
"
0.380805
\n",
"
0.619195
\n",
"
\n",
"
\n",
"
2
\n",
"
0.570397
\n",
"
0.429603
\n",
"
\n",
"
\n",
"
3
\n",
"
0.744711
\n",
"
0.255289
\n",
"
\n",
"
\n",
"
Total
\n",
"
0.618029
\n",
"
0.381971
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
"survivant 0 1\n",
"classe \n",
"1 0.380805 0.619195\n",
"2 0.570397 0.429603\n",
"3 0.744711 0.255289\n",
"Total 0.618029 0.381971"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/markdown": [
"**Pourcentage par colonne : proportion de chaque classe parmi les survivants et les non survivants**"
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
survivant
\n",
"
0
\n",
"
1
\n",
"
Total
\n",
"
\n",
"
\n",
"
classe
\n",
"
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
1
\n",
"
0.152040
\n",
"
0.400
\n",
"
0.246753
\n",
"
\n",
"
\n",
"
2
\n",
"
0.195303
\n",
"
0.238
\n",
"
0.211612
\n",
"
\n",
"
\n",
"
3
\n",
"
0.652658
\n",
"
0.362
\n",
"
0.541635
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
"survivant 0 1 Total\n",
"classe \n",
"1 0.152040 0.400 0.246753\n",
"2 0.195303 0.238 0.211612\n",
"3 0.652658 0.362 0.541635"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/markdown": [
"**Pourcentage total : proportion de chaque combinaison classe-survivant par rapport à l'ensemble des passagers**"
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
survivant
\n",
"
0
\n",
"
1
\n",
"
Total
\n",
"
\n",
"
\n",
"
classe
\n",
"
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
1
\n",
"
0.093965
\n",
"
0.152788
\n",
"
0.246753
\n",
"
\n",
"
\n",
"
2
\n",
"
0.120703
\n",
"
0.090909
\n",
"
0.211612
\n",
"
\n",
"
\n",
"
3
\n",
"
0.403361
\n",
"
0.138273
\n",
"
0.541635
\n",
"
\n",
"
\n",
"
Total
\n",
"
0.618029
\n",
"
0.381971
\n",
"
1.000000
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
"survivant 0 1 Total\n",
"classe \n",
"1 0.093965 0.152788 0.246753\n",
"2 0.120703 0.090909 0.211612\n",
"3 0.403361 0.138273 0.541635\n",
"Total 0.618029 0.381971 1.000000"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Nomenclature :\n",
"# --------------------------\n",
"# tc = tableau croisé\n",
"# cl_sur = classe survivant\n",
"# norm = normalizé\n",
"\n",
"# Pourcentage par ligne\n",
"display(Markdown(\"**Pourcentage par ligne : combien de survivants parmi chaque classe**\"))\n",
"tc_cl_sur_norm_lignes= pd.crosstab(df['classe'], df['survivant'], margins=True, margins_name='Total', normalize='index')\n",
"display(tc_cl_sur_norm_lignes)\n",
"\n",
"# Pourcentage par colonne\n",
"display(Markdown(\"**Pourcentage par colonne : proportion de chaque classe parmi les survivants et les non survivants**\"))\n",
"tc_cl_sur_norm_colonnes = pd.crosstab(df['classe'], df['survivant'], margins=True, margins_name='Total', normalize='columns') \n",
"display(tc_cl_sur_norm_colonnes)\n",
"\n",
"# Pourcentage total\n",
"display(Markdown(\"**Pourcentage total : proportion de chaque combinaison classe-survivant par rapport à l'ensemble des passagers**\"))\n",
"tc_cl_sur_norm_all = pd.crosstab(df['classe'], df['survivant'], margins=True, margins_name='Total', normalize='all')\n",
"display(tc_cl_sur_norm_all)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Le tableau \"par ligne\" indique que les passagers de 1re classe avaient un taux de survie d'environ 62 %, tandis que ceux de 2e classe avaient un taux de survie d'environ 43 %, et ceux de 3e classe seulement environ 26 %. Cela confirme l'idée que les passagers de première classe avaient un avantage significatif en termes de survie par rapport aux autres classes.\n",
"\n",
"Le tableau \"par colonne\" montre que parmi les survivants, une proportion plus élevée venait de la 1re classe (40 %), suivi de la 3e classe (36 %) et enfin de la 2e classe (24 %). La difficulté avec ce tableau est qu'il est influencé par la répartition initiale des passagers dans chaque classe (cf. paradoxe de Simpson).\n",
"\n",
"Le tableau \"total\" montre la répartition globale des passagers par classe, indépendamment de leur survie. Cela donne une idée aussi de la composition initiale des passagers à bord du Titanic."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On peut ensuite afficher ces informations sous forme de graphique :"
]
},
{
"cell_type": "code",
"execution_count": 52,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:38.530945Z",
"iopub.status.busy": "2026-10-30T13:10:38.530723Z",
"iopub.status.idle": "2026-10-30T13:10:38.681855Z",
"shell.execute_reply": "2026-10-30T13:10:38.681436Z"
}
},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Avec Pandas et dans un graphique en barres empilées\n",
"# --------------------------------------------------------------\n",
"tc_cl_sur_norm_lignes.plot(kind='bar', stacked=True, title='Pourcentage par ligne : proportion de survivants dans chaque classe');\n",
"\n",
"# NB Normalisation par ligne : chaque barre a la même hauteur (1 ou 100%) \n",
"# et les segments représentent la proportion de survivants dans chaque classe.\n",
"\n",
"# On ne peut pas faire cela avec plotly.express directement."
]
},
{
"cell_type": "code",
"execution_count": 53,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:38.683615Z",
"iopub.status.busy": "2026-10-30T13:10:38.683423Z",
"iopub.status.idle": "2026-10-30T13:10:38.835763Z",
"shell.execute_reply": "2026-10-30T13:10:38.835261Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Pourcentage total sans les marges\n",
"tc_cl_sur_norm_all = pd.crosstab(df['classe'], df['survivant'], normalize='all')\n",
"display(tc_cl_sur_norm_all)\n",
"tc_cl_sur_norm_all.plot(kind='bar', stacked=True, title='Pourcentage total : proportion de chaque combinaison classe-survivant par rapport à l\\'ensemble des passagers');"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Quel graphe est le plus pertinent pour comparer les survivants et les non-survivants en fonction de la classe ?"
]
},
{
"cell_type": "code",
"execution_count": 54,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:38.837607Z",
"iopub.status.busy": "2026-10-30T13:10:38.837451Z",
"iopub.status.idle": "2026-10-30T13:10:39.111699Z",
"shell.execute_reply": "2026-10-30T13:10:39.111214Z"
}
},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Quel graphe est le plus pertinent pour comparer les survivants et les non-survivants en fonction de la classe ?\n",
"# ===============================================================================================================\n",
"\n",
"# sns.countplot(data=df, x='survivant')\n",
"\n",
"# plt.figure();\n",
"# Chaque statut (survivant / décédé) est divisé en trois barres (classe 1, 2, 3)\n",
"sns.countplot(data=df, x='survivant', hue='classe');\n",
"\n",
"plt.figure();\n",
"# Chaque classe est divisée en deux barres (survivant et non-survivant)\n",
"sns.countplot(data=df, x='classe', hue='survivant');\n",
"\n",
"# plt.figure();\n",
"# sns.histplot(data=df, x='classe', hue='survivant', multiple='stack'); # idem empilées"
]
},
{
"cell_type": "code",
"execution_count": 55,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:39.113654Z",
"iopub.status.busy": "2026-10-30T13:10:39.113494Z",
"iopub.status.idle": "2026-10-30T13:10:39.187712Z",
"shell.execute_reply": "2026-10-30T13:10:39.187205Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"plot_proportions(df, 'port_depart', 'survivant')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"D'après ces graphiques, environ 70 % des passagers ont embarqué depuis S(outhampton), 20 % depuis C(herbourg) et 10 % depuis Q(ueenstown). Cependant, le taux de survie le plus élevé, soit 55 %, est obtenu pour les passagers ayant embarqué à Cherbourg, puis environ 35 % à Queenstown et 30 % à Southampton. "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Bonus : nouvelle colonne femmes / hommes / enfants"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Il pourrait être utile de connaître la répartition entre les hommes, les femmes et les enfants. Pour cela, nous allons créer une nouvelle colonne `personne` qui prendra les valeurs `homme`, `femme` ou `enfant` en fonction de l'âge. \n",
"Nous allons ensuite afficher le taux de survie pour chaque catégorie de personnes.\n",
"\n",
"\n",
"Cependant, la priorité pour l’évacuation selon le fameux principe « les femmes et les enfants d’abord » ne définissait pas exactement un âge officiel pour les enfants. Les archives et témoignages historiques montrent que les enfants étaient généralement considérés comme ceux de moins de 14 ans. Certains témoignages mentionnent que des adolescents (12-15 ans) étaient parfois traités comme des adultes selon le contexte et la cabine. Les décisions étaient aussi un peu arbitraires : la distinction reposait souvent sur l’apparence et la taille plutôt que sur un âge strict. Donc, pour les analyses de survie ou des modèles statistiques sur le Titanic, beaucoup de chercheurs prennent <15 ans comme enfant pour appliquer ce critère."
]
},
{
"cell_type": "code",
"execution_count": 69,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:43.340354Z",
"iopub.status.busy": "2026-10-30T13:10:43.340166Z",
"iopub.status.idle": "2026-10-30T13:10:43.557641Z",
"shell.execute_reply": "2026-10-30T13:10:43.557063Z"
}
},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAk0AAAHkCAYAAAAjARvjAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/P9b71AAAACXBIWXMAAA9hAAAPYQGoP6dpAAA4z0lEQVR4nO3de1hVdd7//9fmsMEDiAeElFDzgJKMqamYpnkaS2XykJa3p9TUPNaYOo6D3uWxUsEU1Ci0zGyUxrEsycq6se4JZu7RGhtHR3EazC8qlBxE5eT+/eGPPe5E+4jAQnw+rquLWGvttd+Li6t5zlqLtW0Oh8MhAAAA3JCb1QMAAADcDogmAAAAA0QTAACAAaIJAADAANEEAABggGgCAAAwQDQBAAAYIJoAAAAMEE0/Y/To0Ro9erTVYwAAAIt5WD1AVZeenm71CAAAoArgTBMAAIABogkAAMAA0QQAAGCAaAIAADBANAEAABggmgAAAAwQTQAAAAaIJgAAAANEEwAAgAGiCQAAwADRBAAAYIBoAgAAMEA0AQAAGCCaAAAADBBNAAAABogmAAAAA0QTAACAAaIJAADAANEEAABggGgCAFjussNh9QhOVWkWVC0eVg8AAICbzaZ9p79SVkGOpXP42X3VJ7CrpTOg6iKaAABVQlZBjjLzz1k9BnBdXJ4DAAAwQDQBAAAYIJoAAAAMEE0AAAAGiCYAAAADRBMAAIABogkAAMAA0QQAAGCAaAIAADBANAEAABggmgAAAAwQTQAAAAaIJgAAAANEEwAAgAGiCQAAwADRBAAAYIBoAgAAMEA0AQAAGCCaAAAADBBNAAAABogmAAAAA0QTAACAAaIJAADAANEEAABggGgCAAAwQDQBAAAYIJoAAAAMEE0AAAAGiCYAAAADRBMAAIABogkAAMAA0QQAAGCAaAIAADBANAEAABggmgAAAAwQTQAAAAaIJgAAAANEEwAAgAGiCQAAwECViqa8vDz16NFDISEhOnTokMu6pKQkDR48WGFhYerXr5/efvvtUvcRHx+v3r17KywsTMOGDVNKSkpljA4AAKq5KhVN69evV3Fx8TXLDx48qGnTpik0NFSvvfaahgwZoqVLlyohIcFlu/j4eEVHR2vUqFGKi4tTkyZNNGnSJB09erSyDgEAAFRTVSaaUlNTtW3bNs2cOfOadbGxsQoNDdXy5csVHh6uadOm6bHHHtMrr7yiy5cvS5IKCgq0YcMGjR07VhMnTlTXrl21cuVKBQUFaePGjZV9OAAAoJqpMtG0bNkyPfHEE2rWrJnL8oKCAiUnJ2vgwIEuyyMiIpSRkaHDhw9Lkg4cOKDc3FwNGjTIuY27u7sGDBigpKQkORyOij8IAABQbVWJaProo4905MgRTZ8+/Zp1aWlpKiws1D333OOyvEWLFpKunKG6+utPt2vevLny8vJ05syZihgdAADcITysHuDixYt68cUXNXv2bNWuXfua9dnZ2ZIkX19fl+Ul35esz8nJkd1ul7e3t8t2derUkSRlZWUpMDCw1Bn69Olz3fnS09N11113GR4NAACoriw/07RhwwbVr19fQ4cOveF2NpvtZ5eXtk3JZbnrvR4AAMCEpWeaTp06pU2bNik2Nlbnz5+XJF24cMH5NS8vz3mmqOSMUomcnBxJ/znj5Ovrq/z8fOXn58vLy+ua7Ur2U5p9+/Zdd92NzkIBAIA7h6XR9P3336uwsFCTJ0++Zt3YsWPVrl07bd26VZ6enjpx4oR69OjhXH/8+HFJV+5ZuvpramqqQkNDndulpqaqVq1aCggIqMhDAQAA1Zyl0dSmTRtt2bLFZdk//vEPrVixQi+88ILCwsJkt9sVHh6uxMREPfnkk87tPvjgA/n7+zsDqUOHDvLx8dGePXucy4qLi5WYmKiePXtyeQ4AANwSS6PJ19dXXbp0KXXdvffeq3vvvVeSNH36dI0ePVqRkZGKiIjQgQMHlJCQoMWLF8vN7cptWXa7XVOnTlV0dLTq1aun0NBQJSQk6OTJk4qKiqq0YwIAANWT5X89Z6J9+/Zav369oqKitGvXLgUGBioyMlLDhw932W7ChAlyOBx66623lJmZqVatWikuLk4hISEWTQ4AAKoLm4OnPt5QyY3gN7pZHABw6/6QtleZ+ecsnaGBV10NC+5v6Qyouix/5AAAAMDtgGgCAAAwQDQBAAAYIJoAAAAMEE0AAAAGiCYAAAADRBMAAIABogkAAMAA0QQAAGCAaAIAADBANAEAABggmgAAAAwQTQAAAAaIJgAAAANEEwAAgAGiCQAAwADRBAAAYIBoAgAAMEA0AQAAGCCaAAAADBBNAAAABogmAAAAA0QTAACAAaIJAADAANEEAABggGgCAAAwQDQBAAAYIJoAAAAMEE0AAAAGiCYAAAADRBMAAIABogkAAMAA0QQAAGCAaAIAADBANAEAABggmgAAAAwQTQAAAAaIJgAAAANEEwAAgAGiCQAAwADRBAAAYIBoAgAAMEA0AQAAGCCaAAAADBBNAAAABogmAAAAA0QTAACAAaIJAADAANEEAABggGgCAAAwQDQBAAAYIJoAAAAMEE0AAAAGiCYAAAADRBMAAIABogkAAMAA0QQAAGCAaAIAADBANAEAABggmgAAAAwQTQAAAAaIJgAAAANEEwAAgAGiCQAAwADRBAAAYIBoAgAAMEA0AQAAGCCaAAAADBBNAAAABogmAAAAA0QTAACAAaIJAADAANEEAABggGgCAAAwYHk0ffHFFxo9erTCw8PVtm1b9enTRytWrFBubq7LdklJSRo8eLDCwsLUr18/vf3226XuLz4+Xr1791ZYWJiGDRumlJSUyjgMAABQzVkeTdnZ2Wrfvr2WLFmi+Ph4jR8/Xrt27dIzzzzj3ObgwYOaNm2aQkND9dprr2nIkCFaunSpEhISXPYVHx+v6OhojRo1SnFxcWrSpIkmTZqko0ePVvZhAQCAasbmcDgcVg/xUzt27NDChQu1f/9+BQQE6KmnnlJ2drZLJC1cuFCff/659u/fLzc3NxUUFOiBBx7QiBEjNG/ePElScXGxIiIiFBISoujo6DLN0qdPH0nSvn37bv3AAADX9Ye0vcrMP2fpDA286mpYcH9LZ0DVZfmZptL4+flJkoqKilRQUKDk5GQNHDjQZZuIiAhlZGTo8OHDkqQDBw4oNzdXgwYNcm7j7u6uAQMGKCkpSVWwDQEAwG2kykRTcXGx8vPz9fe//12xsbHq1auXGjdurLS0NBUWFuqee+5x2b5FixaSpNTUVJevP92uefPmysvL05kzZyrhKAAAQHXlYfUAJXr16uUMmwcffFBRUVGSrtzzJEm+vr4u25d8X7I+JydHdrtd3t7eLtvVqVNHkpSVlaXAwMBS37vkElxp0tPTddddd93s4QAAgGqmykRTXFycLly4oOPHj2v9+vV6+umntXnzZud6m81W6uuuXl7aNiWX5a73egAAABNVJppat24tSerQoYNCQ0M1bNgwffLJJ87LcCVnlErk5ORI+s8ZJ19fX+Xn5ys/P19eXl7XbFdyxqk0N7rJ+0ZnoQAAwJ2jytzTdLU2bdrI3d1daWlpCg4Olqenp06cOOGyzfHjxyVduWfp6q8l9zaVSE1NVa1atRQQEFAJkwMAgOqqSkbTwYMHVVxcrKCgINntdoWHhysxMdFlmw8++ED+/v4KDQ2VdOUMlY+Pj/bs2ePcpri4WImJierZsyeX5wAAwC2x/PLcjBkz1LZtW4WEhMjb21tHjhzR66+/rpCQEPXt21eSNH36dI0ePVqRkZGKiIjQgQMHlJCQoMWLF8vN7Ur32e12TZ06VdHR0apXr55CQ0OVkJCgkydPOm8qBwAAKCvLo+kXv/iF9uzZo7i4ODkcDjVu3FgjRozQxIkTZbfbJUnt27fX+vXrFRUVpV27dikwMFCRkZEaPny4y74mTJggh8Oht956S5mZmWrVqpXi4uIUEhJixaEBAIBqpEo+Ebwq4YngAFA5eCI4qroqeU8TAABAVUM0AQAAGCCaAAAADBBNAAAABogmAAAAA0QTAACAAaIJAADAANEEAABggGgCAAAwQDQBAAAYIJoAAAAMEE0AAAAGiCYAAAADRBMAAIABogkAAMAA0QQAAGCAaAIAADBANAEAABggmgAAAAwQTQAAAAaIJgAAAANEEwAAgAGiCQAAwADRBAAAYKBM0RQTE6MzZ86Uuu7s2bOKiYm5paEAAACqmjJFU2xs7A2jKTY29paGAgAAqGrKFE0Oh+O66y5cuCAPD48yDwQAAFAVGdfNkSNHdOTIEef3SUlJOnHihMs2ly5d0u7duxUcHFx+EwIAAFQBxtH06aefOu9Vstls170E5+3treXLl5fPdAAAAFWEcTSNGDFCDz30kBwOh4YPH64VK1aoZcuWLtvY7XYFBwfL29u73AcFAACwknE0NWzYUA0bNpQkbdmyRffee69q1apVYYMBAABUJWW6Y7tz587lPQcAAKhkFy9eVI0aNawe47ZR5odbvvfee5o0aZIGDhyoPn36uPzTt2/f8pwRAIA7zrp16xQSEqLDhw9rxowZ6tChgzp27Kg5c+boxx9/dNl2z549evzxx3Xfffepffv2mjhxog4fPuyyzfz589W+fXsdPXpUEyZMUPv27fXkk09Kkg4fPqwpU6aoa9euatu2rbp3767Jkyfr9OnTztfn5+dr9erV6t27t9q2basHH3xQL7zwgnJyclzep3fv3poyZYr279+vIUOG6Be/+IUefvhhvfvuuy7b7dy5UyEhIUpOTtZ///d/q0uXLurSpYtmzJhR6mONTI6xopXpTFNcXJyioqLUokULtW7dWna7vbznAgAAkmbMmKGHH35YTzzxhI4fP65XXnlFqamp2rFjhzw9PbVx40atWbNGQ4cO1dSpU1VYWKj4+HiNGjVKCQkJatGihXNfhYWFmjp1qp544glNmjRJxcXFunDhgsaPH6+goCAtWrRIDRo0UEZGhlJSUpSXlyfpyqOGpk2bpuTkZE2ePFn333+/jh49qnXr1unrr7/W9u3bXVrgyJEjeumllzRp0iQ1aNBACQkJ+t3vfqcmTZqoU6dOLscXGRmphx56SKtXr1Z6erpWrlypuXPnasuWLc5tbuYYK1KZomnHjh0aNWqUFi5cWN7zAACAq/Tr10/z5s2TJHXv3l3169fXnDlzlJiYqE6dOmndunUaPXq0IiMjna954IEH1L9/f8XExGjNmjXO5YWFhZo+fbqGDRvmXPbtt98qKytLy5Ytc7lSNGDAAOe/f/nll/ryyy81d+5cPfXUU5Kkbt26KTAwUL/+9a+1a9cujRgxwrn9uXPn9M4776hRo0aSpE6dOik5OVm7d+++JpoefPBBl9mzs7O1cuVKZWRkyN/fX+np6Td1jBWpTJfnMjMzuQQHAEAliIiIcPn+kUcekYeHh1JSUvTll1+qqKhIjz76qIqKipz/eHl5qVOnTvrzn/98zf769+/v8n2TJk1Up04drVq1Su+8846OHz9+zWuSk5MlSUOHDr1mlpo1a+qrr75yWd6mTRtnMEmSl5eXmjZtqv/3//7fNfvu3bu3y/chISGS5Ny2LMdYUcp0punee+/VyZMn1bVr1/KeBwAAXMXf39/lew8PD/n5+SkrK0uZmZmSpMcee6zU17q5uZ4bqVGjhmrXru2yzMfHR2+99ZY2btyo6OhoZWdny9/fXyNGjNDUqVPl6emprKwseXh4qF69ei6vtdlsatCggbKyslyW+/n5XTOL3W5Xfn7+Nct/um3JZb5Lly5J0k0fY0UqUzTNnz9fc+fOVWhoqNq2bVveMwEAgP9fRkaGAgICnN8XFRUpKytLfn5+qlu3riRp7dq1Lmd2rsdms5W6PCQkRNHR0XI4HDp69Kh27typ2NhYeXt7a/LkyfLz81NRUZF+/PFHl3ByOBzKzMxUWFjYLR7l9d3sMVakMkXTggULlJWVpeHDh6tBgwbXVKLNZtP7779fHvMBAHBH2717t8sJisTERBUVFalz587q2LGjPDw8lJaWds1lt7Kw2Wxq3bq1FixYoD/+8Y/6+9//Lknq2rWrXn/9db3//vvOv7iTpL179+rChQsVeuWpe/fu5XqMt6JM0eTn51fqqTcAAFC+PvnkE7m7u6tbt246duyYXnnlFbVu3VqPPPKI7Ha7Zs2apTVr1ujkyZPq0aOHfH19lZmZqUOHDqlGjRqaNWvWDff/+eefa9u2berbt6/uvvtuORwOffzxx8rJyVG3bt0kXbnpu3v37lq1apXOnz+vDh066OjRo1q7dq1CQ0P16KOPVtjxBwUF3fIxlpcyRdNbb71V3nMAAIBSrFu3TuvWrdM777wjm82m3r17a8GCBc57f6ZMmaLmzZtry5Yt+vDDD1VQUCB/f3+1bdtWI0eO/Nn9N2nSRL6+vnr99dd19uxZeXp6qlmzZnrxxRc1ZMgQSVfOQK1fv17r1q3Tzp07tXHjRvn5+enRRx/V7NmzK/zRQ7d6jOXF5nA4HJX2brehPn36SJL27dtn8SQAUL39IW2vMvPPWTpDA6+6GhZs7SWgEuvWrVNMTIy++uqra27AhjXKdKbpL3/5y89u89PnMAAAANzOyhRNY8aMue4d+CX+8Y9/lGkgAACAqqhM0XT1o81LnDt3Tvv27dOBAwe0aNGiWx4MAIA72cyZMzVz5kyrx8BVyhRNnTt3LnV5//79tWjRIn3xxRfq0aPHLQ0GAABQlZT7YzT79eunPXv2lPduAQAALFXu0ZSTk6OCgoLy3i0AAIClynR5rrQP3CsoKNDRo0e1evVqtWvX7pYHAwAAqErKFE29e/cu9a/nHA6HmjVrxo3gAACg2ilTNC1fvvyaaPLy8lLjxo0VFhZWqZ84DAAAUBnKFE1Dhw4t7zkAAEAV8sYbb+iNN97QmTNn1KtXL61fv97qkSxXpmgqcf78eX399dfKyspS3bp11a5dO9WuXbu8ZgMA4LbncDh+9oHQVe29U1NTtWLFCk2ePFm9evVS3bp1K2C620+Zoyk+Pl4xMTG6dOmSSj6+ruSThsePH19uAwIAcDuz2WxKSc1R7sXiSn1fnxru6tLct0yv/de//iVJGjFihO6+++7yHOu2VqZo2rVrl1auXKkePXpoyJAhatiwoc6ePatdu3bp5ZdfVt26dTV48OByHhUAgNtT7sViZV0osnoMI/Pnz9cf//hHSVLfvn0lSStWrFDfvn0VFRWlTz/9VFlZWWrVqpVmz56t7t27O187ZswY1axZUwMGDFBMTIzOnj2r8PBwvfTSS8rLy9OiRYt04MABNWrUSAsXLlR4eLjztb1799ZDDz2koKAgvfHGG8rNzVW/fv20ZMkSnThxQi+88IKOHDmi5s2ba/ny5QoJCXG+1uFwaNOmTdqxY4dOnTqlgIAAjRkzRk8++WS5/mzKFE1vvPGGBg0apFWrVrksf+SRRzRnzhy9+eabRBMAALehadOmqVmzZoqKilJMTIz8/f0VFBSk8ePH64cfftCzzz6rgIAAvf/++5oyZYp27tzpEjCHDx9Wdna25s+fr9zcXC1dulSRkZFKT0/X4MGDNX78eL366quaNWuWPv/8c9WqVcv52n379ikkJERLlizRyZMn9eKLL8put+ubb77Rk08+qQYNGmjVqlV65plntGfPHucfni1btkwJCQl6+umn1a5dOx04cECrVq2Sl5eXRo4cWW4/mzJF04kTJzR79uxS1/3qV7/SjBkzbmkoAABgjeDgYDVp0kSS1KZNGwUFBekPf/iDjhw5ovfee08tWrSQJD344IP67rvvtH79er3yyivO158/f14bNmxw3gd19OhRbdq0Sc8//7wzYBo2bKiIiAh99dVXzrNZ0pVLmTExMbLb7ZKkP//5z0pISNBrr73m/Hi2y5cv6+mnn9Y///lPtW7dWmlpadq6dateeOEFPf7445KkBx54QBcuXFBsbKwef/zxcvur/jLtxdvbW9nZ2aWuy87Olre39y0NBQAAqo7//d//VatWrdS0aVMVFRU5/+natasOHTrksm3r1q1dbhxv2rSppCsh89Nlp0+fdnnt/fff7wymku3c3NxcLuOVvDY9PV2S9Kc//UmS9Mtf/vKa2TIyMpzblYcynWnq2LGjYmJi1LlzZwUEBDiXZ2RkKDY2Vvfff3+5DQgAAKx17tw5HT58WPfee+8169zd3V2+9/V1vfnc09NTkuTj4+NcVhJG+fn5P/tab29vl5Aq2V/Ja8+dOyeHw+ESVldLT09X48aNr39wN6FM0TR79mw9/vjj+uUvf6muXbvK399fGRkZSk5OloeHh2JiYsplOAAAYL06deooJCREy5Yts3qUa9SpU0c2m03btm1zBtXVmjVrVm7vVaZoatmypd59913FxMQoJSVFWVlZ8vPzU58+fTRjxoxyHRAAAFjrgQceUFJSkho2bOhyhakq6Nq1qyQpKytLvXv3rtD3KlM0FRYWKjAwUFFRUdesu3DhggoLC0utPQAAcPsZPHiwfv/732vs2LGaMGGCmjZtqtzcXB0+fFiFhYV67rnnLJutWbNmGjVqlObNm6eJEyeqXbt2Kiws1HfffaeUlJRyfZJ5maJp4cKFKigoKDWaFi5cKG9v7yp5Cg8AACv41HD/+Y2q8Hva7XZt2bJF69at08aNG5WRkSE/Pz+Fhobqv/7rv8rtfcoqMjJSzZo10/bt2xUbG6uaNWuqWbNmeuSRR8r1fWyOksd534RevXrpueee06BBg65Z9+GHH2r16tX67LPPymVAq/Xp00fSlWdHAAAqzh/S9ioz/5ylMzTwqqthwf3LdZ+348eooHRleuRAZmam/P39S13XoEEDZWZm3tJQAABUF1ZGC8FUvsoUTb6+vkpLSyt1XVpamsvTPQEAAKqDMkVTly5d9OqrryorK8tleVZWluLi4q77rAQAAIDbVZluBJ8xY4Yee+wx9e/fX4888ogCAgJ0+vRpffTRRyoqKtLMmTPLe04AAABLlSma7rnnHm3btk0rVqxQQkKCiouL5e7urk6dOmn+/Pm65557yntOAAAAS5UpmqQrny3z5ptv6tKlS8rOzpafn5+8vLzKczYAAIAqo8zRVMLb25sP6AUAANVemW4EBwAAuNMQTQAAAAaIJgAAAAOWR1NiYqKmTZumnj176r777lNERIS2bdumy5cvu2yXlJSkwYMHKywsTP369dPbb79d6v7i4+PVu3dvhYWFadiwYUpJSamMwwAAAKXYvXu3fvnLX+ree+/Vo48+Wq77zsnJ0bp163T8+PFy3e/1WB5Nmzdvlt1u17x587Rx40b17dtXy5Yt08qVK53bHDx4UNOmTVNoaKhee+01DRkyREuXLlVCQoLLvuLj4xUdHa1Ro0YpLi5OTZo00aRJk3T06NHKPiwAACRJDsfln9+oGr63JJ0/f14LFixQx44dtWXLFr388svluv+cnBzFxMRUWjTd8l/P3aqNGzeqXr16zu/Dw8N14cIFvf322/r1r38tu92u2NhYhYaGavny5c5t0tPT9corr2jYsGFyc3NTQUGBNmzYoLFjx2rixImSpM6dOysiIkIbN25UdHS0JccHALiz2Wxuyvtmry7nVe6HEbvVqqta7cr3w4dv1smTJ1VQUKBf/epX6tixo6WzlAfLo+nqYCrRpk0b5efnKysrS35+fkpOTtacOXNctomIiNCOHTt0+PBhtW3bVgcOHFBubq4GDRrk3Mbd3V0DBgzQpk2b+KRnAIBlLuedU3FOhtVj3JSDBw8qOjpaf/vb3+Tu7q6HHnpICxYsUP369fX999+rT58+evnll/XNN99o9+7d8vLyUkREhJ577jl5eHho3bp1iomJkSQ9+eSTkq58osjMmTO1atUqJSUl6fvvv1ft2rWdD8du2LCh8/3HjBmjmjVravDgwVqzZo3Onj2rsLAwLV26VMHBwc4ZJOmZZ55xvm7fvn0KCgqqkJ+J5ZfnSvPXv/5Vfn5+ql+/vtLS0lRYWHjNU8ZbtGghSUpNTXX5+tPtmjdvrry8PJ05c6YSJgcA4PZ38OBBjRkzRj4+PoqOjtaSJUt06NAhTZ061WW7NWvWyM3NTWvWrNHjjz+uTZs2OW+dGT58uFasWCFJWrRokbZv367hw4dLkn744QdNmTJFr776qn73u9/p1KlTGjNmjIqKilz2/49//EObNm3SnDlztGLFCn333XeaO3euJKlhw4bOKJs9e7a2b9+u7du3u4RXebP8TNNPHTp0SDt37tT06dPl7u6u7OxsSZKvr6/LdiXfl6zPycmR3W6/5kGbderUkXTlw4QDAwNLfc+SUi1Nenq67rrrrrIdDAAAt6HVq1erbdu2iomJcV6ladmypSIiIpSUlKTmzZtLkn7xi18oMjJSktStWzf96U9/0t69ezVy5EgFBgaqZcuWkq6c6Ljvvvuc+y+JKUkqLi5W+/bt1aNHDyUnJ6t79+7Odbm5udq1a5fzqlRubq4iIyN1+vRpBQYGqk2bNpKkJk2auOy/olSpM00ZGRmaNWuWwsLCNGnSJJd117u0dvXy0rZxOBw3fD0AAPiPixcv6sCBA3r44YdVXFysoqIiFRUVqVmzZvL399ehQ4ec214dONKVODp9+vTPvkdSUpKeeOIJdezYUaGhoerRo4ck6bvvvnPZrnXr1i638ZTEmsl7VIQqc6YpNzdXkyZNkre3tzZs2CBPT09J/zlTVHJGqUROTo6k/5xx8vX1VX5+vvLz810+A69ku5L9lGbfvn3XXXejs1AAAFQ3OTk5Ki4u1ooVK1zOCJVIT093/ruPj4/LOk9PTxUUFNxw/3/72980bdo09enTR5MmTVL9+vVls9k0YsQI5efnu2z706tMJW3w0+0qS5WIpvz8fE2dOlWZmZnavn276tat61wXHBwsT09PnThxwlmikpx/XlhSnSVfU1NTFRoa6twuNTVVtWrVUkBAQGUcCgAAtzUfHx/ZbDZNmTJFffv2vWb91f8bXRaffvqpateu7bwfSpJOnTp1S/usLJZHU1FRkZ555hkdOXJEW7duVePGjV3W2+12hYeHKzEx0Xn3vSR98MEH8vf3dwZShw4d5OPjoz179jiXFRcXKzExUT179uTyHAAABmrWrKn77rtPJ06cUFhYWKnbfP/992Xe/6VLl+Tp6enyv8u7d+8u074q+8yT5dG0ePFiff7555o7d64uXbqkr7/+2rmuRYsWql27tqZPn67Ro0crMjJSEREROnDggBISErR48WJnpdrtdk2dOlXR0dGqV6+eQkNDlZCQoJMnTyoqKsqiowMA4PYzb948jRs3Ts8++6wGDhwoX19fnT59Wn/60580dOjQa05w3Ixu3brpzTff1JIlS9SvXz8dPHhQ7733Xpn25e/vL19fX3344YcKCgqS3W5XSEiI7HZ7mee7Ecuj6csvv5QklyeAl9iyZYu6dOmi9u3ba/369YqKitKuXbsUGBioyMhI558ulpgwYYIcDofeeustZWZmqlWrVoqLi1NISEilHAsAAKVxq3Vrl7Qq+z07dOigbdu2ad26dfrtb3+rwsJCBQYGKjw8XE2aNLnm0QA3o2fPnpozZ462bt2qnTt3qkOHDnr11VfVv//NP4jTzc1Ny5cvV3R0tJ588kkVFBRU6HOabI6SPy9DqUpuBL/RzeIAgFv3h7S9ysyv3Kdm/1QDr7oaFly+T9F2OC7LZrPmj9WtfO/qiJ8kAAAVyMpoIZjKFz9NAAAAA0QTAACAAaIJAADAANEEAABggGgCAAAwQDQBQCWqSk95qUqzALcDyx9uCQB3EpvNppTUHOVeLLZ0Dp8a7urS3PfnNwTgRDQBQCXLvVisrAtlf6IyAGtweQ4AALiYP3++Bg0aZPUYVQ7RBAAAYIBoAgCgAl228IZ7K9+7OuKeJgAAKpCbzaZ9p79SVkFOpb6vn91XfQK73tI+UlJStGLFCn333Xdq0aKFnn/+ebVt21aSlJ+fr+joaH344Yc6d+6cmjVrpsmTJysiIsL5+vnz5+vbb7/Vb37zG7388sv697//rbZt2+qll16Sj4+Pnn/+ee3fv1/16tXT7NmzNWDAAOdrx4wZo5o1a2rAgAGKiYnR2bNnFR4erpdeekl5eXlatGiRDhw4oEaNGmnhwoUKDw93mX3nzp3avHmzvvvuO/n5+Wno0KGaOXOmPDzKnj5EEwAAFSyrIEeZ+eesHuOmZGRkaOnSpZo8ebJq166t1atXa8aMGfrkk0/k6empOXPmKCkpSc8++6xatmypPXv2aM6cOSouLtbgwYNd9rNq1SpNnTpVHh4eWrp0qebOnauaNWuqY8eOGj58uHbs2KG5c+eqXbt2aty4sfO1hw8fVnZ2tubPn6/c3FwtXbpUkZGRSk9P1+DBgzV+/Hi9+uqrmjVrlj7//HPVqlVLkrR582atXLlS48aN0/z585Wamqro6GgVFxdrzpw5Zf6ZEE0AAOAa2dnZ2rp1q1q2bClJ8vLy0vjx4/XNN9+odu3a+vjjj7Vo0SKNGjVKkvTggw/q7NmzWrt2rUs0ZWdna9u2bWrevLkk6ezZs1qyZIkmTZqk6dOnS5LCwsL0ySef6NNPP9W4ceOcrz1//rw2bNigunXrSpKOHj2qTZs26fnnn9fIkSMlSQ0bNlRERIS++uor9e3bV+fPn9fatWv11FNPafbs2ZKkbt26yd3dXS+//LImTpzo3N/N4p4mVKqqdH29Ks0CAFVNw4YNncEkyRk9Z86c0V//+ldJcrmcJkkDBw7UqVOnlJ6e7rKfktdKUtOmTSVJDzzwgHOZr6+v6tWrp9OnT7vsr3Xr1i6BU9prS5aVvPbgwYO6cOGCHn74YRUVFTn/CQ8P16VLl3Ts2LGb+jlcjTNNqFRWXdv/qfK41g8A1Zmvr+vDTz09PSVduZcpOztbHh4e15yxadCggaQrZ5fuuuuuG+7Hx8fHZbndbld+fr7RDFe/1m63O+eSpHPnrlwGHTJkSKnHdXXQ3SyiCZXudry2DwD4jzp16qioqEhZWVny8/NzLs/MzHSut0rJe8fExCgwMPCa9UFBQWXeN9EEAABuSseOHSVJiYmJznuLJGnPnj1q3Lix8yyTFTp06KAaNWro9OnT6tevX7num2gCAAA3pXXr1urfv79efPFFXbp0SS1atFBiYqK++OILvfTSS5bO5uPjo1mzZmnlypU6ffq0unTpIjc3N508eVL79u3TunXrVKNGjTLtm2gCAKCC+dkr/8ORK/o9V65cqejoaMXHxysrK0tNmzbVypUr9atf/apC39fEhAkTFBAQoM2bN2vr1q3y8PBQcHCwHnroIed9UWVhczj4E6Ib6dOnjyRp3759Fk9Sffwhba/l9zQ18KqrYcH9LZ0Bd65Pvz1n+Qf2+tX0UN+2Zfuz64pSXf/bcNnhkJvNVq77vB3euzrikQMAAFQgK6OFYCpfRBMAAIABogkAAMAA0QQAAGCAaAIAADBANAGwTFX6/L+qNAuAqonnNAGwDJ9FCOB2QjQBsBSfRQjgdsHlOQAAAANEEwAAgAGiCQAAwADRBAAAYIBoAgAAMEA0AQAAGCCaAAAADBBNAAAABogmAAAAA0QTAACAAaIJAADAANEEAABggGiqIA6Hw+oRnKrSLAAA3K48rB6gurLZbEpJzVHuxWJL5/Cp4a4uzX0tnQEAgOqAaKpAuReLlXWhyOoxAABAOeDyHAAAgAGiCQAAwADRBAAAYIBoAgAAMEA0AQAAGCCaAAAADBBNAAAABogmAAAAA0QTAACAAaIJAADAANEEAABggGgCAAAwQDQBAAAYIJoAAAAMEE0AAAAGiCYAAAADRBMAAIABogkAAMAA0QQAAGCAaAIAADBANAEAABggmgAAAAwQTQAAAAaIJgAAAANEEwAAgAGiCQAAwADRBAAAYIBoAgAAMEA0AQAAGLA8mv79739r0aJFevTRRxUaGqpBgwaVul1SUpIGDx6ssLAw9evXT2+//Xap28XHx6t3794KCwvTsGHDlJKSUpHjAwCAO4Tl0XTs2DElJSWpSZMmat68eanbHDx4UNOmTVNoaKhee+01DRkyREuXLlVCQoLLdvHx8YqOjtaoUaMUFxenJk2aaNKkSTp69GhlHAoAAKjGPKweoHfv3urbt68kaf78+fr222+v2SY2NlahoaFavny5JCk8PFzp6el65ZVXNGzYMLm5uamgoEAbNmzQ2LFjNXHiRElS586dFRERoY0bNyo6OrryDgoAAFQ7lp9pcnO78QgFBQVKTk7WwIEDXZZHREQoIyNDhw8fliQdOHBAubm5Lpf33N3dNWDAACUlJcnhcJT/8AAA4I5heTT9nLS0NBUWFuqee+5xWd6iRQtJUmpqqsvXn27XvHlz5eXl6cyZM5UwLQAAqK4svzz3c7KzsyVJvr6+LstLvi9Zn5OTI7vdLm9vb5ft6tSpI0nKyspSYGBgqe/Rp0+f675/enq67rrrrrINDwAAqo0qf6aphM1m+9nlpW1Tclnueq8HAAAwUeXPNJWcKSo5o1QiJydH0n/OOPn6+io/P1/5+fny8vK6ZruS/ZRm37591113o7NQAADgzlHlzzQFBwfL09NTJ06ccFl+/PhxSXI+pqDka8m9TSVSU1NVq1YtBQQEVMK0AACguqry0WS32xUeHq7ExESX5R988IH8/f0VGhoqSerQoYN8fHy0Z88e5zbFxcVKTExUz549uTwHAABuieWX5y5evKikpCRJ0qlTp3T+/Hl99NFHkq48Z6levXqaPn26Ro8ercjISEVEROjAgQNKSEjQ4sWLnY8ssNvtmjp1qqKjo1WvXj2FhoYqISFBJ0+eVFRUlGXHBwAAqgfLo+mHH37QM88847Ks5PstW7aoS5cuat++vdavX6+oqCjt2rVLgYGBioyM1PDhw11eN2HCBDkcDr311lvKzMxUq1atFBcXp5CQkEo7HgAAUD1ZHk1BQUFGH3PSs2dP9ezZ84bb2Gw2PfXUU3rqqafKazygXDgcjipzibgqzQIAtxPLowm4E9hsNqWk5ij3YrGlc/jUcFeX5r4/vyEA4BpEE1BJci8WK+tCkdVjAADKqMr/9RwAAEBVQDQBAAAYIJoAAAAMEE0AAAAGiCYAAAADRBMAAIABogkAAMAA0QQAAGCAaAIAADBANAEAABggmgAAAAwQTQAAAAaIJgAAAANEEwAAgAGiCQAAwADRBAAAYIBoAgAAMEA0AQAAGCCaAOAO5OVpk8Nx2eoxgNuKh9UDAAAqn93dTTabm/K+2avLeecsncWjQbBqtHrA0hkAE0QTANzBLuedU3FOhqUzuNWqa+n7A6a4PAcAAGCAaAIAADBANAEAABggmgAAAAwQTQAAAAaIJgAAAANEEwAAgAGiCQAAwADRBAAAYIBoAgAAMEA0AQAAGCCaAAAADBBNAAAABogmAAAAA0QTAACAAaIJAADAANEEAABggGgCAAAwQDQBAAAYIJoAAAAMEE0AAAAGiCYAAAADRBMAAIABogkAAMAA0QQAAGCAaAIAADBANAEAABggmgAAAAwQTQAAAAaIJgAAAANEEwAAgAGiCQAAwADRBAAAYIBoAgAAMEA0AQAAGCCagDuIl6dNDsdlq8cAgNuSh9UDAKg8dnc32Wxuyvtmry7nnbN0Fo8GwarR6gFLZwCAm0E0AXegy3nnVJyTYekMbrXqWvr+AHCzuDwHAABggGgCAAAwQDQBAAAYIJoAAAAMEE0AAAAGiCYAAAADRBMAAIABogkAAMAA0QQAAGCAaAIAADBANAEAABggmgAAAAxUu2j617/+pYkTJ+q+++5T165dtXTpUl26dMnqsQAAwG3Ow+oBylNOTo7GjRunRo0aae3atfrxxx+1YsUKZWVladWqVVaPZwkvT5scjsuy2apdHwMAUKmqVTT9/ve/V05Ojnbt2qV69epJktzd3TVnzhxNnTpVzZs3t3jCymd3d5PN5qa8b/bqct45S2fxaBCsGq0esHQGAADKqlpF0/79+9W1a1dnMElS//79tWDBAiUlJd2R0VTict45FedkWDqDW626lr4/AAC3olpds0lNTb0mjOx2u4KDg5WammrRVAAAoDqoVmeacnJy5Ovre81yX19fZWdnX/d1ffr0ue6677//Xu7u7jfc5nryiy7r8uWbflm5cnezye5hk6PgouSweBg3D9k8X9bF4nxdtngWN5ub1ru/XKnvye/DT9zBvw/8LvxENflduOuuu7R169ZynghVSbWKputxOByy2Wxleq3NZpOHR9l+TF4eVedEns1e45Zen56eLunKfxRuVQ13r1vex+2ouvw+lOfvgnRn/j5Ul98Fif824M5SraLJ19dXOTk51yzPzc294f1M+/btq8ixqoWSM238rMDvAq7G7wPuJFXn/+6Ug+bNm19z71JBQYHS0tLu6JvAAQDAratW0dSjRw8lJyfr3Ln//Gn9J598ooKCAvXs2dPCyQAAwO2uWkXTE088IR8fH02bNk1ffPGFdu3apSVLligiIoIzTQAA4JZUu3ua3nzzTS1dulQzZ86Ut7e3Bg0apDlz5lg9GgAAuM1Vq2iSpGbNmik+Pt7qMQAAQDVTrS7PAQAAVBSbw+FwWD0EAABAVceZJgAAAANEEwAAgAGiCQAAwADRBAAAYKDaPXIA5evf//634uPj9c033+jYsWO655579MEHH1g9FiyQmJio3bt36+9//7uys7N19913a+TIkXriiSfk5sb//7qTfPHFF3r11Vd1/PhxnT9/XgEBAerbt69mzJghHx8fq8cDKgzRhBs6duyYkpKS1K5dO12+fFn8seWda/PmzWrUqJHmzZun+vXrKyUlRcuWLdPJkyf1m9/8xurxUImys7PVvn17jRs3Tr6+vjp27JjWrVunY8eOadOmTVaPB1QYHjmAG7p8+bLzLML8+fP17bffcqbpDvXjjz+qXr16LstWrFihd955R//3f/8nu91u0WSoCnbs2KGFCxdq//79CggIsHocoEJwTh03xGUXlPhpMElSmzZtlJ+fr6ysrMofCFWKn5+fJKmoqMjaQYAKxOU5AGX217/+VX5+fqpfv77Vo8ACxcXFKioq0vHjxxUbG6tevXqpcePGVo8FVBiiCUCZHDp0SDt37tT06dPl7u5u9TiwQK9evXTmzBlJ0oMPPqioqCiLJwIqFtEE4KZlZGRo1qxZCgsL06RJk6weBxaJi4vThQsXdPz4ca1fv15PP/20Nm/eTESj2iKaANyU3NxcTZo0Sd7e3tqwYYM8PT2tHgkWad26tSSpQ4cOCg0N1bBhw/TJJ5/o4YcftngyoGIQTQCM5efna+rUqcrMzNT27dtVt25dq0dCFdGmTRu5u7srLS3N6lGACkM0ATBSVFSkZ555RkeOHNHWrVu54RcuDh48qOLiYgUFBVk9ClBhiCbc0MWLF5WUlCRJOnXqlM6fP6+PPvpIktS5c+dS/wwd1dPixYv1+eefa+7cubp06ZK+/vpr57oWLVqodu3a1g2HSjVjxgy1bdtWISEh8vb21pEjR/T6668rJCREffv2tXo8oMLwcEvc0Pfff68+ffqUum7Lli3q0qVLJU8Eq/Tu3VunTp0qdR2/C3eWuLg47dmzR2lpaXI4HGrcuLH69euniRMnEs+o1ogmAAAAAzzuGQAAwADRBAAAYIBoAgAAMEA0AQAAGCCaAAAADBBNAAAABogmAAAAA0QTAACAAaIJuAOlpKQoJCREKSkpVo8CALcNogkAAMAA0QQAAGDAw+oBAFSM1NRUxcbGKjk5WTk5OWrQoIE6d+6spUuXlrr9oUOHtGnTJn3zzTfKzMxUgwYNdN999+m5555T48aNndtdvHhRa9eu1d69e5WRkaEaNWooKChIEyZM0KBBgyRJJ0+eVFRUlP7yl78oKytLvr6+atmypebPn682bdo497Vnzx69+eabOnr0qGw2mzp06KDnnntOoaGhFfvDAYAyIJqAaujIkSMaOXKk6tatq1mzZqlJkybKyMjQZ599poKCglJfc+rUKTVr1kwDBw5UnTp1lJGRoXfeeUePPfaYPvzwQ9WrV0+StGLFCr3//vt69tln1aZNG128eFH//Oc/lZWV5dzXpEmTdPnyZc2dO1eNGjXSuXPndPDgQeXk5Di32bhxo9asWaOhQ4dq6tSpKiwsVHx8vEaNGqWEhAS1aNGiQn9GAHCzbA6Hw2H1EADK17hx43T48GHt3bvXGTtXS0lJ0dixY7VlyxZ16dKl1H0UFxcrPz9f3bp1069//WuNHTtWkhQREaHg4GDFxsaW+rpz584pPDxcCxYs0Lhx40rdJj09XX379tXIkSMVGRnpXJ6Xl6f+/fvr/vvv15o1a27yqAGgYnGmCahmLl68qL/85S967LHHSg2m68nLy9P69ev18ccf69SpUyouLnauS01Ndf57WFiYdu/erVWrVunBBx9Uu3bt5O3t7Vzv5+en4OBgxcfH6/Lly+rSpYtat24tN7f/3EL55ZdfqqioSI8++qiKioqcy728vNSpUyf+qg9AlUQ0AdVMTk6OiouLFRAQcFOve+6555ScnKxp06YpLCxMtWrVks1m0+TJk5Wfn+/cLjIyUoGBgdqzZ49ee+01eXl5qXv37po3b56aNm0qm82mN954Q7GxsXr99df14osvys/PTxEREXr22WdVu3ZtZWZmSpIee+yxUme5OrAAoKogmoBqpk6dOnJ3d9eZM2eMX5Obm6v/+Z//0YwZMzR58mTn8oKCAmVnZ7tsW7NmTc2aNUuzZs1SZmam9u/fr9WrV+vpp5/WRx99JElq3Lixli9fLkn617/+pcTERMXExKigoECLFy9W3bp1JUlr165Vo0aNbvWQAaBSEE1ANePt7a1OnTrpo48+0rPPPmt0ic5ms8nhcMhut7ssT0hIcLlM91MNGjTQ0KFDdeTIEb355pu6ePGiatSo4bJNs2bNNG3aNH388cc6fPiwJKl79+7y8PBQWlqa+vfvX4ajBIDKRzQB1dBvf/tbjRw5UiNGjNDkyZMVHBysH374QZ999pleeOGFa7avXbu2OnXqpPj4eNWtW1eNGzfWn//8Z7377rvy9fV12Xb48OF66KGHFBISojp16ig1NVXvvfee2rdvrxo1aujIkSNasmSJHn74YTVp0kSenp5KTk7W0aNHnWexgoKCNGvWLK1Zs0YnT55Ujx495Ovrq8zMTB06dEg1atTQrFmzKuVnBQCmiCagGmrdurXeffddrV27VqtXr1ZeXp78/f0VHh5+zdmkEqtXr9ayZcu0cuVKFRUVqUOHDtq8ebOmTJnisl14eLg+++wz55mlgIAADR48WE8//bQkyd/fX8HBwdq2bZtOnz4tSbr77rv1m9/8RmPGjHHuZ8qUKWrevLm2bNmiDz/8UAUFBfL391fbtm01cuTICvrJAEDZ8cgBAAAAA/yJCgAAgAGiCQAAwADRBAAAYIBoAgAAMEA0AQAAGCCaAAAADBBNAAAABogmAAAAA0QTAACAAaIJAADAANEEAABggGgCAAAw8P8BelZuvHdcTf8AAAAASUVORK5CYII=",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"df_survivants = df[df['survivant'] == 1]\n",
"px.histogram(df_survivants, x='classe', color='personne', text_auto=True).update_layout(bargap=0.2).show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Classe / Âge / Survie"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Quelle est la distribution de l'âge des passagers en fonction de la classe ?"
]
},
{
"cell_type": "code",
"execution_count": 73,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:43.875433Z",
"iopub.status.busy": "2026-10-30T13:10:43.875245Z",
"iopub.status.idle": "2026-10-30T13:10:43.882110Z",
"shell.execute_reply": "2026-10-30T13:10:43.881765Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
age
\n",
"
\n",
"
\n",
"
classe
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
1
\n",
"
39.159918
\n",
"
\n",
"
\n",
"
2
\n",
"
29.506705
\n",
"
\n",
"
\n",
"
3
\n",
"
24.816367
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" age\n",
"classe \n",
"1 39.159918\n",
"2 29.506705\n",
"3 24.816367"
]
},
"execution_count": 73,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# df.groupby('classe')['age'].mean()\n",
"df.pivot_table(values='age', index='classe', aggfunc='mean')"
]
},
{
"cell_type": "code",
"execution_count": 74,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:43.883905Z",
"iopub.status.busy": "2026-10-30T13:10:43.883745Z",
"iopub.status.idle": "2026-10-30T13:10:43.885924Z",
"shell.execute_reply": "2026-10-30T13:10:43.885540Z"
}
},
"outputs": [],
"source": [
"# plot_proportions(df, 'age_group', 'classe')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Nous souhaitons analyser comment une variable continue (comme l'âge) varie au sein de groupes catégoriels (comme la classe), tout en distinguant les sous-groupes à l'intérieur de chaque catégorie à l'aide d'une variable de différenciation (ici, le statut de survivant). Ce type de graphique permet d'avoir une vue claire des distributions, de voir si certaines catégories ont des valeurs extrêmes ou des tendances particulières, et de faire des comparaisons entre des groupes sous différents critères."
]
},
{
"cell_type": "code",
"execution_count": 75,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:43.887568Z",
"iopub.status.busy": "2026-10-30T13:10:43.887421Z",
"iopub.status.idle": "2026-10-30T13:10:44.073275Z",
"shell.execute_reply": "2026-10-30T13:10:44.072790Z"
}
},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"sns.FacetGrid(df, \n",
" col='classe', \n",
" hue='survivant', \n",
" sharex=True, \n",
" sharey=False, \n",
" ).map(sns.kdeplot, 'age', fill=True, alpha=0.15).add_legend();\n",
"plt.tight_layout()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Avantage de la première classe à tous les âges : pour les passagers de première classe, la courbe de densité de survie (orange) reste systématiquement supérieure à la courbe des non-survivants (bleue) dans presque toutes les tranches d'âge. Cela indique que le statut de première classe conférait un fort avantage en termes de survie, quel que soit l'âge. \n",
"Dans les classes moyenne et inférieure, la courbe de survie atteint un pic marqué chez les plus jeunes (0-10 ans), confirmant que la politique « les femmes et les enfants d'abord » offrait un avantage en termes de survie aux enfants, même dans les classes inférieures. \n",
"Les graphiques mettent en évidence comment la combinaison de l'âge et de la classe sociale a déterminé les chances de survie : un jeune enfant, quelle que soit sa classe sociale, avait plus de chances de survivre qu'un jeune adulte de la même classe sociale. Un jeune adulte de troisième classe avait beaucoup moins de chances de survivre qu'un jeune adulte de première classe, comme le montre clairement le pic bleu proéminent pour Pclass=3 chez les 20-30 ans."
]
},
{
"cell_type": "code",
"execution_count": 77,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:44.727951Z",
"iopub.status.busy": "2026-10-30T13:10:44.727789Z",
"iopub.status.idle": "2026-10-30T13:10:44.960354Z",
"shell.execute_reply": "2026-10-30T13:10:44.959880Z"
}
},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"sns.FacetGrid(df, \n",
" col='sexe', \n",
" hue='survivant', \n",
" sharex=True, \n",
" sharey=False, \n",
" ).map(sns.kdeplot, 'age', fill=True, alpha=0.15).add_legend();\n",
"plt.tight_layout()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Le graphique de survie des passagers masculins révèle un désavantage prononcé. La courbe des non-survivants (en bleu) reste systématiquement plus élevée que celle des survivants (en orange) pour presque tous les âges, du jeune âge adulte à la vieillesse. La seule exception est un petit pic dans la courbe de survie autour de 0 à 5 ans, qui représente les jeunes garçons ayant bénéficié de la politique « les femmes et les enfants d'abord ». Le pic le plus marqué de la courbe des hommes non survivants se situe au début de la vingtaine, ce qui souligne que les jeunes hommes constituaient le groupe le plus vulnérable. Ils étaient les derniers à avoir accès aux canots de sauvetage et devaient faire face à des difficultés physiques importantes pour s'échapper des cabines situées sur les ponts inférieurs.\n",
"\n",
"En revanche, le graphique de survie des passagères montre que la courbe des survivantes (en orange) est supérieure à celle des non-survivantes (en bleu) pour la majorité des tranches d'âge. Les chances de survie des femmes étant nettement supérieures à celles des hommes.\n",
"\n"
]
},
{
"cell_type": "code",
"execution_count": 80,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:45.478516Z",
"iopub.status.busy": "2026-10-30T13:10:45.478319Z",
"iopub.status.idle": "2026-10-30T13:10:45.691859Z",
"shell.execute_reply": "2026-10-30T13:10:45.691331Z"
}
},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"px.scatter(df, x='prix', y='age', color='survivant', marginal_x='histogram', marginal_y='histogram')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Classe / Sexe / Survie"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Regroupons les données d'abord par sexe et classe, puis sélectionnons le taux de survie et enfin calculons le taux de survie moyen pour chaque groupe."
]
},
{
"cell_type": "code",
"execution_count": 86,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:46.470046Z",
"iopub.status.busy": "2026-10-30T13:10:46.469842Z",
"iopub.status.idle": "2026-10-30T13:10:46.477673Z",
"shell.execute_reply": "2026-10-30T13:10:46.477112Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
classe
\n",
"
1
\n",
"
2
\n",
"
3
\n",
"
\n",
"
\n",
"
sexe
\n",
"
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
F
\n",
"
0.965278
\n",
"
0.886792
\n",
"
0.490741
\n",
"
\n",
"
\n",
"
H
\n",
"
0.340782
\n",
"
0.146199
\n",
"
0.152130
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
"classe 1 2 3\n",
"sexe \n",
"F 0.965278 0.886792 0.490741\n",
"H 0.340782 0.146199 0.152130"
]
},
"execution_count": 86,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.groupby(['sexe','classe'])['survivant'].agg('mean').unstack()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On note que la classe sociale a un impact sur la survie. Les femmes de première classe ont survécu à 96 %, contre 50 % pour celle de troisième classe. En ce qui concerne les hommes, 36 % des hommes de première classe ont survécu, contre 13 % pour ceux de troisième classe."
]
},
{
"cell_type": "code",
"execution_count": 87,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:46.479416Z",
"iopub.status.busy": "2026-10-30T13:10:46.479229Z",
"iopub.status.idle": "2026-10-30T13:10:46.614856Z",
"shell.execute_reply": "2026-10-30T13:10:46.614333Z"
}
},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"sns.barplot(data=df, x='classe', y='survivant', hue='sexe', \n",
" errorbar=None, palette={'F': 'pink', 'H': 'blue'});"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Pour obtenir les mêmes résultats que précédemment, on peut utiliser la méthode `pivot_table()` qui permet de créer un tableau croisé dynamique. Voici un petit dessin pour comprendre :"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Combien de survivants par classe et sexe ?"
]
},
{
"cell_type": "code",
"execution_count": 88,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:46.616764Z",
"iopub.status.busy": "2026-10-30T13:10:46.616576Z",
"iopub.status.idle": "2026-10-30T13:10:46.625564Z",
"shell.execute_reply": "2026-10-30T13:10:46.625134Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
"
],
"text/plain": [
"classe 1 2 3\n",
"sexe \n",
"F 0.965278 0.886792 0.490741\n",
"H 0.340782 0.146199 0.152130"
]
},
"execution_count": 89,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.pivot_table(index='sexe', columns='classe', values='survivant', aggfunc='mean')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"La même information peut être affichée sous forme de mappe de chaleur (heatmap) :"
]
},
{
"cell_type": "code",
"execution_count": 90,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:46.638032Z",
"iopub.status.busy": "2026-10-30T13:10:46.637853Z",
"iopub.status.idle": "2026-10-30T13:10:46.785019Z",
"shell.execute_reply": "2026-10-30T13:10:46.784513Z"
}
},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"pivot = df.pivot_table(values=\"survivant\", index=\"sexe\", columns=\"classe\", aggfunc=\"mean\")\n",
"sns.heatmap(pivot, annot=True, cmap=\"Blues\", fmt=\".2f\")\n",
"plt.title(\"Survival Rate by Sex & Class\")\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On peut spécifier plusieurs niveaux d'index pour obtenir des résultats plus détaillés, ici on a ajouté le groupe d'âge et, pour chaque groupe, le sexe."
]
},
{
"cell_type": "code",
"execution_count": 91,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:46.786895Z",
"iopub.status.busy": "2026-10-30T13:10:46.786706Z",
"iopub.status.idle": "2026-10-30T13:10:46.799870Z",
"shell.execute_reply": "2026-10-30T13:10:46.799386Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
classe
\n",
"
1
\n",
"
2
\n",
"
3
\n",
"
\n",
"
\n",
"
age_group
\n",
"
sexe
\n",
"
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0-10
\n",
"
F
\n",
"
0.0
\n",
"
11.0
\n",
"
14.0
\n",
"
\n",
"
\n",
"
H
\n",
"
3.0
\n",
"
11.0
\n",
"
11.0
\n",
"
\n",
"
\n",
"
10-20
\n",
"
F
\n",
"
15.0
\n",
"
15.0
\n",
"
20.0
\n",
"
\n",
"
\n",
"
H
\n",
"
3.0
\n",
"
2.0
\n",
"
9.0
\n",
"
\n",
"
\n",
"
20-30
\n",
"
F
\n",
"
32.0
\n",
"
34.0
\n",
"
25.0
\n",
"
\n",
"
\n",
"
H
\n",
"
12.0
\n",
"
5.0
\n",
"
26.0
\n",
"
\n",
"
\n",
"
30-40
\n",
"
F
\n",
"
33.0
\n",
"
19.0
\n",
"
9.0
\n",
"
\n",
"
\n",
"
H
\n",
"
14.0
\n",
"
3.0
\n",
"
11.0
\n",
"
\n",
"
\n",
"
40-50
\n",
"
F
\n",
"
22.0
\n",
"
11.0
\n",
"
3.0
\n",
"
\n",
"
\n",
"
H
\n",
"
13.0
\n",
"
1.0
\n",
"
2.0
\n",
"
\n",
"
\n",
"
50-60
\n",
"
F
\n",
"
21.0
\n",
"
2.0
\n",
"
NaN
\n",
"
\n",
"
\n",
"
H
\n",
"
7.0
\n",
"
0.0
\n",
"
0.0
\n",
"
\n",
"
\n",
"
60-70
\n",
"
F
\n",
"
4.0
\n",
"
NaN
\n",
"
1.0
\n",
"
\n",
"
\n",
"
H
\n",
"
0.0
\n",
"
1.0
\n",
"
0.0
\n",
"
\n",
"
\n",
"
70-80
\n",
"
F
\n",
"
1.0
\n",
"
NaN
\n",
"
NaN
\n",
"
\n",
"
\n",
"
H
\n",
"
1.0
\n",
"
NaN
\n",
"
0.0
\n",
"
\n",
"
\n",
"
Unknown
\n",
"
F
\n",
"
11.0
\n",
"
2.0
\n",
"
34.0
\n",
"
\n",
"
\n",
"
H
\n",
"
8.0
\n",
"
2.0
\n",
"
16.0
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
"classe 1 2 3\n",
"age_group sexe \n",
"0-10 F 0.0 11.0 14.0\n",
" H 3.0 11.0 11.0\n",
"10-20 F 15.0 15.0 20.0\n",
" H 3.0 2.0 9.0\n",
"20-30 F 32.0 34.0 25.0\n",
" H 12.0 5.0 26.0\n",
"30-40 F 33.0 19.0 9.0\n",
" H 14.0 3.0 11.0\n",
"40-50 F 22.0 11.0 3.0\n",
" H 13.0 1.0 2.0\n",
"50-60 F 21.0 2.0 NaN\n",
" H 7.0 0.0 0.0\n",
"60-70 F 4.0 NaN 1.0\n",
" H 0.0 1.0 0.0\n",
"70-80 F 1.0 NaN NaN\n",
" H 1.0 NaN 0.0\n",
"Unknown F 11.0 2.0 34.0\n",
" H 8.0 2.0 16.0"
]
},
"execution_count": 91,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.pivot_table(index=['age_group','sexe'], columns='classe', values='survivant', aggfunc='sum', observed=False)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"La même technique peut s'appliquer pour les colonnes :"
]
},
{
"cell_type": "code",
"execution_count": 92,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:46.801626Z",
"iopub.status.busy": "2026-10-30T13:10:46.801443Z",
"iopub.status.idle": "2026-10-30T13:10:46.823422Z",
"shell.execute_reply": "2026-10-30T13:10:46.822819Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
classe
\n",
"
1
\n",
"
2
\n",
"
3
\n",
"
\n",
"
\n",
"
\n",
"
port_depart
\n",
"
C
\n",
"
Q
\n",
"
S
\n",
"
C
\n",
"
Q
\n",
"
S
\n",
"
C
\n",
"
Q
\n",
"
S
\n",
"
\n",
"
\n",
"
age_group
\n",
"
sexe
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0-10
\n",
"
F
\n",
"
NaN
\n",
"
NaN
\n",
"
0.0
\n",
"
2.0
\n",
"
NaN
\n",
"
9.0
\n",
"
6.0
\n",
"
NaN
\n",
"
8.0
\n",
"
\n",
"
\n",
"
H
\n",
"
1.0
\n",
"
NaN
\n",
"
2.0
\n",
"
1.0
\n",
"
NaN
\n",
"
10.0
\n",
"
2.0
\n",
"
0.0
\n",
"
9.0
\n",
"
\n",
"
\n",
"
10-20
\n",
"
F
\n",
"
5.0
\n",
"
NaN
\n",
"
10.0
\n",
"
2.0
\n",
"
NaN
\n",
"
13.0
\n",
"
7.0
\n",
"
4.0
\n",
"
9.0
\n",
"
\n",
"
\n",
"
H
\n",
"
2.0
\n",
"
NaN
\n",
"
1.0
\n",
"
1.0
\n",
"
NaN
\n",
"
1.0
\n",
"
3.0
\n",
"
0.0
\n",
"
6.0
\n",
"
\n",
"
\n",
"
20-30
\n",
"
F
\n",
"
16.0
\n",
"
NaN
\n",
"
16.0
\n",
"
7.0
\n",
"
1.0
\n",
"
26.0
\n",
"
2.0
\n",
"
4.0
\n",
"
19.0
\n",
"
\n",
"
\n",
"
H
\n",
"
7.0
\n",
"
NaN
\n",
"
5.0
\n",
"
2.0
\n",
"
NaN
\n",
"
3.0
\n",
"
5.0
\n",
"
2.0
\n",
"
19.0
\n",
"
\n",
"
\n",
"
30-40
\n",
"
F
\n",
"
15.0
\n",
"
2.0
\n",
"
15.0
\n",
"
NaN
\n",
"
NaN
\n",
"
19.0
\n",
"
1.0
\n",
"
0.0
\n",
"
8.0
\n",
"
\n",
"
\n",
"
H
\n",
"
6.0
\n",
"
NaN
\n",
"
8.0
\n",
"
0.0
\n",
"
0.0
\n",
"
3.0
\n",
"
1.0
\n",
"
0.0
\n",
"
10.0
\n",
"
\n",
"
\n",
"
40-50
\n",
"
F
\n",
"
15.0
\n",
"
NaN
\n",
"
7.0
\n",
"
NaN
\n",
"
NaN
\n",
"
11.0
\n",
"
1.0
\n",
"
NaN
\n",
"
2.0
\n",
"
\n",
"
\n",
"
H
\n",
"
6.0
\n",
"
0.0
\n",
"
7.0
\n",
"
0.0
\n",
"
NaN
\n",
"
1.0
\n",
"
0.0
\n",
"
0.0
\n",
"
2.0
\n",
"
\n",
"
\n",
"
50-60
\n",
"
F
\n",
"
11.0
\n",
"
NaN
\n",
"
10.0
\n",
"
NaN
\n",
"
NaN
\n",
"
2.0
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
\n",
"
\n",
"
H
\n",
"
4.0
\n",
"
NaN
\n",
"
3.0
\n",
"
NaN
\n",
"
0.0
\n",
"
0.0
\n",
"
NaN
\n",
"
NaN
\n",
"
0.0
\n",
"
\n",
"
\n",
"
60-70
\n",
"
F
\n",
"
1.0
\n",
"
NaN
\n",
"
2.0
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
1.0
\n",
"
\n",
"
\n",
"
H
\n",
"
0.0
\n",
"
NaN
\n",
"
0.0
\n",
"
NaN
\n",
"
0.0
\n",
"
1.0
\n",
"
NaN
\n",
"
0.0
\n",
"
0.0
\n",
"
\n",
"
\n",
"
70-80
\n",
"
F
\n",
"
NaN
\n",
"
NaN
\n",
"
1.0
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
\n",
"
\n",
"
H
\n",
"
0.0
\n",
"
NaN
\n",
"
1.0
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
0.0
\n",
"
0.0
\n",
"
\n",
"
\n",
"
Unknown
\n",
"
F
\n",
"
6.0
\n",
"
NaN
\n",
"
5.0
\n",
"
NaN
\n",
"
1.0
\n",
"
1.0
\n",
"
5.0
\n",
"
25.0
\n",
"
4.0
\n",
"
\n",
"
\n",
"
H
\n",
"
2.0
\n",
"
NaN
\n",
"
6.0
\n",
"
1.0
\n",
"
0.0
\n",
"
1.0
\n",
"
4.0
\n",
"
5.0
\n",
"
7.0
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
"classe 1 2 3 \n",
"port_depart C Q S C Q S C Q S\n",
"age_group sexe \n",
"0-10 F NaN NaN 0.0 2.0 NaN 9.0 6.0 NaN 8.0\n",
" H 1.0 NaN 2.0 1.0 NaN 10.0 2.0 0.0 9.0\n",
"10-20 F 5.0 NaN 10.0 2.0 NaN 13.0 7.0 4.0 9.0\n",
" H 2.0 NaN 1.0 1.0 NaN 1.0 3.0 0.0 6.0\n",
"20-30 F 16.0 NaN 16.0 7.0 1.0 26.0 2.0 4.0 19.0\n",
" H 7.0 NaN 5.0 2.0 NaN 3.0 5.0 2.0 19.0\n",
"30-40 F 15.0 2.0 15.0 NaN NaN 19.0 1.0 0.0 8.0\n",
" H 6.0 NaN 8.0 0.0 0.0 3.0 1.0 0.0 10.0\n",
"40-50 F 15.0 NaN 7.0 NaN NaN 11.0 1.0 NaN 2.0\n",
" H 6.0 0.0 7.0 0.0 NaN 1.0 0.0 0.0 2.0\n",
"50-60 F 11.0 NaN 10.0 NaN NaN 2.0 NaN NaN NaN\n",
" H 4.0 NaN 3.0 NaN 0.0 0.0 NaN NaN 0.0\n",
"60-70 F 1.0 NaN 2.0 NaN NaN NaN NaN NaN 1.0\n",
" H 0.0 NaN 0.0 NaN 0.0 1.0 NaN 0.0 0.0\n",
"70-80 F NaN NaN 1.0 NaN NaN NaN NaN NaN NaN\n",
" H 0.0 NaN 1.0 NaN NaN NaN NaN 0.0 0.0\n",
"Unknown F 6.0 NaN 5.0 NaN 1.0 1.0 5.0 25.0 4.0\n",
" H 2.0 NaN 6.0 1.0 0.0 1.0 4.0 5.0 7.0"
]
},
"execution_count": 92,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.pivot_table(index=['age_group','sexe'], columns=['classe','port_depart'], values='survivant', aggfunc='sum', observed=False)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Pourcentage de survivants par sexe et age :"
]
},
{
"cell_type": "code",
"execution_count": 93,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:46.825358Z",
"iopub.status.busy": "2026-10-30T13:10:46.825186Z",
"iopub.status.idle": "2026-10-30T13:10:46.835725Z",
"shell.execute_reply": "2026-10-30T13:10:46.835222Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
age_group
\n",
"
0-10
\n",
"
10-20
\n",
"
20-30
\n",
"
30-40
\n",
"
40-50
\n",
"
50-60
\n",
"
60-70
\n",
"
70-80
\n",
"
Unknown
\n",
"
\n",
"
\n",
"
sexe
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
F
\n",
"
25
\n",
"
50
\n",
"
91
\n",
"
61
\n",
"
36
\n",
"
23
\n",
"
5
\n",
"
1
\n",
"
47
\n",
"
\n",
"
\n",
"
H
\n",
"
25
\n",
"
14
\n",
"
43
\n",
"
28
\n",
"
16
\n",
"
7
\n",
"
1
\n",
"
1
\n",
"
26
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
"age_group 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 Unknown\n",
"sexe \n",
"F 25 50 91 61 36 23 5 1 47\n",
"H 25 14 43 28 16 7 1 1 26"
]
},
"execution_count": 93,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.pivot_table(index='sexe', columns='age_group', values='survivant', aggfunc='sum', observed=False) \n",
"# Comme la colonne 'survivant' vaut 0 si décédé et 1 si survivant, la somme donne le nombre de survivants"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Classe / Prix / Survie"
]
},
{
"cell_type": "code",
"execution_count": 94,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:46.837415Z",
"iopub.status.busy": "2026-10-30T13:10:46.837227Z",
"iopub.status.idle": "2026-10-30T13:10:47.011056Z",
"shell.execute_reply": "2026-10-30T13:10:47.010643Z"
}
},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"sns.boxplot(data=df, x='classe', y='prix', hue='survivant');"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Le diagramme en boîte « Répartition des tarifs par classe et survie » offre une perspective plus détaillée sur la manière dont la richesse, telle que reflétée par le prix du billet, a influencé les chances de survie.\n",
"\n",
"Première classe (classe = 1) : parmi les passagers de première classe, les survivants (boîte orange) avaient un tarif médian nettement plus élevé que les non-survivants (boîte bleue). Cela suggère que même au sein de ce groupe privilégié, les passagers ayant payé plus cher ont bénéficié de meilleures chances de survie, probablement grâce à des cabines situées plus près des ponts supérieurs et des canots de sauvetage. Les valeurs extrêmes à l'extrémité supérieure du graphique des tarifs des survivants indiquent que les passagers les plus riches, qui ont payé des tarifs exceptionnellement élevés, ont presque tous survécu.\n",
"\n",
"Deuxième classe (classe = 2) : la différence entre les tarifs médians des survivants et des non-survivants est moins frappante, mais toujours présente. Les survivants ont tendance à avoir des tarifs légèrement plus élevés, ce qui suggère un léger avantage lié au fait de payer plus cher, bien que cet effet soit moins marqué qu'en première classe.\n",
"\n",
"Troisième classe (classe = 3) : pour les passagers de troisième classe, la répartition des tarifs est uniformément basse, avec presque aucune différence entre les survivants et les non-survivants. Dans ce groupe, le prix du billet a eu un impact minime sur la survie. Les obstacles physiques importants que représentaient les ponts inférieurs et l'accès limité aux canots de sauvetage ont éclipsé les différences économiques mineures."
]
},
{
"cell_type": "code",
"execution_count": 95,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:47.013102Z",
"iopub.status.busy": "2026-10-30T13:10:47.012759Z",
"iopub.status.idle": "2026-10-30T13:10:47.234498Z",
"shell.execute_reply": "2026-10-30T13:10:47.233945Z"
}
},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"\n",
"sns.kdeplot(df[df['survivant']==0]['prix'], fill=True, alpha=0.25, label='Décédés', color='red')\n",
"sns.kdeplot(df[df['survivant']==1]['prix'], fill=True, alpha=0.25, label='Survivants', color='green')\n",
"plt.legend();"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Prix / Âge / Sexe / Survie"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Nous souhaitons explorer la relation entre le prix payé et l'âge, tout en tenant compte du statut de survie (`survivant`) et du sexe.\n",
"\n",
"La commande `lmplot()` est utilisée pour afficher une **régression linéaire** entre deux variables continues tout en permettant de visualiser la relation à l'aide de couleurs et de facettes.\n",
"\n",
"- Le graphique affiche une régression linéaire entre les deux variables continues `prix` et `age`. Cela signifie qu'une droite (ou un autre modèle de régression si spécifié) est tracée pour chaque groupe de la variable `survivant`. La ligne de régression montre la tendance générale de la relation entre prix et âge. Par exemple, on observe que les personnes plus âgées ont tendance à payer plus cher.\n",
"- En fonction de la variable `hue='survivant'`, chaque groupe (survivant vs non-survivant) aura une couleur différente. Cela permet de comparer les relations entre prix et âge pour chaque groupe de manière visuellement distincte.\n",
"- Grâce à l'argument `col='sexe'`, on a des graphiques séparés pour chaque niveau de la variable sexe. Cela permet de visualiser comment la relation entre prix et âge varie entre les hommes et les femmes, par exemple.\n",
"- Le paramètre `markers=['o', 'x']` permet d'utiliser différents types de marqueurs pour les points représentant les survivants et les non-survivants, facilitant ainsi la distinction visuelle entre les deux groupes.\n",
"\n",
"Ce graphique est pertinent si on pense que la relation entre deux variables continues est linéaire (ou presque linéaire) et souhaitons l'examiner graphiquement tout en incluant des groupes distincts (par exemple, selon `survivant`)."
]
},
{
"cell_type": "code",
"execution_count": 96,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:47.236505Z",
"iopub.status.busy": "2026-10-30T13:10:47.236286Z",
"iopub.status.idle": "2026-10-30T13:10:48.262934Z",
"shell.execute_reply": "2026-10-30T13:10:48.262464Z"
}
},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# On ajoute une quatrième dimension : le sexe\n",
"sns.lmplot(data=df, x='prix', y='age', hue='survivant', col='sexe', markers=['o', 'x'], palette='Set1');"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Un graphique 3D avec la couleur comme 4e dimension est également possible :"
]
},
{
"cell_type": "code",
"execution_count": 97,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:48.265066Z",
"iopub.status.busy": "2026-10-30T13:10:48.264853Z",
"iopub.status.idle": "2026-10-30T13:10:48.328963Z",
"shell.execute_reply": "2026-10-30T13:10:48.328512Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# df['survivant'] = df['survivant'].astype('category')\n",
"\n",
"df['survivant_code'] = df['survivant'].map({0: 'Non', 1: 'Oui'})\n",
"\n",
"px.scatter_3d(\n",
" df, \n",
" x='prix', \n",
" y='age', \n",
" z='sexe', \n",
" # color='survivant',\n",
" color='survivant_code', \n",
" # Si la colonne est numérique, on peut la colorer mais il faut color_continuous_scale\n",
" # color_continuous_scale='sunset',\n",
" # On l'a transformée en catégorielle pour utiliser color_discrete_sequence\n",
" color_discrete_sequence=['green', 'red'], # il faut que 'survivant' soit une variable catégorielle !\n",
" # labels={'prix': 'Prix', 'age': 'Âge', 'sexe': 'Sexe', 'survivant_code': 'Survivant'},\n",
" # title='Classe vs Age vs Sexe selon le statut de survie',\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Miscellanea"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### `groupby()`"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Cout moyen (en livres d'époque) des billets par classe.**\n",
"\n",
"Selon (Wikipedia)[https://fr.wikipedia.org/wiki/Passagers_du_Titanic#Le_naufrage_et_son_bilan], le prix moyen des billets était de 7 livres de l'époque pour la troisième classe. Notre jeu de données indique un prix moyen de 13,30 livres pour la troisième classe, ce qui semble un peu élevé. Cependant, il est possible que notre jeu de données ne contienne pas tous les passagers de troisième classe.\n"
]
},
{
"cell_type": "code",
"execution_count": 98,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:48.330745Z",
"iopub.status.busy": "2026-10-30T13:10:48.330588Z",
"iopub.status.idle": "2026-10-30T13:10:48.336320Z",
"shell.execute_reply": "2026-10-30T13:10:48.335853Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
prix
\n",
"
\n",
"
\n",
"
classe
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
1
\n",
"
87.508992
\n",
"
\n",
"
\n",
"
2
\n",
"
21.179196
\n",
"
\n",
"
\n",
"
3
\n",
"
13.302889
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" prix\n",
"classe \n",
"1 87.508992\n",
"2 21.179196\n",
"3 13.302889"
]
},
"execution_count": 98,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.groupby('classe')['prix'].mean().to_frame()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Filtres et `value_counts`"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Parmi les enfants, les filles ont-elles été plus sauvées que les garçons ? "
]
},
{
"cell_type": "code",
"execution_count": 99,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:48.337995Z",
"iopub.status.busy": "2026-10-30T13:10:48.337822Z",
"iopub.status.idle": "2026-10-30T13:10:48.342875Z",
"shell.execute_reply": "2026-10-30T13:10:48.342414Z"
}
},
"outputs": [
{
"data": {
"text/plain": [
"survivant\n",
"1 25\n",
"0 18\n",
"Name: count, dtype: int64"
]
},
"execution_count": 99,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"mask_mineur_homme = (df['age'] < 10) & (df['sexe']==\"H\")\n",
"df[mask_mineur_homme]['survivant'].value_counts()"
]
},
{
"cell_type": "code",
"execution_count": 100,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:48.344628Z",
"iopub.status.busy": "2026-10-30T13:10:48.344446Z",
"iopub.status.idle": "2026-10-30T13:10:48.349376Z",
"shell.execute_reply": "2026-10-30T13:10:48.348915Z"
}
},
"outputs": [
{
"data": {
"text/plain": [
"survivant\n",
"1 25\n",
"0 14\n",
"Name: count, dtype: int64"
]
},
"execution_count": 100,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"mask_mineur_femme = (df['age'] < 10) & (df['sexe']==\"F\")\n",
"df[mask_mineur_femme]['survivant'].value_counts()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Corrélation entre les données : `corr`"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"La corrélation est le degré auquel deux ou plusieurs attributs ou mesures sur le même groupe d'éléments ont tendance à varier ensemble.\n",
"\n",
"Les corrélations sont parmi les éléments les plus courants et les plus utiles de l'analyse des données. Qu'est-ce qui bouge avec quoi ? Quelles variables sont « dépendantes » et lesquelles sont « indépendantes » ? Quelles sont donc les questions pour lesquelles nous pourrions vouloir trouver des corrélations ?\n",
"\n",
"Quelles sont les corrélations que nous pourrions rechercher ?\n",
"\n",
"- Existe-t-il une corrélation entre l'âge et le prix du billet ?\n",
"- Existe-t-il une corrélation entre la classe et la survie ? Les riches ont-ils survécu davantage que les travailleurs ?\n",
"- Y a-t-il une corrélation entre l'âge et la survie ? \n",
"- Les femmes et les enfants ont-ils vraiment le droit de passer en premier ?\n",
"\n",
"Nous connaissons tous le vieil adage selon lequel, lorsqu'un navire coule, ce sont « les femmes et les enfants d'abord » qui montent à bord des canots de sauvetage. Ce vieil adage s'est-il vérifié sur le Titanic ? Les femmes et les enfants ont-ils été plus nombreux à survivre que les hommes ? Le sexe, l'âge ou la classe sociale des passagers du Titanic ont-ils joué un rôle déterminant dans leur survie ?\n",
"\n",
"Si les femmes et les enfants ont survécu plus que les hommes, il devrait y avoir une corrélation positive entre la survie et le sexe et la survie et l'âge.\n",
"\n",
"La méthode `.corr()` de pandas permet de trouver la corrélation entre deux caractéristiques quelconques d'un jeu de données."
]
},
{
"cell_type": "code",
"execution_count": 101,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:48.351016Z",
"iopub.status.busy": "2026-10-30T13:10:48.350839Z",
"iopub.status.idle": "2026-10-30T13:10:48.355745Z",
"shell.execute_reply": "2026-10-30T13:10:48.355238Z"
}
},
"outputs": [
{
"data": {
"text/plain": [
"-0.055512520192146246"
]
},
"execution_count": 101,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df['age'].corr(df['survivant'])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Une valeur +1 signifie qu'il existe une corrélation positive parfaite entre deux caractéristiques, c'est-à-dire que lorsqu'une caractéristique augmente, l'autre augmente exactement dans les mêmes proportions.\n",
"\n",
"Zéro signifie qu'il n'y a pas de corrélation entre deux caractéristiques. Elles se déplacent complètement au hasard l'une par rapport à l'autre.\n",
"\n",
"Une valeur -1 signifie qu'il existe une corrélation parfaite, négative ou inverse entre deux caractéristiques. Lorsqu'une caractéristique augmente, l'autre diminue et vice versa.\n",
"\n",
"D'après la corrélation entre le sexe et la survie (53 %) et l'âge et la survie (-0,05 %), nous pouvons affirmer certaines choses :\n",
"\n",
"- le fait qu'un passager soit un homme était fortement corrélé négativement avec sa survie à bord du Titanic.\n",
"- le fait qu'un passager soit plus âgé est très faiblement corrélé négativement avec sa survie.\n",
"\n",
"Mais « l'âge » n'est pas « les enfants », n'est-ce pas ? Comment savoir si le fait d'avoir moins de 10 ans augmentait les chances de survie ? Quelle était la probabilité qu'une personne de moins de 10 ans survive par rapport à une personne de plus de 10 ans ?"
]
},
{
"cell_type": "code",
"execution_count": 102,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:48.357553Z",
"iopub.status.busy": "2026-10-30T13:10:48.357372Z",
"iopub.status.idle": "2026-10-30T13:10:48.362092Z",
"shell.execute_reply": "2026-10-30T13:10:48.361716Z"
}
},
"outputs": [
{
"data": {
"text/plain": [
"-0.14006219496281994"
]
},
"execution_count": 102,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"masque = df['age'] < 10\n",
"df[masque]['age'].corr(df['survivant'])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On ne peut pas calculer la corrélation avec le sexe car c'est une variable catégorielle. On peut utiliser la méthode `groupby()` pour regrouper les données par sexe et calculer la moyenne de survie pour chaque groupe. Ou sinon ajouter une colonne où les valeurs catégorielles sont encodées par des valeurs numériques, par exemple 0 pour les hommes et 1 pour les femmes."
]
},
{
"cell_type": "code",
"execution_count": 103,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:48.363913Z",
"iopub.status.busy": "2026-10-30T13:10:48.363748Z",
"iopub.status.idle": "2026-10-30T13:10:48.375128Z",
"shell.execute_reply": "2026-10-30T13:10:48.374676Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
classe
\n",
"
survivant
\n",
"
sexe
\n",
"
age
\n",
"
prix
\n",
"
port_depart
\n",
"
age_group
\n",
"
sexe_num
\n",
"
personne
\n",
"
survivant_code
\n",
"
sexe_code
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
1
\n",
"
1
\n",
"
F
\n",
"
29.0000
\n",
"
211.3375
\n",
"
S
\n",
"
20-30
\n",
"
0
\n",
"
femme
\n",
"
Oui
\n",
"
0
\n",
"
\n",
"
\n",
"
1
\n",
"
1
\n",
"
1
\n",
"
H
\n",
"
0.9167
\n",
"
151.5500
\n",
"
S
\n",
"
0-10
\n",
"
1
\n",
"
enfant
\n",
"
Oui
\n",
"
1
\n",
"
\n",
"
\n",
"
2
\n",
"
1
\n",
"
0
\n",
"
F
\n",
"
2.0000
\n",
"
151.5500
\n",
"
S
\n",
"
0-10
\n",
"
0
\n",
"
enfant
\n",
"
Non
\n",
"
0
\n",
"
\n",
"
\n",
"
3
\n",
"
1
\n",
"
0
\n",
"
H
\n",
"
30.0000
\n",
"
151.5500
\n",
"
S
\n",
"
20-30
\n",
"
1
\n",
"
homme
\n",
"
Non
\n",
"
1
\n",
"
\n",
"
\n",
"
4
\n",
"
1
\n",
"
0
\n",
"
F
\n",
"
25.0000
\n",
"
151.5500
\n",
"
S
\n",
"
20-30
\n",
"
0
\n",
"
femme
\n",
"
Non
\n",
"
0
\n",
"
\n",
"
\n",
"
...
\n",
"
...
\n",
"
...
\n",
"
...
\n",
"
...
\n",
"
...
\n",
"
...
\n",
"
...
\n",
"
...
\n",
"
...
\n",
"
...
\n",
"
...
\n",
"
\n",
"
\n",
"
1304
\n",
"
3
\n",
"
0
\n",
"
F
\n",
"
14.5000
\n",
"
14.4542
\n",
"
C
\n",
"
10-20
\n",
"
0
\n",
"
femme
\n",
"
Non
\n",
"
0
\n",
"
\n",
"
\n",
"
1305
\n",
"
3
\n",
"
0
\n",
"
F
\n",
"
NaN
\n",
"
14.4542
\n",
"
C
\n",
"
Unknown
\n",
"
0
\n",
"
femme
\n",
"
Non
\n",
"
0
\n",
"
\n",
"
\n",
"
1306
\n",
"
3
\n",
"
0
\n",
"
H
\n",
"
26.5000
\n",
"
7.2250
\n",
"
C
\n",
"
20-30
\n",
"
1
\n",
"
homme
\n",
"
Non
\n",
"
1
\n",
"
\n",
"
\n",
"
1307
\n",
"
3
\n",
"
0
\n",
"
H
\n",
"
27.0000
\n",
"
7.2250
\n",
"
C
\n",
"
20-30
\n",
"
1
\n",
"
homme
\n",
"
Non
\n",
"
1
\n",
"
\n",
"
\n",
"
1308
\n",
"
3
\n",
"
0
\n",
"
H
\n",
"
29.0000
\n",
"
7.8750
\n",
"
S
\n",
"
20-30
\n",
"
1
\n",
"
homme
\n",
"
Non
\n",
"
1
\n",
"
\n",
" \n",
"
\n",
"
1309 rows × 11 columns
\n",
"
"
],
"text/plain": [
" classe survivant sexe age prix port_depart age_group \\\n",
"0 1 1 F 29.0000 211.3375 S 20-30 \n",
"1 1 1 H 0.9167 151.5500 S 0-10 \n",
"2 1 0 F 2.0000 151.5500 S 0-10 \n",
"3 1 0 H 30.0000 151.5500 S 20-30 \n",
"4 1 0 F 25.0000 151.5500 S 20-30 \n",
"... ... ... ... ... ... ... ... \n",
"1304 3 0 F 14.5000 14.4542 C 10-20 \n",
"1305 3 0 F NaN 14.4542 C Unknown \n",
"1306 3 0 H 26.5000 7.2250 C 20-30 \n",
"1307 3 0 H 27.0000 7.2250 C 20-30 \n",
"1308 3 0 H 29.0000 7.8750 S 20-30 \n",
"\n",
" sexe_num personne survivant_code sexe_code \n",
"0 0 femme Oui 0 \n",
"1 1 enfant Oui 1 \n",
"2 0 enfant Non 0 \n",
"3 1 homme Non 1 \n",
"4 0 femme Non 0 \n",
"... ... ... ... ... \n",
"1304 0 femme Non 0 \n",
"1305 0 femme Non 0 \n",
"1306 1 homme Non 1 \n",
"1307 1 homme Non 1 \n",
"1308 1 homme Non 1 \n",
"\n",
"[1309 rows x 11 columns]"
]
},
"execution_count": 103,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Catégorisation de la variable 'sexe' en 'sexe_code'\n",
"df['sexe_code'] = df['sexe'].astype('category').cat.codes\n",
"# df['sexe_code'] = df['sexe'].map({'H': 0, 'F': 1})\n",
"# df['sexe_code'] = df['sexe'].replace({'H': 0, 'F': 1})\n",
"df"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Pour l'instant, examinons la matrice de corrélation."
]
},
{
"cell_type": "code",
"execution_count": 104,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:48.377020Z",
"iopub.status.busy": "2026-10-30T13:10:48.376836Z",
"iopub.status.idle": "2026-10-30T13:10:48.383292Z",
"shell.execute_reply": "2026-10-30T13:10:48.382852Z"
}
},
"outputs": [
{
"data": {
"text/plain": [
"survivant 1.000000\n",
"sexe_code -0.528693\n",
"age -0.055513\n",
"classe -0.312469\n",
"prix 0.244265\n",
"Name: survivant, dtype: float64"
]
},
"execution_count": 104,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"corr_matrix = df[['survivant', 'sexe_code', 'age', 'classe', 'prix']].corr()\n",
"\n",
"# Ordonner les colonnes et les lignes selon la corrélation avec la variable 'survivant'\n",
"sorted_columns = corr_matrix['survivant'].abs().sort_values(ascending=False).index\n",
"sorted_corr_matrix = corr_matrix.loc[sorted_columns, sorted_columns]\n",
"\n",
"corr_matrix['survivant']"
]
},
{
"cell_type": "code",
"execution_count": 105,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:48.384954Z",
"iopub.status.busy": "2026-10-30T13:10:48.384804Z",
"iopub.status.idle": "2026-10-30T13:10:48.412835Z",
"shell.execute_reply": "2026-10-30T13:10:48.412381Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Avec Seaborn\n",
"# ---------------------------\n",
"# sns.heatmap(sorted_corr_matrix, linewidths=0.5, annot=True, cbar=True, cmap=\"coolwarm\")\n",
"# plt.show()\n",
"\n",
"#Avec plotly.express\n",
"# ---------------------------\n",
"px.imshow(sorted_corr_matrix, \n",
" text_auto='.2f', \n",
" color_continuous_scale='RdBu_r', \n",
" aspect='auto').show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Attention : toutes les variables codées ne se valent pas"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Nous venons de remplacer une variable catégorielle par des codes numériques pour la faire entrer dans une matrice de corrélation. C'est une manipulation **très courante** : et qui n'est légitime que dans certains cas. Il faut distinguer trois situations.\n",
"\n",
"| Type de variable | Exemple ici | Corrélation de Pearson ? |\n",
"|---|---|---|\n",
"| **Numérique** | `age`, `prix` | ✔️ oui, c'est son usage normal |\n",
"| **Binaire** (2 modalités) | `sexe` → 0 / 1 | ✔️ oui : c'est la *corrélation point-bisériale*. Seul le **signe** dépend de la convention de codage |\n",
"| **Ordinale** (modalités ordonnées) | `classe` → 1, 2, 3 | ⚠️ acceptable, car l'ordre a un sens : mais on ne mesure que la part **monotone** de la relation |\n",
"| **Nominale** (≥ 3 modalités non ordonnées) | `port_depart` → C, Q, S | ❌ **non**, le résultat n'a aucun sens |"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"
\n",
"\n",
"**Pourquoi c'est faux pour une variable nominale.** Coder C, Q, S par 0, 1, 2 revient à affirmer que Queenstown est « entre » Cherbourg et Southampton, et que l'écart C→Q est le même que l'écart Q→S. Ces affirmations n'ont aucun sens : l'ordre choisi est purement arbitraire, hérité de l'alphabet.\n",
"\n",
"La conséquence est facile à vérifier : **changez l'ordre du codage, et le coefficient change.**\n",
"\n",
"
"
]
},
{
"cell_type": "code",
"execution_count": 106,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:48.414694Z",
"iopub.status.busy": "2026-10-30T13:10:48.414482Z",
"iopub.status.idle": "2026-10-30T13:10:48.422294Z",
"shell.execute_reply": "2026-10-30T13:10:48.421761Z"
}
},
"outputs": [
{
"data": {
"text/markdown": [
"codage C,Q,S = 0,1,2 -> corrélation avec la survie = -0.1768"
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/markdown": [
"codage S,C,Q = 0,1,2 -> corrélation avec la survie = 0.0998"
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Le même port de départ, codé de deux façons également arbitraires\n",
"codage_1 = df['port_depart'].map({'C': 0, 'Q': 1, 'S': 2}) # ordre alphabétique\n",
"codage_2 = df['port_depart'].map({'S': 0, 'C': 1, 'Q': 2}) # un autre ordre, tout aussi valable\n",
"\n",
"display(Markdown(f\"codage C,Q,S = 0,1,2 -> corrélation avec la survie = {codage_1.corr(df['survivant']): .4f}\"))\n",
"display(Markdown(f\"codage S,C,Q = 0,1,2 -> corrélation avec la survie = {codage_2.corr(df['survivant']): .4f}\"))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Mêmes données, même question : et non seulement la valeur change, mais **le signe s'inverse**. Le coefficient ne mesure donc rien du port d'embarquement : il mesure l'ordre dans lequel nous avons écrit les lettres.\n",
"\n",
"Notez que `.astype('category').cat.codes`, bien pratique, applique exactement le codage alphabétique : il produit silencieusement le premier des deux résultats."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**La bonne façon de faire** avec une variable nominale est celle que nous avons déjà employée dans la section *Port de départ / Survie* : comparer le **taux de survie modalité par modalité**, avec `groupby` ou un tableau croisé."
]
},
{
"cell_type": "code",
"execution_count": 107,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:48.424318Z",
"iopub.status.busy": "2026-10-30T13:10:48.424120Z",
"iopub.status.idle": "2026-10-30T13:10:48.430958Z",
"shell.execute_reply": "2026-10-30T13:10:48.430563Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
taux_de_survie
\n",
"
effectif
\n",
"
\n",
"
\n",
"
port_depart
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
C
\n",
"
0.556
\n",
"
270
\n",
"
\n",
"
\n",
"
Q
\n",
"
0.358
\n",
"
123
\n",
"
\n",
"
\n",
"
S
\n",
"
0.333
\n",
"
914
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" taux_de_survie effectif\n",
"port_depart \n",
"C 0.556 270\n",
"Q 0.358 123\n",
"S 0.333 914"
]
},
"execution_count": 107,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.groupby('port_depart')['survivant'].agg(\n",
" taux_de_survie = 'mean',\n",
" effectif = 'count'\n",
").round(3)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On lit directement le résultat, sans aucune convention arbitraire : **55,6 %** de survie pour Cherbourg, **35,8 %** pour Queenstown, **33,3 %** pour Southampton. L'information est là, et elle est interprétable.\n",
"\n",
"À retenir pour vos propres analyses : une matrice de corrélation ne doit contenir que des variables **numériques, binaires ou ordinales**. Pour les autres, un tableau croisé des taux dit bien davantage : et dit vrai."
]
},
{
"cell_type": "code",
"execution_count": 108,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:48.432712Z",
"iopub.status.busy": "2026-10-30T13:10:48.432527Z",
"iopub.status.idle": "2026-10-30T13:10:48.435251Z",
"shell.execute_reply": "2026-10-30T13:10:48.434738Z"
}
},
"outputs": [],
"source": [
"# df['personne_code'] = df['personne'].astype('category').cat.codes\n",
"# corr_matrix = df[['survivant','personne_code','age','classe']].corr()\n",
"# sorted_columns = corr_matrix['survivant'].abs().sort_values(ascending=False).index\n",
"# sorted_corr_matrix = corr_matrix.loc[sorted_columns, sorted_columns]\n",
"# plt.figure(figsize=(10, 10))\n",
"# sns.heatmap(sorted_corr_matrix, linewidths=0.5, annot=True, cbar=True, cmap=\"coolwarm\");\n"
]
},
{
"cell_type": "code",
"execution_count": 109,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:48.437109Z",
"iopub.status.busy": "2026-10-30T13:10:48.436924Z",
"iopub.status.idle": "2026-10-30T13:10:48.439486Z",
"shell.execute_reply": "2026-10-30T13:10:48.438973Z"
}
},
"outputs": [],
"source": [
"# dftemp = df[['survivant','sexe_code','age','classe']]\n",
"# corr_matrix = dftemp.corr()['survivant'].sort_values(ascending=False)\n",
"\n",
"# # Sélection des colonnes ordonnées par corrélation avec 'survivant' (à l'exclusion de 'survivant' elle-même)\n",
"# sorted_columns = corr_matrix.index.tolist()[1:]\n",
"\n",
"# # Création du heatmap avec Seaborn\n",
"# sns.heatmap(dftemp[sorted_columns].corr(), annot=True, linewidths=0.5, cmap=\"coolwarm\", cbar=True);"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Autres ressources"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Voici trois excellentes vidéos (en français) de la chaîne YouTube **Machine Learnia** qui expliquent comment faire une analyse statistique du jeu de données du Titanic :\n",
"\n",
"1. [Vidéo 1 : Analyse exploratoire de données](https://www.youtube.com/watch?v=zZkNOdBWgFQ)\n",
"\n",
"\n",
"\n",
"2. [Vidéo 2 : Data visualization (à partir de 19:50)](https://www.youtube.com/watch?v=qHRLG5hsW9I&list=PLO_fdPEVlfKqMDNmCFzQISI2H_nJcEDJq&index=18)\n",
"\n",
"\n",
"\n",
"3. [Vidéo 3 : Feature engineering (à partir de 6:35)](https://www.youtube.com/watch?v=xYgfIRzNPlo&list=PLO_fdPEVlfKqMDNmCFzQISI2H_nJcEDJq&index=19)\n",
"\n",
""
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Pour aller plus loin : 🧹 analyse exploratoire 🧹 → 🔮 prédiction 🔮"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"
\n",
"\n",
"**Hors séance : pour votre curiosité.** Ce qui suit dépasse le cadre de ce cours : il s'agit d'une ouverture vers l'apprentissage automatique, présentée pour montrer à quoi sert le travail de préparation des données que nous venons de faire.\n",
"\n",
"Rien de cette section n'est exigible, et l'examen ne porte que sur Pandas. Le code utilise la bibliothèque `scikit-learn`, qui n'est pas au programme.\n",
"\n",
"
"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Dans cette exploraiton du dataset Titanic avec Pandas, nous avons vu comment filtrer et sélectionner des données, faire des statistiques descriptives, croiser et agréger des variables, visualiser les résultats avec différents types de graphiques. Tout cela nous permet de **comprendre les facteurs qui influencent la survie** des passagers. Mais ces outils restent des analyses descriptives : ils montrent des relations, mais ne permettent pas de prédire directement le destin d’un passager inconnu.\n",
"\n",
"Une fois que l’on a identifié les facteurs importants, on peut se poser la question : **\"Peut-on prédire si un passager va survivre ou non à partir de ses caractéristiques (âge, sexe, classe, tarif, etc.) ?\"**\n",
"\n",
"La prédiction cherche à **estimer une valeur future ou inconnue** sur la base de données existantes.\n",
"Les étapes générales pour prédire sont les suivantes :\n",
"\n",
"1. **Préparer et explorer les données**. \n",
"2. **Choisir un modèle** : régression logistique, arbres de décision, réseaux de neurones, etc. \n",
"3. Séparer les données en **train set** (pour entraîner le modèle) et **test set** (pour évaluer la performance)\n",
"4. **Entraîner le modèle** sur les données existantes (*train set*) \n",
"5. **Évaluer le modèle** sur des données jamais vues (*test set*) et évaluer la performance avec des métriques adaptées (accuracy, precision, recall) \n",
"\n",
"Pandas est utilisé pour la **préparation des données** avant de les passer à un modèle de machine learning. \n",
"\n",
"Dans le cas du Titanic : \n",
"\n",
"- Variable cible : `survivant` (0 = mort, 1 = survécu) \n",
"- Features (variables explicatives) : `classe`, `sexe`, `age`, `port_depart`, etc. \n",
"\n",
"C'est un **problème de classification** : prédire une catégorie (`0` ou `1`) à partir des autres caractéristiques.\n",
"\n",
"Observations après l'analyse exploratoire : \n",
"\n",
"- les femmes ont survécu plus souvent que les hommes \n",
"- les passagers de première classe ont survécu plus souvent que ceux de troisième classe \n",
"- les enfants ont un taux de survie légèrement supérieur à celui des adultes \n",
"\n",
"Un modèle de prédiction pourrait apprendre ces patterns et estimer la probabilité de survie d’un passager inconnu.\n",
"\n",
"\n",
"\n",
"\n",
"**Conclusion** \n",
"L’analyse exploratoire avec Pandas est la première étape pour comprendre les données et identifier les facteurs influençant la survie des passagers. Pour aller plus loin, on peut passer à la prédiction : estimer si un passager inconnu survivra ou non à partir de ses caractéristiques.\n",
"\n",
"Cette étape implique le **_machine learning_**, où l’on choisit un modèle (régression logistique, arbres de décision, réseau de neurones...), on entraîne ce modèle sur un train set, puis on évalue sa performance sur un test set jamais vu.\n",
"\n",
"Les modèles de machine learning permettent non seulement de quantifier l’impact de chaque variable, mais aussi de capturer des relations complexes entre elles. Les réseaux de neurones (*deep learning*) sont particulièrement utiles lorsque les patterns sont non linéaires ou que les datasets sont volumineux.\n",
"\n",
"Ainsi, le workflow complet va de l’observation des données à la prise de décision prédictive, illustrant comment l’intelligence artificielle appliquée aux données réelles peut aider à estimer des probabilités et guider des actions."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Voici plusieurs modèles de prédiction avec la bibliothèque Scikit-Learn pour les entraîner sur le dataset Titanic : "
]
},
{
"cell_type": "code",
"execution_count": 110,
"metadata": {
"execution": {
"iopub.execute_input": "2026-10-30T13:10:48.441256Z",
"iopub.status.busy": "2026-10-30T13:10:48.441055Z",
"iopub.status.idle": "2026-10-30T13:10:49.736777Z",
"shell.execute_reply": "2026-10-30T13:10:49.736180Z"
}
},
"outputs": [
{
"data": {
"text/markdown": [
"Dimension de X_train : (1047, 3), Dimension de X_test : (262, 3)"
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"
\n",
"\n",
"Par exemple, dans le cas du modèle SVC appliqué au dataset Titanic, la matrice de confusion est la suivante :\n",
"\n",
"- **Vrais négatifs (VN)** : le modèle a correctement prédit que 153 passagers n’ont pas survécu. \n",
"- **Faux positifs (FP)** : le modèle a prédit que 9 passagers survivraient alors qu’ils sont morts. \n",
"- **Faux négatifs (FN)** : le modèle a prédit que 34 passagers ne survivraient pas alors qu’ils ont survécu. \n",
"- **Vrais positifs (VP)** : le modèle a correctement prédit que 66 passagers ont survécu. \n",
"\n",
"\n",
"**Métriques de performance**\n",
"\n",
"1. **Accuracy** (précision globale) \n",
"Proportion de bonnes prédictions parmi l’ensemble des prédictions : \n",
"$$ \\frac{VN+VP}{VN+VP+FP+FN} $$\n",
"\n",
"2. **Précision (precision)** \n",
"Mesure la proportion de prédictions positives correctes. Elle pénalise les **faux positifs**. \n",
"$$ \\text{precision}=\\frac{VP}{VP+FP} $$\n",
"\n",
"3. **Rappel (recall ou sensibilité)** \n",
"Mesure la capacité du modèle à identifier correctement les survivants (classe positive). \n",
"$$ \\text{recall}=\\frac{VP}{VP+FN} $$\n",
"\n",
"4. **F1-score** \n",
"Moyenne harmonique entre la précision et le rappel. \n",
"$$ F_1=2\\dfrac{\\text{precision}\\times \\text{recall}}{\\text{precision}+\\text{recall}} $$\n",
"\n",
"5. **Spécificité** \n",
"Proportion de vrais négatifs correctement identifiés. \n",
"$$ \\text{spécificité}=\\frac{VN}{VN+FP} $$\n",
"\n",
"\n",
"\n",
"**Interprétation des résultats**\n",
"\n",
"**Accuracy** : parmi les modèles testés, le SVC obtient la meilleure performance avec une précision globale de **83,6%**. Cela signifie que **83,6% des prédictions du modèle sont correctes**. \n",
"Cependant, cette mesure ne distingue pas les erreurs concernant les **survivants** et les **non-survivants**. \n",
"La matrice de confusion montre que le modèle dentifie correctement une majorité de non-survivants, **sous-prédit la survie**, avec **34 faux négatifs**, ce qui signifie qu’un nombre significatif de survivants réels sont classés comme non-survivants. \n",
"Globalement, le modèle est **performant**, mais la sous-détection des survivants peut être problématique selon les objectifs de l’étude.\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.12.3"
}
},
"nbformat": 4,
"nbformat_minor": 2
}