{
"cells": [
{
"cell_type": "markdown",
"id": "3c94213f",
"metadata": {},
"source": [
"## Importation des bibliothéques "
]
},
{
"cell_type": "code",
"execution_count": 1,
"id": "d248e3e8",
"metadata": {},
"outputs": [],
"source": [
"import numpy as np\n",
"import pandas as pd\n",
"import matplotlib.pyplot as plt\n",
"import seaborn as sns\n",
"#from numpy import random #Aléatoire"
]
},
{
"cell_type": "markdown",
"id": "579cedc3",
"metadata": {},
"source": [
"## Partie 1"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "bf705b81",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" gender | \n",
" ethnicity | \n",
" age | \n",
" weight | \n",
" protein | \n",
" protein2 | \n",
" protein3 | \n",
" n_visits | \n",
" X | \n",
"
\n",
" \n",
" \n",
" \n",
" 0 | \n",
" Female | \n",
" White | \n",
" 72.0 | \n",
" 76.0 | \n",
" 246 | \n",
" 88 | \n",
" 136 | \n",
" 8 | \n",
" 75.252141 | \n",
"
\n",
" \n",
" 1 | \n",
" Female | \n",
" Black | \n",
" 84.1 | \n",
" 59.8 | \n",
" 210 | \n",
" 85 | \n",
" 86 | \n",
" 6 | \n",
" 55.366956 | \n",
"
\n",
" \n",
" 2 | \n",
" Female | \n",
" Black | \n",
" 79.7 | \n",
" 56.0 | \n",
" 205 | \n",
" 91 | \n",
" 110 | \n",
" 7 | \n",
" 53.413395 | \n",
"
\n",
" \n",
" 3 | \n",
" Female | \n",
" White | \n",
" 75.7 | \n",
" 66.7 | \n",
" 286 | \n",
" 68 | \n",
" 54 | \n",
" 2 | \n",
" 65.672425 | \n",
"
\n",
" \n",
" 4 | \n",
" Female | \n",
" White | \n",
" 74.6 | \n",
" 72.1 | \n",
" 171 | \n",
" 81 | \n",
" 99 | \n",
" 8 | \n",
" 67.774991 | \n",
"
\n",
" \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
"
\n",
" \n",
" 812 | \n",
" Male | \n",
" White | \n",
" 29.6 | \n",
" 84.0 | \n",
" 298 | \n",
" 150 | \n",
" 95 | \n",
" 1 | \n",
" 87.491425 | \n",
"
\n",
" \n",
" 813 | \n",
" Female | \n",
" Hispanic | \n",
" 43.2 | \n",
" 69.6 | \n",
" 243 | \n",
" 138 | \n",
" 107 | \n",
" 2 | \n",
" 75.175035 | \n",
"
\n",
" \n",
" 814 | \n",
" Male | \n",
" White | \n",
" 64.8 | \n",
" 78.2 | \n",
" 240 | \n",
" 86 | \n",
" 106 | \n",
" 6 | \n",
" 79.772296 | \n",
"
\n",
" \n",
" 815 | \n",
" Male | \n",
" Unknown | \n",
" 57.2 | \n",
" 84.6 | \n",
" 264 | \n",
" 132 | \n",
" 99 | \n",
" 0 | \n",
" 84.550631 | \n",
"
\n",
" \n",
" 816 | \n",
" Female | \n",
" White | \n",
" 60.9 | \n",
" 68.4 | \n",
" 266 | \n",
" 137 | \n",
" 70 | \n",
" 8 | \n",
" 67.878950 | \n",
"
\n",
" \n",
"
\n",
"
817 rows × 9 columns
\n",
"
"
],
"text/plain": [
" gender ethnicity age weight protein protein2 protein3 n_visits \\\n",
"0 Female White 72.0 76.0 246 88 136 8 \n",
"1 Female Black 84.1 59.8 210 85 86 6 \n",
"2 Female Black 79.7 56.0 205 91 110 7 \n",
"3 Female White 75.7 66.7 286 68 54 2 \n",
"4 Female White 74.6 72.1 171 81 99 8 \n",
".. ... ... ... ... ... ... ... ... \n",
"812 Male White 29.6 84.0 298 150 95 1 \n",
"813 Female Hispanic 43.2 69.6 243 138 107 2 \n",
"814 Male White 64.8 78.2 240 86 106 6 \n",
"815 Male Unknown 57.2 84.6 264 132 99 0 \n",
"816 Female White 60.9 68.4 266 137 70 8 \n",
"\n",
" X \n",
"0 75.252141 \n",
"1 55.366956 \n",
"2 53.413395 \n",
"3 65.672425 \n",
"4 67.774991 \n",
".. ... \n",
"812 87.491425 \n",
"813 75.175035 \n",
"814 79.772296 \n",
"815 84.550631 \n",
"816 67.878950 \n",
"\n",
"[817 rows x 9 columns]"
]
},
"execution_count": 2,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"#1. Importer le fichier csv\n",
"data= pd.read_csv(\"data.csv\",delimiter=\",\")\n",
"data"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "3862dfc0",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n",
"RangeIndex: 817 entries, 0 to 816\n",
"Data columns (total 9 columns):\n",
" # Column Non-Null Count Dtype \n",
"--- ------ -------------- ----- \n",
" 0 gender 817 non-null object \n",
" 1 ethnicity 817 non-null object \n",
" 2 age 817 non-null float64\n",
" 3 weight 817 non-null float64\n",
" 4 protein 817 non-null int64 \n",
" 5 protein2 817 non-null int64 \n",
" 6 protein3 817 non-null int64 \n",
" 7 n_visits 817 non-null int64 \n",
" 8 X 817 non-null float64\n",
"dtypes: float64(3), int64(4), object(2)\n",
"memory usage: 57.6+ KB\n"
]
}
],
"source": [
"#Afficher des informations \n",
"data.info()"
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "3c48a804",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" gender | \n",
" ethnicity | \n",
" age | \n",
" weight | \n",
" protein | \n",
" protein2 | \n",
" protein3 | \n",
" n_visits | \n",
" X | \n",
"
\n",
" \n",
" \n",
" \n",
" 0 | \n",
" Female | \n",
" White | \n",
" 72.0 | \n",
" 76.0 | \n",
" 246 | \n",
" 88 | \n",
" 136 | \n",
" 8 | \n",
" 75.252141 | \n",
"
\n",
" \n",
" 1 | \n",
" Female | \n",
" Black | \n",
" 84.1 | \n",
" 59.8 | \n",
" 210 | \n",
" 85 | \n",
" 86 | \n",
" 6 | \n",
" 55.366956 | \n",
"
\n",
" \n",
" 2 | \n",
" Female | \n",
" Black | \n",
" 79.7 | \n",
" 56.0 | \n",
" 205 | \n",
" 91 | \n",
" 110 | \n",
" 7 | \n",
" 53.413395 | \n",
"
\n",
" \n",
" 3 | \n",
" Female | \n",
" White | \n",
" 75.7 | \n",
" 66.7 | \n",
" 286 | \n",
" 68 | \n",
" 54 | \n",
" 2 | \n",
" 65.672425 | \n",
"
\n",
" \n",
" 4 | \n",
" Female | \n",
" White | \n",
" 74.6 | \n",
" 72.1 | \n",
" 171 | \n",
" 81 | \n",
" 99 | \n",
" 8 | \n",
" 67.774991 | \n",
"
\n",
" \n",
" 5 | \n",
" Female | \n",
" Black | \n",
" 70.5 | \n",
" 76.4 | \n",
" 204 | \n",
" 56 | \n",
" 112 | \n",
" 5 | \n",
" 77.983575 | \n",
"
\n",
" \n",
" 6 | \n",
" Male | \n",
" White | \n",
" 94.6 | \n",
" 71.9 | \n",
" 217 | \n",
" 85 | \n",
" 72 | \n",
" 1 | \n",
" 74.539839 | \n",
"
\n",
" \n",
" 7 | \n",
" Female | \n",
" Black | \n",
" 79.5 | \n",
" 59.1 | \n",
" 252 | \n",
" 63 | \n",
" 115 | \n",
" 3 | \n",
" 56.244106 | \n",
"
\n",
" \n",
" 8 | \n",
" Male | \n",
" Hispanic | \n",
" 87.5 | \n",
" 60.5 | \n",
" 240 | \n",
" 59 | \n",
" 120 | \n",
" 7 | \n",
" 57.310157 | \n",
"
\n",
" \n",
" 9 | \n",
" Female | \n",
" White | \n",
" 83.8 | \n",
" 67.0 | \n",
" 268 | \n",
" 82 | \n",
" 72 | \n",
" 8 | \n",
" 66.779214 | \n",
"
\n",
" \n",
" 10 | \n",
" Male | \n",
" White | \n",
" 74.7 | \n",
" 70.2 | \n",
" 270 | \n",
" 102 | \n",
" 133 | \n",
" 0 | \n",
" 69.724581 | \n",
"
\n",
" \n",
" 11 | \n",
" Female | \n",
" Black | \n",
" 84.6 | \n",
" 65.5 | \n",
" 269 | \n",
" 47 | \n",
" 103 | \n",
" 4 | \n",
" 67.474976 | \n",
"
\n",
" \n",
" 12 | \n",
" Male | \n",
" White | \n",
" 69.5 | \n",
" 73.6 | \n",
" 241 | \n",
" 121 | \n",
" 74 | \n",
" 3 | \n",
" 76.139005 | \n",
"
\n",
" \n",
" 13 | \n",
" Male | \n",
" White | \n",
" 20.5 | \n",
" 58.7 | \n",
" 309 | \n",
" 128 | \n",
" 71 | \n",
" 1 | \n",
" 56.083953 | \n",
"
\n",
" \n",
" 14 | \n",
" Female | \n",
" White | \n",
" 16.5 | \n",
" 58.7 | \n",
" 190 | \n",
" 196 | \n",
" 147 | \n",
" 3 | \n",
" 61.100016 | \n",
"
\n",
" \n",
" 15 | \n",
" Male | \n",
" White | \n",
" 19.8 | \n",
" 80.1 | \n",
" 204 | \n",
" 132 | \n",
" 125 | \n",
" 2 | \n",
" 85.021952 | \n",
"
\n",
" \n",
" 16 | \n",
" Male | \n",
" Black | \n",
" 15.7 | \n",
" 59.6 | \n",
" 203 | \n",
" 159 | \n",
" 52 | \n",
" 1 | \n",
" 60.947737 | \n",
"
\n",
" \n",
" 17 | \n",
" Female | \n",
" Black | \n",
" 15.7 | \n",
" 47.2 | \n",
" 196 | \n",
" 205 | \n",
" 148 | \n",
" 3 | \n",
" 49.128628 | \n",
"
\n",
" \n",
" 18 | \n",
" Female | \n",
" White | \n",
" 20.9 | \n",
" 48.6 | \n",
" 236 | \n",
" 180 | \n",
" 59 | \n",
" 1 | \n",
" 43.906114 | \n",
"
\n",
" \n",
" 19 | \n",
" Female | \n",
" White | \n",
" 21.3 | \n",
" 48.8 | \n",
" 212 | \n",
" 212 | \n",
" 139 | \n",
" 4 | \n",
" 52.640694 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" gender ethnicity age weight protein protein2 protein3 n_visits \\\n",
"0 Female White 72.0 76.0 246 88 136 8 \n",
"1 Female Black 84.1 59.8 210 85 86 6 \n",
"2 Female Black 79.7 56.0 205 91 110 7 \n",
"3 Female White 75.7 66.7 286 68 54 2 \n",
"4 Female White 74.6 72.1 171 81 99 8 \n",
"5 Female Black 70.5 76.4 204 56 112 5 \n",
"6 Male White 94.6 71.9 217 85 72 1 \n",
"7 Female Black 79.5 59.1 252 63 115 3 \n",
"8 Male Hispanic 87.5 60.5 240 59 120 7 \n",
"9 Female White 83.8 67.0 268 82 72 8 \n",
"10 Male White 74.7 70.2 270 102 133 0 \n",
"11 Female Black 84.6 65.5 269 47 103 4 \n",
"12 Male White 69.5 73.6 241 121 74 3 \n",
"13 Male White 20.5 58.7 309 128 71 1 \n",
"14 Female White 16.5 58.7 190 196 147 3 \n",
"15 Male White 19.8 80.1 204 132 125 2 \n",
"16 Male Black 15.7 59.6 203 159 52 1 \n",
"17 Female Black 15.7 47.2 196 205 148 3 \n",
"18 Female White 20.9 48.6 236 180 59 1 \n",
"19 Female White 21.3 48.8 212 212 139 4 \n",
"\n",
" X \n",
"0 75.252141 \n",
"1 55.366956 \n",
"2 53.413395 \n",
"3 65.672425 \n",
"4 67.774991 \n",
"5 77.983575 \n",
"6 74.539839 \n",
"7 56.244106 \n",
"8 57.310157 \n",
"9 66.779214 \n",
"10 69.724581 \n",
"11 67.474976 \n",
"12 76.139005 \n",
"13 56.083953 \n",
"14 61.100016 \n",
"15 85.021952 \n",
"16 60.947737 \n",
"17 49.128628 \n",
"18 43.906114 \n",
"19 52.640694 "
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"data.head(20)"
]
},
{
"cell_type": "code",
"execution_count": 5,
"id": "ad47ac4f",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" age | \n",
" weight | \n",
" protein | \n",
" protein2 | \n",
" protein3 | \n",
" n_visits | \n",
" X | \n",
"
\n",
" \n",
" \n",
" \n",
" count | \n",
" 817.000000 | \n",
" 817.000000 | \n",
" 817.000000 | \n",
" 817.000000 | \n",
" 817.000000 | \n",
" 817.000000 | \n",
" 817.000000 | \n",
"
\n",
" \n",
" mean | \n",
" 41.994002 | \n",
" 67.997307 | \n",
" 244.293758 | \n",
" 137.565483 | \n",
" 100.981640 | \n",
" 2.395349 | \n",
" 68.045217 | \n",
"
\n",
" \n",
" std | \n",
" 21.623043 | \n",
" 10.386467 | \n",
" 46.767645 | \n",
" 39.445960 | \n",
" 29.033465 | \n",
" 1.987492 | \n",
" 10.880718 | \n",
"
\n",
" \n",
" min | \n",
" 15.100000 | \n",
" 45.800000 | \n",
" 140.000000 | \n",
" 30.000000 | \n",
" 50.000000 | \n",
" 0.000000 | \n",
" 40.661973 | \n",
"
\n",
" \n",
" 25% | \n",
" 23.200000 | \n",
" 60.600000 | \n",
" 208.000000 | \n",
" 111.000000 | \n",
" 76.000000 | \n",
" 1.000000 | \n",
" 60.389961 | \n",
"
\n",
" \n",
" 50% | \n",
" 37.100000 | \n",
" 67.200000 | \n",
" 245.000000 | \n",
" 139.000000 | \n",
" 102.000000 | \n",
" 2.000000 | \n",
" 66.986613 | \n",
"
\n",
" \n",
" 75% | \n",
" 60.100000 | \n",
" 74.400000 | \n",
" 279.000000 | \n",
" 164.000000 | \n",
" 124.000000 | \n",
" 3.000000 | \n",
" 75.017939 | \n",
"
\n",
" \n",
" max | \n",
" 94.600000 | \n",
" 95.700000 | \n",
" 361.000000 | \n",
" 227.000000 | \n",
" 150.000000 | \n",
" 8.000000 | \n",
" 99.431199 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" age weight protein protein2 protein3 n_visits \\\n",
"count 817.000000 817.000000 817.000000 817.000000 817.000000 817.000000 \n",
"mean 41.994002 67.997307 244.293758 137.565483 100.981640 2.395349 \n",
"std 21.623043 10.386467 46.767645 39.445960 29.033465 1.987492 \n",
"min 15.100000 45.800000 140.000000 30.000000 50.000000 0.000000 \n",
"25% 23.200000 60.600000 208.000000 111.000000 76.000000 1.000000 \n",
"50% 37.100000 67.200000 245.000000 139.000000 102.000000 2.000000 \n",
"75% 60.100000 74.400000 279.000000 164.000000 124.000000 3.000000 \n",
"max 94.600000 95.700000 361.000000 227.000000 150.000000 8.000000 \n",
"\n",
" X \n",
"count 817.000000 \n",
"mean 68.045217 \n",
"std 10.880718 \n",
"min 40.661973 \n",
"25% 60.389961 \n",
"50% 66.986613 \n",
"75% 75.017939 \n",
"max 99.431199 "
]
},
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"data.describe()"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "9e1ce894",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"['gender', 'ethnicity', 'age', 'weight', 'protein', 'protein2', 'protein3', 'n_visits', 'X']\n"
]
}
],
"source": [
"#print(data.columns)\n",
"#Afficher les variables\n",
"colnames = list(data.columns)\n",
"print(colnames)"
]
},
{
"cell_type": "code",
"execution_count": 7,
"id": "81a33eef",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" | \n",
" weight | \n",
" protein | \n",
" protein2 | \n",
" protein3 | \n",
" n_visits | \n",
" X | \n",
"
\n",
" \n",
" gender | \n",
" age | \n",
" | \n",
" | \n",
" | \n",
" | \n",
" | \n",
" | \n",
"
\n",
" \n",
" \n",
" \n",
" Female | \n",
" 15.1 | \n",
" 51.800000 | \n",
" 189.000000 | \n",
" 200.5 | \n",
" 95.000000 | \n",
" 2.500000 | \n",
" 53.114162 | \n",
"
\n",
" \n",
" 15.2 | \n",
" 47.800000 | \n",
" 189.000000 | \n",
" 127.0 | \n",
" 72.000000 | \n",
" 3.000000 | \n",
" 46.340041 | \n",
"
\n",
" \n",
" 15.3 | \n",
" 56.900000 | \n",
" 165.500000 | \n",
" 186.5 | \n",
" 119.500000 | \n",
" 1.000000 | \n",
" 58.908039 | \n",
"
\n",
" \n",
" 15.4 | \n",
" 53.666667 | \n",
" 205.333333 | \n",
" 186.0 | \n",
" 83.000000 | \n",
" 1.666667 | \n",
" 54.578518 | \n",
"
\n",
" \n",
" 15.7 | \n",
" 50.200000 | \n",
" 184.333333 | \n",
" 188.0 | \n",
" 106.333333 | \n",
" 2.666667 | \n",
" 49.955718 | \n",
"
\n",
" \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
"
\n",
" \n",
" Male | \n",
" 89.2 | \n",
" 72.000000 | \n",
" 252.000000 | \n",
" 95.0 | \n",
" 68.000000 | \n",
" 1.000000 | \n",
" 70.480425 | \n",
"
\n",
" \n",
" 89.5 | \n",
" 72.800000 | \n",
" 208.000000 | \n",
" 77.0 | \n",
" 126.000000 | \n",
" 8.000000 | \n",
" 75.726986 | \n",
"
\n",
" \n",
" 91.8 | \n",
" 75.700000 | \n",
" 254.000000 | \n",
" 48.0 | \n",
" 54.000000 | \n",
" 7.000000 | \n",
" 74.487246 | \n",
"
\n",
" \n",
" 93.3 | \n",
" 53.200000 | \n",
" 172.000000 | \n",
" 58.0 | \n",
" 107.000000 | \n",
" 0.000000 | \n",
" 54.101774 | \n",
"
\n",
" \n",
" 94.6 | \n",
" 71.900000 | \n",
" 217.000000 | \n",
" 85.0 | \n",
" 72.000000 | \n",
" 1.000000 | \n",
" 74.539839 | \n",
"
\n",
" \n",
"
\n",
"
563 rows × 6 columns
\n",
"
"
],
"text/plain": [
" weight protein protein2 protein3 n_visits X\n",
"gender age \n",
"Female 15.1 51.800000 189.000000 200.5 95.000000 2.500000 53.114162\n",
" 15.2 47.800000 189.000000 127.0 72.000000 3.000000 46.340041\n",
" 15.3 56.900000 165.500000 186.5 119.500000 1.000000 58.908039\n",
" 15.4 53.666667 205.333333 186.0 83.000000 1.666667 54.578518\n",
" 15.7 50.200000 184.333333 188.0 106.333333 2.666667 49.955718\n",
"... ... ... ... ... ... ...\n",
"Male 89.2 72.000000 252.000000 95.0 68.000000 1.000000 70.480425\n",
" 89.5 72.800000 208.000000 77.0 126.000000 8.000000 75.726986\n",
" 91.8 75.700000 254.000000 48.0 54.000000 7.000000 74.487246\n",
" 93.3 53.200000 172.000000 58.0 107.000000 0.000000 54.101774\n",
" 94.6 71.900000 217.000000 85.0 72.000000 1.000000 74.539839\n",
"\n",
"[563 rows x 6 columns]"
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# regrouper une variable suivant la moyenne \n",
"data.groupby(['age']).max()\n",
"data.groupby(['gender','age']).mean()# regrouper deux variables suivant la moyenne "
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "daa82a2b",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" gender | \n",
" ethnicity | \n",
" age | \n",
" weight | \n",
" protein | \n",
" protein2 | \n",
" protein3 | \n",
" n_visits | \n",
" X | \n",
"
\n",
" \n",
" \n",
" \n",
" 14 | \n",
" Female | \n",
" White | \n",
" 16.5 | \n",
" 58.7 | \n",
" 190 | \n",
" 196 | \n",
" 147 | \n",
" 3 | \n",
" 61.100016 | \n",
"
\n",
" \n",
" 16 | \n",
" Male | \n",
" Black | \n",
" 15.7 | \n",
" 59.6 | \n",
" 203 | \n",
" 159 | \n",
" 52 | \n",
" 1 | \n",
" 60.947737 | \n",
"
\n",
" \n",
" 17 | \n",
" Female | \n",
" Black | \n",
" 15.7 | \n",
" 47.2 | \n",
" 196 | \n",
" 205 | \n",
" 148 | \n",
" 3 | \n",
" 49.128628 | \n",
"
\n",
" \n",
" 20 | \n",
" Female | \n",
" White | \n",
" 16.4 | \n",
" 51.7 | \n",
" 184 | \n",
" 225 | \n",
" 106 | \n",
" 1 | \n",
" 47.921351 | \n",
"
\n",
" \n",
" 22 | \n",
" Female | \n",
" White | \n",
" 15.4 | \n",
" 46.0 | \n",
" 169 | \n",
" 227 | \n",
" 53 | \n",
" 1 | \n",
" 48.368520 | \n",
"
\n",
" \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
"
\n",
" \n",
" 470 | \n",
" Male | \n",
" White | \n",
" 16.6 | \n",
" 56.4 | \n",
" 292 | \n",
" 157 | \n",
" 88 | \n",
" 2 | \n",
" 53.491081 | \n",
"
\n",
" \n",
" 478 | \n",
" Male | \n",
" White | \n",
" 17.4 | \n",
" 57.1 | \n",
" 280 | \n",
" 134 | \n",
" 107 | \n",
" 3 | \n",
" 53.628482 | \n",
"
\n",
" \n",
" 483 | \n",
" Male | \n",
" White | \n",
" 17.4 | \n",
" 68.1 | \n",
" 288 | \n",
" 135 | \n",
" 126 | \n",
" 2 | \n",
" 66.857853 | \n",
"
\n",
" \n",
" 484 | \n",
" Female | \n",
" White | \n",
" 16.8 | \n",
" 53.9 | \n",
" 193 | \n",
" 163 | \n",
" 117 | \n",
" 3 | \n",
" 55.625794 | \n",
"
\n",
" \n",
" 486 | \n",
" Female | \n",
" White | \n",
" 15.9 | \n",
" 47.3 | \n",
" 174 | \n",
" 198 | \n",
" 84 | \n",
" 2 | \n",
" 51.745720 | \n",
"
\n",
" \n",
"
\n",
"
103 rows × 9 columns
\n",
"
"
],
"text/plain": [
" gender ethnicity age weight protein protein2 protein3 n_visits \\\n",
"14 Female White 16.5 58.7 190 196 147 3 \n",
"16 Male Black 15.7 59.6 203 159 52 1 \n",
"17 Female Black 15.7 47.2 196 205 148 3 \n",
"20 Female White 16.4 51.7 184 225 106 1 \n",
"22 Female White 15.4 46.0 169 227 53 1 \n",
".. ... ... ... ... ... ... ... ... \n",
"470 Male White 16.6 56.4 292 157 88 2 \n",
"478 Male White 17.4 57.1 280 134 107 3 \n",
"483 Male White 17.4 68.1 288 135 126 2 \n",
"484 Female White 16.8 53.9 193 163 117 3 \n",
"486 Female White 15.9 47.3 174 198 84 2 \n",
"\n",
" X \n",
"14 61.100016 \n",
"16 60.947737 \n",
"17 49.128628 \n",
"20 47.921351 \n",
"22 48.368520 \n",
".. ... \n",
"470 53.491081 \n",
"478 53.628482 \n",
"483 66.857853 \n",
"484 55.625794 \n",
"486 51.745720 \n",
"\n",
"[103 rows x 9 columns]"
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"data[data['age']<18]#Afficher seulement les mineurs"
]
},
{
"cell_type": "code",
"execution_count": 9,
"id": "3de9ad09",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"Female 56\n",
"Male 47\n",
"Name: gender, dtype: int64"
]
},
"execution_count": 9,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
" data[data['age']<18]['gender'].value_counts()#Afficher somme des mineurs suivant le sex"
]
},
{
"cell_type": "code",
"execution_count": 10,
"id": "8c76ae68",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"gender False\n",
"ethnicity False\n",
"age False\n",
"weight False\n",
"protein False\n",
"protein2 False\n",
"protein3 False\n",
"n_visits False\n",
"X False\n",
"dtype: bool\n"
]
},
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" gender | \n",
" ethnicity | \n",
" age | \n",
" weight | \n",
" protein | \n",
" protein2 | \n",
" protein3 | \n",
" n_visits | \n",
" X | \n",
"
\n",
" \n",
" \n",
" \n",
" 0 | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
"
\n",
" \n",
" 1 | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
"
\n",
" \n",
" 2 | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
"
\n",
" \n",
" 3 | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
"
\n",
" \n",
" 4 | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
"
\n",
" \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
" ... | \n",
"
\n",
" \n",
" 812 | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
"
\n",
" \n",
" 813 | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
"
\n",
" \n",
" 814 | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
"
\n",
" \n",
" 815 | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
"
\n",
" \n",
" 816 | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
" False | \n",
"
\n",
" \n",
"
\n",
"
817 rows × 9 columns
\n",
"
"
],
"text/plain": [
" gender ethnicity age weight protein protein2 protein3 n_visits \\\n",
"0 False False False False False False False False \n",
"1 False False False False False False False False \n",
"2 False False False False False False False False \n",
"3 False False False False False False False False \n",
"4 False False False False False False False False \n",
".. ... ... ... ... ... ... ... ... \n",
"812 False False False False False False False False \n",
"813 False False False False False False False False \n",
"814 False False False False False False False False \n",
"815 False False False False False False False False \n",
"816 False False False False False False False False \n",
"\n",
" X \n",
"0 False \n",
"1 False \n",
"2 False \n",
"3 False \n",
"4 False \n",
".. ... \n",
"812 False \n",
"813 False \n",
"814 False \n",
"815 False \n",
"816 False \n",
"\n",
"[817 rows x 9 columns]"
]
},
"execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"#Vérifier qu’il n’y a pas de valeurs manquantes dans le DataFrame\n",
"print(data.isna().any())\n",
"#ou bien \n",
"data.isna()"
]
},
{
"cell_type": "code",
"execution_count": 11,
"id": "7b9c4c21",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Observations n= 817\n",
"Variables= 9\n"
]
}
],
"source": [
"#2. Quel est le nombre de variables n et le nombre d’observations (dimension du DataFrame) p?\n",
"[n,p]=data.shape;\n",
"print(\"Observations n=\",n)\n",
"print('Variables=',p)"
]
},
{
"cell_type": "code",
"execution_count": 17,
"id": "744c037b",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"['gender',\n",
" 'ethnicity',\n",
" 'age',\n",
" 'weight',\n",
" 'protein',\n",
" 'protein2',\n",
" 'protein3',\n",
" 'n_visits',\n",
" 'X']"
]
},
"execution_count": 17,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"list(data.columns)"
]
},
{
"cell_type": "code",
"execution_count": 18,
"id": "325a2b67",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"0 Female\n",
"1 Female\n",
"2 Female\n",
"3 Female\n",
"4 Female\n",
" ... \n",
"812 Male\n",
"813 Female\n",
"814 Male\n",
"815 Male\n",
"816 Female\n",
"Name: gender, Length: 817, dtype: object 0 White\n",
"1 Black\n",
"2 Black\n",
"3 White\n",
"4 White\n",
" ... \n",
"812 White\n",
"813 Hispanic\n",
"814 White\n",
"815 Unknown\n",
"816 White\n",
"Name: ethnicity, Length: 817, dtype: object\n"
]
}
],
"source": [
"print(data['gender'],data['ethnicity'])"
]
},
{
"cell_type": "code",
"execution_count": 19,
"id": "e9d07139",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n",
"RangeIndex: 817 entries, 0 to 816\n",
"Data columns (total 9 columns):\n",
" # Column Non-Null Count Dtype \n",
"--- ------ -------------- ----- \n",
" 0 gender 817 non-null category\n",
" 1 ethnicity 817 non-null category\n",
" 2 age 817 non-null float64 \n",
" 3 weight 817 non-null float64 \n",
" 4 protein 817 non-null int64 \n",
" 5 protein2 817 non-null int64 \n",
" 6 protein3 817 non-null int64 \n",
" 7 n_visits 817 non-null int64 \n",
" 8 X 817 non-null float64 \n",
"dtypes: category(2), float64(3), int64(4)\n",
"memory usage: 46.9 KB\n"
]
}
],
"source": [
"data.gender = pd.Categorical(data.gender,categories=['Female','Male']) #data.gender.astype('category') # ou bien pd.Categorical(data.gender) \n",
"data.ethnicity = data.ethnicity.astype('category') \n",
"data.info()"
]
},
{
"cell_type": "code",
"execution_count": 20,
"id": "cd6cf64a",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" age | \n",
" weight | \n",
" protein | \n",
" protein2 | \n",
" protein3 | \n",
" n_visits | \n",
" X | \n",
"
\n",
" \n",
" \n",
" \n",
" count | \n",
" 817.00 | \n",
" 817.00 | \n",
" 817.00 | \n",
" 817.00 | \n",
" 817.00 | \n",
" 817.00 | \n",
" 817.00 | \n",
"
\n",
" \n",
" mean | \n",
" 41.99 | \n",
" 68.00 | \n",
" 244.29 | \n",
" 137.57 | \n",
" 100.98 | \n",
" 2.40 | \n",
" 68.05 | \n",
"
\n",
" \n",
" std | \n",
" 21.62 | \n",
" 10.39 | \n",
" 46.77 | \n",
" 39.45 | \n",
" 29.03 | \n",
" 1.99 | \n",
" 10.88 | \n",
"
\n",
" \n",
" min | \n",
" 15.10 | \n",
" 45.80 | \n",
" 140.00 | \n",
" 30.00 | \n",
" 50.00 | \n",
" 0.00 | \n",
" 40.66 | \n",
"
\n",
" \n",
" 25% | \n",
" 23.20 | \n",
" 60.60 | \n",
" 208.00 | \n",
" 111.00 | \n",
" 76.00 | \n",
" 1.00 | \n",
" 60.39 | \n",
"
\n",
" \n",
" 50% | \n",
" 37.10 | \n",
" 67.20 | \n",
" 245.00 | \n",
" 139.00 | \n",
" 102.00 | \n",
" 2.00 | \n",
" 66.99 | \n",
"
\n",
" \n",
" 75% | \n",
" 60.10 | \n",
" 74.40 | \n",
" 279.00 | \n",
" 164.00 | \n",
" 124.00 | \n",
" 3.00 | \n",
" 75.02 | \n",
"
\n",
" \n",
" max | \n",
" 94.60 | \n",
" 95.70 | \n",
" 361.00 | \n",
" 227.00 | \n",
" 150.00 | \n",
" 8.00 | \n",
" 99.43 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" age weight protein protein2 protein3 n_visits X\n",
"count 817.00 817.00 817.00 817.00 817.00 817.00 817.00\n",
"mean 41.99 68.00 244.29 137.57 100.98 2.40 68.05\n",
"std 21.62 10.39 46.77 39.45 29.03 1.99 10.88\n",
"min 15.10 45.80 140.00 30.00 50.00 0.00 40.66\n",
"25% 23.20 60.60 208.00 111.00 76.00 1.00 60.39\n",
"50% 37.10 67.20 245.00 139.00 102.00 2.00 66.99\n",
"75% 60.10 74.40 279.00 164.00 124.00 3.00 75.02\n",
"max 94.60 95.70 361.00 227.00 150.00 8.00 99.43"
]
},
"execution_count": 20,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"#3.Description des données pour les variables quantitatifs\n",
"data.describe().round(2)"
]
},
{
"cell_type": "code",
"execution_count": 21,
"id": "6ce6e8a9",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 21,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"#Tracer les boites a moustaches des variables quantitatives (\"gender et ethnicity\")\n",
"sns.catplot(y = 'age', x = 'ethnicity', kind ='box', data = data)\n",
"sns.catplot(y = 'age', x = 'gender', kind ='box', data = data)"
]
},
{
"cell_type": "code",
"execution_count": 22,
"id": "8ba49163",
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"#Une représentation graphique, de votre choix, de cettes dernières variables.\n",
"labels = 'Female', 'Male'\n",
"sizes = [np.sum(data.gender==\"Male\"), np.sum(data.gender==\"Female\")]\n",
"explode = (0, 0.1,) # only \"explode\" the 2nd slice (i.e. 'Hogs')\n",
"fig1, ax1 = plt.subplots()\n",
"ax1.pie(sizes, explode=explode, labels=labels, autopct='%1.1f%%', shadow=True, startangle=90)\n",
"ax1.axis('equal') # Equal aspect ratio ensures that pie is drawn as a circle.\n",
"plt.show()\n"
]
},
{
"cell_type": "code",
"execution_count": 23,
"id": "60d63c0d",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" ni | \n",
" fi | \n",
" Fi | \n",
"
\n",
" \n",
" \n",
" \n",
" Asian | \n",
" 27 | \n",
" 0.033048 | \n",
" 0.033048 | \n",
"
\n",
" \n",
" Black | \n",
" 88 | \n",
" 0.107711 | \n",
" 0.140759 | \n",
"
\n",
" \n",
" Dominican | \n",
" 1 | \n",
" 0.001224 | \n",
" 0.141983 | \n",
"
\n",
" \n",
" Hispanic | \n",
" 75 | \n",
" 0.091799 | \n",
" 0.233782 | \n",
"
\n",
" \n",
" Indian | \n",
" 1 | \n",
" 0.001224 | \n",
" 0.235006 | \n",
"
\n",
" \n",
" Other | \n",
" 6 | \n",
" 0.007344 | \n",
" 0.242350 | \n",
"
\n",
" \n",
" Unknown | \n",
" 32 | \n",
" 0.039168 | \n",
" 0.281518 | \n",
"
\n",
" \n",
" White | \n",
" 587 | \n",
" 0.718482 | \n",
" 1.000000 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" ni fi Fi\n",
"Asian 27 0.033048 0.033048\n",
"Black 88 0.107711 0.140759\n",
"Dominican 1 0.001224 0.141983\n",
"Hispanic 75 0.091799 0.233782\n",
"Indian 1 0.001224 0.235006\n",
"Other 6 0.007344 0.242350\n",
"Unknown 32 0.039168 0.281518\n",
"White 587 0.718482 1.000000"
]
},
"execution_count": 23,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"xi, ni = np.unique(data.ethnicity, return_counts=True)\n",
"table=pd.DataFrame( data=ni, columns=[\"ni\"],index=xi)\n",
"N=sum(table.ni)\n",
"table.insert(1, \"fi\", table.ni / N, True)\n",
"table.insert(2,\"Fi\",np.cumsum(table.fi), True)\n",
"table"
]
},
{
"cell_type": "code",
"execution_count": 215,
"id": "0838ba33",
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"labels = 'Asian', 'Black', 'Dominician', 'Hispanic','Indian', 'other','Unknown', 'White'\n",
"sizes = [np.sum(data.ethnicity==\"Indian\"),np.sum(data.ethnicity==\"White\"), np.sum(data.ethnicity==\"Dominican\"),np.sum(data.ethnicity==\"Hispanic\"), np.sum(data.ethnicity==\"Other\"), np.sum(data.ethnicity==\"Unknown\"),np.sum(data.ethnicity==\"Black\"),np.sum(data.ethnicity==\"Asian\")]\n",
"#explode = (0, 0.1,) # only \"explode\" the 2nd slice (i.e. 'Hogs')\n",
"fig1, ax1 = plt.subplots()\n",
"ax1.pie(sizes, labels=labels, autopct='%1.1f%%', shadow=True,startangle=90)\n",
"ax1.axis('equal') # Equal aspect ratio ensures that pie is drawn as a circle.\n",
"plt.show()"
]
},
{
"cell_type": "code",
"execution_count": 24,
"id": "0514a3d5",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" ni | \n",
" fi | \n",
" Fi | \n",
"
\n",
" \n",
" \n",
" \n",
" 0 | \n",
" 154 | \n",
" 0.188494 | \n",
" 0.188494 | \n",
"
\n",
" \n",
" 1 | \n",
" 160 | \n",
" 0.195838 | \n",
" 0.384333 | \n",
"
\n",
" \n",
" 2 | \n",
" 145 | \n",
" 0.177479 | \n",
" 0.561812 | \n",
"
\n",
" \n",
" 3 | \n",
" 156 | \n",
" 0.190942 | \n",
" 0.752754 | \n",
"
\n",
" \n",
" 4 | \n",
" 111 | \n",
" 0.135863 | \n",
" 0.888617 | \n",
"
\n",
" \n",
" 5 | \n",
" 18 | \n",
" 0.022032 | \n",
" 0.910649 | \n",
"
\n",
" \n",
" 6 | \n",
" 27 | \n",
" 0.033048 | \n",
" 0.943696 | \n",
"
\n",
" \n",
" 7 | \n",
" 25 | \n",
" 0.030600 | \n",
" 0.974296 | \n",
"
\n",
" \n",
" 8 | \n",
" 21 | \n",
" 0.025704 | \n",
" 1.000000 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" ni fi Fi\n",
"0 154 0.188494 0.188494\n",
"1 160 0.195838 0.384333\n",
"2 145 0.177479 0.561812\n",
"3 156 0.190942 0.752754\n",
"4 111 0.135863 0.888617\n",
"5 18 0.022032 0.910649\n",
"6 27 0.033048 0.943696\n",
"7 25 0.030600 0.974296\n",
"8 21 0.025704 1.000000"
]
},
"execution_count": 24,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"#4 variable n_visits\n",
"xi, ni = np.unique(data.n_visits, return_counts=True)\n",
"table=pd.DataFrame( data=ni, columns=[\"ni\"],index=xi)\n",
"N=sum(table.ni)\n",
"table.insert(1, \"fi\", table.ni / N, True)\n",
"table.insert(2,\"Fi\",np.cumsum(table.fi), True)\n",
"table"
]
},
{
"cell_type": "code",
"execution_count": 54,
"id": "78c34143",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"1\n",
"Dans la plupart de temps, il y une seule visite pour un patient au cours des 24 derniers mois\n"
]
}
],
"source": [
"#5. Le mode ou valeur dominante \n",
"print(ni.argmax())\n",
"ni.max()\n",
"print(\"Dans la plupart de temps, il y une seule visite pour un patient au cours des 24 derniers mois\")"
]
},
{
"cell_type": "markdown",
"id": "7e7511d7",
"metadata": {},
"source": [
"## Partie 2"
]
},
{
"cell_type": "code",
"execution_count": 35,
"id": "bd41a69c",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"Text(0.5, 1.0, 'Nuage de points')"
]
},
"execution_count": 35,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"