PSante_02_notebookexploration.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "oNEsWsBl6ncs"
   },
   "source": [
    "# Import des librairies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "id": "RDTKaUY-6LVY"
   },
   "outputs": [],
   "source": [
    "# fonctions statistiques\n",
    "import scipy.stats as st\n",
    "\n",
    "# algèbre linéaire\n",
    "import numpy as np \n",
    "\n",
    "# traitement des données\n",
    "import pandas as pd\n",
    "\n",
    "# graphiques et visualisation des données\n",
    "import seaborn as sns\n",
    "sns.set_theme(style=\"whitegrid\") # style graphique des visualisations\n",
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# identification des types de données\n",
    "from pandas.api.types import is_string_dtype, is_numeric_dtype\n",
    "\n",
    "# imputation k-NN\n",
    "import librairies.knn_impute as knn"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "xVsSaoXN7au7"
   },
   "source": [
    "# Chargement des données"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "path = 'data/openfoodfacts_data_for_analysis.csv'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "PVa0j7gk9VEB",
    "outputId": "e98b309b-810c-45e3-9b18-e68c4a862136"
   },
   "outputs": [],
   "source": [
    "open_food_facts_df = pd.read_csv(path, sep=\"\\t\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Exploration préliminaire des données"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 728
    },
    "id": "N1nxUijk9jgU",
    "outputId": "ffd18d09-b4a6-4c34-bdd1-cb84559783c7"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 104300 entries, 0 to 104299\n",
      "Data columns (total 10 columns):\n",
      "Unnamed: 0          104300 non-null int64\n",
      "brands_tags         103844 non-null object\n",
      "labels_en           71866 non-null object\n",
      "additives_tags      73811 non-null object\n",
      "nutriscore_grade    104300 non-null object\n",
      "pnns_groups_1       104159 non-null object\n",
      "pnns_groups_2       104300 non-null object\n",
      "additives_n         104300 non-null float64\n",
      "nutriscore_score    104300 non-null float64\n",
      "nova_group          104067 non-null float64\n",
      "dtypes: float64(3), int64(1), object(6)\n",
      "memory usage: 8.0+ MB\n"
     ]
    }
   ],
   "source": [
    "open_food_facts_df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "L'index apparaît comme une variable dans le jeu de données exporté. Il sera supprimé car un nouvel index est créé lors du chargement."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "open_food_facts_df.drop(columns=['Unnamed: 0'], inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "dxJtHnYpAX94",
    "outputId": "42f9a8ab-8132-4bf8-bef9-c058133879db"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 104300 entries, 0 to 104299\n",
      "Data columns (total 9 columns):\n",
      "brands_tags         103844 non-null object\n",
      "labels_en           71866 non-null object\n",
      "additives_tags      73811 non-null object\n",
      "nutriscore_grade    104300 non-null object\n",
      "pnns_groups_1       104159 non-null object\n",
      "pnns_groups_2       104300 non-null object\n",
      "additives_n         104300 non-null float64\n",
      "nutriscore_score    104300 non-null float64\n",
      "nova_group          104067 non-null float64\n",
      "dtypes: float64(3), object(6)\n",
      "memory usage: 45.1 MB\n"
     ]
    }
   ],
   "source": [
    "open_food_facts_df.info(memory_usage='deep')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "id": "ZDHUmQgCCWnX"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>brands_tags</th>\n",
       "      <th>labels_en</th>\n",
       "      <th>additives_tags</th>\n",
       "      <th>nutriscore_grade</th>\n",
       "      <th>pnns_groups_1</th>\n",
       "      <th>pnns_groups_2</th>\n",
       "      <th>additives_n</th>\n",
       "      <th>nutriscore_score</th>\n",
       "      <th>nova_group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>count</td>\n",
       "      <td>103844</td>\n",
       "      <td>71866</td>\n",
       "      <td>73811</td>\n",
       "      <td>104300</td>\n",
       "      <td>104159</td>\n",
       "      <td>104300</td>\n",
       "      <td>104300.000000</td>\n",
       "      <td>104300.000000</td>\n",
       "      <td>104067.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>unique</td>\n",
       "      <td>17089</td>\n",
       "      <td>21046</td>\n",
       "      <td>19093</td>\n",
       "      <td>5</td>\n",
       "      <td>13</td>\n",
       "      <td>45</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>top</td>\n",
       "      <td>u</td>\n",
       "      <td>Green Dot</td>\n",
       "      <td>en:e322,en:e322i</td>\n",
       "      <td>d</td>\n",
       "      <td>Sugary snacks</td>\n",
       "      <td>Biscuits and cakes</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>freq</td>\n",
       "      <td>3576</td>\n",
       "      <td>4116</td>\n",
       "      <td>3834</td>\n",
       "      <td>30566</td>\n",
       "      <td>22711</td>\n",
       "      <td>11070</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>mean</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.140489</td>\n",
       "      <td>9.488926</td>\n",
       "      <td>3.512872</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>std</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.476332</td>\n",
       "      <td>8.933405</td>\n",
       "      <td>0.875439</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>min</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-14.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>25%</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>3.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>50%</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>4.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>75%</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>16.000000</td>\n",
       "      <td>4.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>max</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>30.000000</td>\n",
       "      <td>40.000000</td>\n",
       "      <td>4.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       brands_tags  labels_en    additives_tags nutriscore_grade  \\\n",
       "count       103844      71866             73811           104300   \n",
       "unique       17089      21046             19093                5   \n",
       "top              u  Green Dot  en:e322,en:e322i                d   \n",
       "freq          3576       4116              3834            30566   \n",
       "mean           NaN        NaN               NaN              NaN   \n",
       "std            NaN        NaN               NaN              NaN   \n",
       "min            NaN        NaN               NaN              NaN   \n",
       "25%            NaN        NaN               NaN              NaN   \n",
       "50%            NaN        NaN               NaN              NaN   \n",
       "75%            NaN        NaN               NaN              NaN   \n",
       "max            NaN        NaN               NaN              NaN   \n",
       "\n",
       "        pnns_groups_1       pnns_groups_2    additives_n  nutriscore_score  \\\n",
       "count          104159              104300  104300.000000     104300.000000   \n",
       "unique             13                  45            NaN               NaN   \n",
       "top     Sugary snacks  Biscuits and cakes            NaN               NaN   \n",
       "freq            22711               11070            NaN               NaN   \n",
       "mean              NaN                 NaN       2.140489          9.488926   \n",
       "std               NaN                 NaN       2.476332          8.933405   \n",
       "min               NaN                 NaN       0.000000        -14.000000   \n",
       "25%               NaN                 NaN       0.000000          2.000000   \n",
       "50%               NaN                 NaN       1.000000         10.000000   \n",
       "75%               NaN                 NaN       3.000000         16.000000   \n",
       "max               NaN                 NaN      30.000000         40.000000   \n",
       "\n",
       "           nova_group  \n",
       "count   104067.000000  \n",
       "unique            NaN  \n",
       "top               NaN  \n",
       "freq              NaN  \n",
       "mean         3.512872  \n",
       "std          0.875439  \n",
       "min          1.000000  \n",
       "25%          3.000000  \n",
       "50%          4.000000  \n",
       "75%          4.000000  \n",
       "max          4.000000  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df.describe(include='all')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Le jeu de données contient des données manquantes parmi les variables *brands_tags*, *labels_en*, *additives_tags*, *pnns_groups_1* et *nova_group*.\n",
    "\n",
    "Nous allons les visualiser ci-dessus :"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>nombre</th>\n",
       "      <th>total</th>\n",
       "      <th>pourcentage</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>labels_en</td>\n",
       "      <td>32434</td>\n",
       "      <td>104300</td>\n",
       "      <td>31.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>additives_tags</td>\n",
       "      <td>30489</td>\n",
       "      <td>104300</td>\n",
       "      <td>29.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>brands_tags</td>\n",
       "      <td>456</td>\n",
       "      <td>104300</td>\n",
       "      <td>0.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>nova_group</td>\n",
       "      <td>233</td>\n",
       "      <td>104300</td>\n",
       "      <td>0.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>pnns_groups_1</td>\n",
       "      <td>141</td>\n",
       "      <td>104300</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>nutriscore_grade</td>\n",
       "      <td>0</td>\n",
       "      <td>104300</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>pnns_groups_2</td>\n",
       "      <td>0</td>\n",
       "      <td>104300</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>additives_n</td>\n",
       "      <td>0</td>\n",
       "      <td>104300</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>nutriscore_score</td>\n",
       "      <td>0</td>\n",
       "      <td>104300</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  nombre   total  pourcentage\n",
       "labels_en          32434  104300         31.1\n",
       "additives_tags     30489  104300         29.2\n",
       "brands_tags          456  104300          0.4\n",
       "nova_group           233  104300          0.2\n",
       "pnns_groups_1        141  104300          0.1\n",
       "nutriscore_grade       0  104300          0.0\n",
       "pnns_groups_2          0  104300          0.0\n",
       "additives_n            0  104300          0.0\n",
       "nutriscore_score       0  104300          0.0"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nombres_donnees_manquantes = open_food_facts_df.isnull().sum()\n",
    "nombres_donnees_totales = open_food_facts_df.isnull().count()\n",
    "pourcentages_donnees_manquantes = round(nombres_donnees_manquantes/nombres_donnees_totales*100, 1)\n",
    "\n",
    "donnees_manquantes_df = pd.DataFrame({'nombre': nombres_donnees_manquantes, \n",
    "                                      'total': nombres_donnees_totales, \n",
    "                                      'pourcentage': pourcentages_donnees_manquantes})\n",
    "\n",
    "donnees_manquantes_df.sort_values(by=['nombre'], ascending=False, inplace=True)\n",
    "\n",
    "donnees_manquantes_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "donnees_manquantes_df = donnees_manquantes_df[donnees_manquantes_df['nombre'] > 0]\n",
    "\n",
    "plt.rcParams[\"figure.figsize\"] = (15, 6)\n",
    "sns.barplot(x=donnees_manquantes_df.index.values, y='pourcentage', data=donnees_manquantes_df, color='tab:blue')\n",
    "plt.title('Pourcentage de données manquantes par variable')\n",
    "plt.ylabel('Pourcentage de données manquantes')\n",
    "plt.xlabel('Variables')\n",
    "\n",
    "for variable, pourcentage in enumerate(donnees_manquantes_df.pourcentage):\n",
    "    plt.text(variable, pourcentage, str(pourcentage)+\"\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Les variables contenant une plus grande proportion de données manquantes sont *labels_en* et *additives_tags*, avec chacune 31.1% et 29.2% de données manquantes.\n",
    "\n",
    "Il faudra porter une attention particulière lors de l'imputation de ces variables car la nouvelle valeur représentera environ 30% des données."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>brands_tags</th>\n",
       "      <th>labels_en</th>\n",
       "      <th>additives_tags</th>\n",
       "      <th>nutriscore_grade</th>\n",
       "      <th>pnns_groups_1</th>\n",
       "      <th>pnns_groups_2</th>\n",
       "      <th>additives_n</th>\n",
       "      <th>nutriscore_score</th>\n",
       "      <th>nova_group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>crousresto</td>\n",
       "      <td>NaN</td>\n",
       "      <td>en:e14xx,en:e262,en:e262i,en:e326</td>\n",
       "      <td>a</td>\n",
       "      <td>Cereals and potatoes</td>\n",
       "      <td>Bread</td>\n",
       "      <td>3.0</td>\n",
       "      <td>-4.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>comme-j-aime</td>\n",
       "      <td>fr:viande-francaise,Made in France</td>\n",
       "      <td>en:e14xx,en:e451,en:e451i</td>\n",
       "      <td>b</td>\n",
       "      <td>Composite foods</td>\n",
       "      <td>One-dish meals</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>comme-j-aime</td>\n",
       "      <td>Green Dot</td>\n",
       "      <td>en:e102,en:e110,en:e407,en:e471</td>\n",
       "      <td>b</td>\n",
       "      <td>Milk and dairy products</td>\n",
       "      <td>Dairy desserts</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>mx3</td>\n",
       "      <td>Green Dot</td>\n",
       "      <td>en:e322,en:e322i</td>\n",
       "      <td>b</td>\n",
       "      <td>Fruits and vegetables</td>\n",
       "      <td>Fruits</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>wonka,nerds</td>\n",
       "      <td>No artificial flavors,Contains GMOs</td>\n",
       "      <td>en:e102,en:e110,en:e129,en:e132,en:e133,en:e29...</td>\n",
       "      <td>d</td>\n",
       "      <td>Sugary snacks</td>\n",
       "      <td>Sweets</td>\n",
       "      <td>7.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>alice-delice</td>\n",
       "      <td>NaN</td>\n",
       "      <td>en:e171,en:e339ii,en:e420,en:e420i,en:e422,en:...</td>\n",
       "      <td>d</td>\n",
       "      <td>unknown</td>\n",
       "      <td>unknown</td>\n",
       "      <td>5.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>alice-delice</td>\n",
       "      <td>Gluten-free</td>\n",
       "      <td>en:e330,en:e415,en:e420,en:e420ii,en:e422,en:e471</td>\n",
       "      <td>d</td>\n",
       "      <td>Sugary snacks</td>\n",
       "      <td>Sweets</td>\n",
       "      <td>5.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>vichy</td>\n",
       "      <td>Low or no sugar,Contains a source of phenylala...</td>\n",
       "      <td>en:e420,en:e951</td>\n",
       "      <td>c</td>\n",
       "      <td>Sugary snacks</td>\n",
       "      <td>Sweets</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>m-m-s,mars,mars-chocolat</td>\n",
       "      <td>Contains GMOs</td>\n",
       "      <td>en:e414</td>\n",
       "      <td>e</td>\n",
       "      <td>Sugary snacks</td>\n",
       "      <td>Sweets</td>\n",
       "      <td>1.0</td>\n",
       "      <td>36.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>pepsi</td>\n",
       "      <td>NaN</td>\n",
       "      <td>en:e150d,en:e338</td>\n",
       "      <td>e</td>\n",
       "      <td>Beverages</td>\n",
       "      <td>Sweetened beverages</td>\n",
       "      <td>2.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10</td>\n",
       "      <td>heinz</td>\n",
       "      <td>Vegetarian,No artificial colors,No flavors,No ...</td>\n",
       "      <td>en:e101,en:e101i</td>\n",
       "      <td>d</td>\n",
       "      <td>Fat and sauces</td>\n",
       "      <td>Dressings and sauces</td>\n",
       "      <td>1.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11</td>\n",
       "      <td>coca-cola</td>\n",
       "      <td>NaN</td>\n",
       "      <td>en:e338</td>\n",
       "      <td>e</td>\n",
       "      <td>Beverages</td>\n",
       "      <td>Sweetened beverages</td>\n",
       "      <td>1.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 brands_tags  \\\n",
       "0                 crousresto   \n",
       "1               comme-j-aime   \n",
       "2               comme-j-aime   \n",
       "3                        mx3   \n",
       "4                wonka,nerds   \n",
       "5               alice-delice   \n",
       "6               alice-delice   \n",
       "7                      vichy   \n",
       "8   m-m-s,mars,mars-chocolat   \n",
       "9                      pepsi   \n",
       "10                     heinz   \n",
       "11                 coca-cola   \n",
       "\n",
       "                                            labels_en  \\\n",
       "0                                                 NaN   \n",
       "1                  fr:viande-francaise,Made in France   \n",
       "2                                           Green Dot   \n",
       "3                                           Green Dot   \n",
       "4                 No artificial flavors,Contains GMOs   \n",
       "5                                                 NaN   \n",
       "6                                         Gluten-free   \n",
       "7   Low or no sugar,Contains a source of phenylala...   \n",
       "8                                       Contains GMOs   \n",
       "9                                                 NaN   \n",
       "10  Vegetarian,No artificial colors,No flavors,No ...   \n",
       "11                                                NaN   \n",
       "\n",
       "                                       additives_tags nutriscore_grade  \\\n",
       "0                   en:e14xx,en:e262,en:e262i,en:e326                a   \n",
       "1                           en:e14xx,en:e451,en:e451i                b   \n",
       "2                     en:e102,en:e110,en:e407,en:e471                b   \n",
       "3                                    en:e322,en:e322i                b   \n",
       "4   en:e102,en:e110,en:e129,en:e132,en:e133,en:e29...                d   \n",
       "5   en:e171,en:e339ii,en:e420,en:e420i,en:e422,en:...                d   \n",
       "6   en:e330,en:e415,en:e420,en:e420ii,en:e422,en:e471                d   \n",
       "7                                     en:e420,en:e951                c   \n",
       "8                                             en:e414                e   \n",
       "9                                    en:e150d,en:e338                e   \n",
       "10                                   en:e101,en:e101i                d   \n",
       "11                                            en:e338                e   \n",
       "\n",
       "              pnns_groups_1         pnns_groups_2  additives_n  \\\n",
       "0      Cereals and potatoes                 Bread          3.0   \n",
       "1           Composite foods        One-dish meals          2.0   \n",
       "2   Milk and dairy products        Dairy desserts          4.0   \n",
       "3     Fruits and vegetables                Fruits          1.0   \n",
       "4             Sugary snacks                Sweets          7.0   \n",
       "5                   unknown               unknown          5.0   \n",
       "6             Sugary snacks                Sweets          5.0   \n",
       "7             Sugary snacks                Sweets          2.0   \n",
       "8             Sugary snacks                Sweets          1.0   \n",
       "9                 Beverages   Sweetened beverages          2.0   \n",
       "10           Fat and sauces  Dressings and sauces          1.0   \n",
       "11                Beverages   Sweetened beverages          1.0   \n",
       "\n",
       "    nutriscore_score  nova_group  \n",
       "0               -4.0         4.0  \n",
       "1                0.0         4.0  \n",
       "2                2.0         4.0  \n",
       "3                1.0         4.0  \n",
       "4               14.0         4.0  \n",
       "5               15.0         4.0  \n",
       "6               18.0         4.0  \n",
       "7                3.0         4.0  \n",
       "8               36.0         4.0  \n",
       "9               13.0         4.0  \n",
       "10              14.0         4.0  \n",
       "11              14.0         4.0  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df.head(12)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['additives_n', 'nutriscore_score', 'nova_group']\n",
      "['brands_tags', 'labels_en', 'additives_tags', 'nutriscore_grade', 'pnns_groups_1', 'pnns_groups_2']\n"
     ]
    }
   ],
   "source": [
    "# identification des types de variables\n",
    "variables_numeriques = []\n",
    "variables_qualitatives = []\n",
    "\n",
    "for variable in open_food_facts_df:\n",
    "    if is_numeric_dtype(open_food_facts_df[variable]):\n",
    "        variables_numeriques.append(variable)\n",
    "    elif is_string_dtype(open_food_facts_df[variable]):\n",
    "        variables_qualitatives.append(variable)\n",
    "        \n",
    "print(variables_numeriques)\n",
    "print(variables_qualitatives)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Analyses univariées"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Avant de réaliser les analyses, je vais d'abord modifier le type de donnée de *nova_group*."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "open_food_facts_df = open_food_facts_df.astype({'nova_group': 'object'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 104300 entries, 0 to 104299\n",
      "Data columns (total 9 columns):\n",
      " #   Column            Non-Null Count   Dtype  \n",
      "---  ------            --------------   -----  \n",
      " 0   brands_tags       103844 non-null  object \n",
      " 1   labels_en         71866 non-null   object \n",
      " 2   additives_tags    73811 non-null   object \n",
      " 3   nutriscore_grade  104300 non-null  object \n",
      " 4   pnns_groups_1     104159 non-null  object \n",
      " 5   pnns_groups_2     104300 non-null  object \n",
      " 6   additives_n       104300 non-null  float64\n",
      " 7   nutriscore_score  104300 non-null  float64\n",
      " 8   nova_group        104067 non-null  object \n",
      "dtypes: float64(2), object(7)\n",
      "memory usage: 7.2+ MB\n"
     ]
    }
   ],
   "source": [
    "open_food_facts_df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## *Nova_group* : classification NOVA des produits"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4.0    71633\n",
       "3.0    22792\n",
       "1.0     8618\n",
       "2.0     1024\n",
       "Name: nova_group, dtype: int64"
      ]
     },
     "execution_count": 106,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df['nova_group'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Nombre de produits')"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "nombre_produits = open_food_facts_df['nova_group'].count()\n",
    "plt.rcParams[\"figure.figsize\"] = (8, 6)\n",
    "sns.countplot(x='nova_group', data=open_food_facts_df, color='tab:blue', order=[1,2,3,4])\n",
    "plt.title('Répartition des {} produits selon leur classification NOVA'.format(nombre_produits))\n",
    "plt.xlabel('Classification NOVA')\n",
    "plt.ylabel('Nombre de produits')\n",
    "\n",
    "#for variable, pourcentage in enumerate(donnees_manquantes_df.pourcentage):\n",
    " #   plt.text(variable, pourcentage, str(pourcentage)+\"\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "La plupart des produits (environ 70%) sont ultra-transformés."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## *nutriscore_grade* : classement Nutri-Score des produits"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "d    30566\n",
       "c    23016\n",
       "e    22130\n",
       "a    14474\n",
       "b    14114\n",
       "Name: nutriscore_grade, dtype: int64"
      ]
     },
     "execution_count": 107,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df['nutriscore_grade'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Nombre de produits')"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgoAAAGJCAYAAADrDRu+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nO3dd5glVZmA8XcYYERyMjBLUIHPlUHREUEXxBxWUUQMKElMqCwGBFwERVcUARVJkiXjYkAFRRAVEcE0ijigH8IKDEHiEJUBZ2b/OOfCpenqrp7u27eneX/P00/fW6fCV+FWfXXqVNWUhQsXIkmSNJgl+h2AJEmauEwUJElSIxMFSZLUyERBkiQ1MlGQJEmNTBQkSVIjE4UxFBFHRsTnxnma74yI84co3zwicpxiuTYiXjEe09IjIuLEtttdRBwVEfv2OqYhpr9fRJzag/HuFBEXj/V4FyGO1utCYyMi7ouIp/c7jslsyX4HMNFFxLXAk4H5wH3Aj4BdM/O+Af29D5iXmfv0MJZ1gL8BS2XmvwAy8zTgtK5+FgLrZebVtfwXQPQqpl6LiF2BnYANgTMyc6cB5S8HjgDWAn4N7JSZ1w3oZxUggczMzWq3ZwEnA8+ovc0CdsvMK2v5FOAA4D21/Hhgr8xcWMs3qt3+Hfgz8O7MvGzMZrxHMnOXzueIeAlwamb+W/8i0uKo7hffk5kXNJS/BPgZcGRmfqir+8XAcZl5YotpXEjZPo8bqr/MXG6Y8ewNvBdYHbgL+GVmvm246fdSnbdNKfvqObXbKyjLZp2u/nYCdqfsp+4BzgL+OzPviohtgS8AT+vsl+owSwI3ATtn5jm129OAa4CjMvODI43XGoV2tqwb40bAc4H/HthDZh6TmR/tVQB15T8e3QR8DjhhYEFErAZ8B9gXWAX4HfC/g4zji5SD+cDxblOHWw34PvCNrvL3AVsBzwGeDbweeH+d7tLA94BTgZWBk4Dv1e6j8jhezxOS62NU7gd2qCc4Y67NuomIHYHtgVfUffjzgZ+MdxwN7qfsu5rGuztl37UHsCIlsVgb+HHd15wFrARsMWDQ1wALKSe1HTsAc4G3R8S0kQbqj2AEMvPvEXEeJWEAoC70/YG3AtMoK++jmfnPzhkbcCTwMUqNxCdrLQAR8TrKQfAZwN3A8Zm5Xy1bh1J78B7g08C1wDp1sndFBMArKbUF78nMzSLiolr+x1qz8G7gFrrOGiPi34Gv1Xm4kZKdfr+WnUjZeNcBXgxcCbwjM68ZbHlExPY1/uWALw8oWwLYk5LJr0T5ce6SmXdGxBOA44DXAlOBvwKvz8xbBlnm36njez4w8Mx3a+CKzPxm7Wc/4PaIeGZm/qV2eyEwAzimLo/OeO+inF10ag/mA+t2jXtH4EuZeUPt50t1Xo4CXkL57RxSM/lDI+LjwMt49I+zsywuBC4FXk5ZXxcC76rLYh0eu55fHBFvoJwtTAcuAz6QmX+u43supTZjPeCHlJ1CZ1o7UbeHrm4P1zLVdXxDHfe5wLSI6NSOrV+X8ZH18z+B0zLzY4PM02rAicBmwALgCmCLzFwQEWsAh1G2ofuAr2TmoQPHUccz1HxeCxxO2cmtXZftjpn5wGDjGjDeZ9YYZgK3Aftm5pm17EK6zlQHLrO6vHYFPkJZz08bZPybAQcCzwLureM/cUA/KwOnAJvU8fyS8hvobFM7AZ+inOneDuyTmadFxLqU9bsR8BDwk84Z8DDzdSLwjxrv5sAfgTcDn6Bsz7cA22bmH2r/jeup/paeBTwAvAm4vi7730XEKZQavLMjYj7w2cw8cJDVcBdlf/hp4F2DLMP9gHUzc7v6fR1qjSnwmToPm0bEIcCJmbnrYOtmYC3qABsD53X2YZn5d8q+oBPDKsCXgFcDywA/z8ytatl7gb0oJxMXU9bdTbVssDga102DQ4GPR8SBA2OPiBXqMtg5Mzv7lGsj4q3A/wHbZeYJEXEm5fdxYdfgO1B+t/8a0G0fYD9gS+BbQ8T1GNYojEBE/Bvl4Na9Ur9I2aluRDnQTKf8+DueQjljnU75sR4T9ShPzbgpB9LXAR+IiK0GTHYLSvX2qyk/aICVMnO5zLy0u8fM7JQ/p5Y/6uw6IpYCzgbOB54E/BdwWlc8ANtSNtCV63zu37AsnkVJOLYH1gBW5dEH8t0oZ+Rb1PK5lEsE1OWwIrBmHW4XykFppDag7AwByMz7KdVrG9QYp9Zp7krXwXTAfNxF2RkeBny+adz18wZdZZd3V/cBl3eVD2YHYGfKsvgXZSfR7eH1HBHrA2dQdkKrU5KBsyNi6Xom8V3KAWgV4JuUg8GI1GX1WuCmuq0sV3eCXwW+mpkrUBLYph3d7pSEY3XKpbm9gYU1QTybsrymU5Kjj0TEqweOYKj57OrtrZQzpKdRanZ2Gm7eImJZ4MfA6ZTtfFvgyIgYav0MtBXlAP+sQca/FiXJOqzGvRElyRloCeDrlCRnLco2fnhXjIcCr83M5YEXdY3jfyi/0ZUpv6nDRjBfb6UcEFYD5lES1N/X79+iJvQt19MbKLVsK1Fq3A4HyMztKYnDlnW7GSxJ6NgfePOAfcywMvOTwC8ol3mXy8xdu4ob180gfkWp1dgjIp5f9wndTgGeSPntPgn4CkBEvIySwL4VeCpwHY+ucXxUHIu4zd0IHEs5eA/0IuAJlBrTh9VL3udSThKh1GZuExHL1LhXpCQCJ3eGiYjNKdvRNyi/5x2GiGlQ1ii0892aQS4H/JSSIXfORN8LPDsz76zdPk/ZWLovT+ybmfOAn0fEDygb3/9k5oVd/VweEWdQDhjf7eq+X92pM8Lf2mA2rfNwQGYuAH4aEedQNur9aj/fyczf1OmdxoCagi7bAOdk5kW1330pB+SO91N+5J2zp/2A62stxEOUBGHdzLyc0j5gUSxHydy73Q0sXz/vBvw6M2dFxIaDjSAzV6o/8h0pO4Pucd89YLzL1XU+sGzgdAdzSmbOhoeX1WW1WrSjez2/DfhBZv64fj8Y+DBl57GAcsbVqc34VkQ85ox/FB4C1o2I1TLzdsqOtqm/pwJr17OhX9RYXwCsnpmfrf39X0QcC7wdOG/AOIaazwtrP4d2ncWdTVdt3hBeD1ybmV+v338fEd+mbLNXtBge4Aud3/Qg3glckJln1O931L9Hycw7gG93vkfE/pTr9h0LgBkRcX1m3gzcXLs/REku1qi/n04jzTbzdVZmzqrTOwv4YGaeXL//L4/8Rjdm+PV0cWb+sA57CiWhG5FaC3sU8FnK+h4LQ62bgdM/te6730XZxz0QEQdl5gER8VRKsrxqZs6tg/y8/n8ncEJm/h4gIv4bmBsR62TmtQPjqL/ZRdnmvgBcPUhCsRpw+4BagY6bKbUWZOYvI+IWSq3P6ZRjy1X56PZSOwLnZubciDgduCginpSZtw4R16OYKLSzVWZeEBFbUFbGapRqtdUp2eisroP4FEp1esfczgGguo5yVklEbEJpMDcDWJpy6eKbA6Y9ZwznYw1gTk0SuuOZ3vX9712f/0E5KDaOq/MlM++PiO6d5drAWRHRPa35lLPPUyi1Cd+IiJUol2c+mZkPjXB+7gNWGNBtBeDeWq26G/UHNZQa+1HAbRHx7/UHNHDcKwD3ZebCWlU/6HSHmEz3eryOcrBfraF8DbqSllqdP4eynuYDNw6ozXhU481Rejdlp/6XiPgb8JmsDaIGOIiy4z2/bvvHZOYB1ANcranpmEpNJAYYaj47Bm6Pa7SYh7WBTQbEsCRlu2trqN/dmpSaqyFFxBMpZ6ivodQOACwfEVPrNvc24OPA8RHxS2D3eslsT0qtwm8iYi7lEtgJLeer+/LdPwf53vk9t1lPA5f9EyJiyYaD11C+CFwTEc8Z4XBNBl03tabnys73rI0cszb4rjWqW9XPf6DUct7ZlSR0W4NSE9MZ1311/zadcnlwYByLtM1l5m0RcTjlN/e1rqLbgdUalvdTa3nHyZRagtMpNbwndQpqTcNbqI2yM/PSiLgeeAdwyFCxdTNRGIHM/HmU64AHUza42yk/vg0y88aGwVaOiGW7koW1gNn18+mU6rzXZuYDUa7FrTZg+IUNnxfFTcCaEbFEV7KwFnDVIozrZkpVOfDwTnHVrvI5lOtrv2wY/jPAZ6Jcl/wh5a6E40cYwxWUbLkTw7KU6vIrgBdQflBX1gPZMsAyEfF3YHpmzh8wriUoSd904NY6jucAv6nlz+GRM4MrgN0jYkrXAfvZPHJpZTBrdn1ei3LWeHtX9+51exPlLo/OfE2p/d1Y+5s+YNpr8ciB6/46H51hnzJETI/ZnjLzr8C2tWp6a0qNxaoDkl0y817K5Yfd69nQzyLit5T1/rfMXG+I6baZz9GYQ7nW/MqG8kctI8rlwYGG+q3NoWxfw9md0iZlk3pmvRHwB8rJBJl5HnBe3Zl/jlINvXm9jv5eeLgtxAVR2h8NN18jMZL1NJjW+6LMvKPu2/5nQNFw66FpGoN2z8zraT6xoZ6IfDMi9qKcnJ0OrBIRK2Vps9TtJsrBH3h437Iqj942u+MYzbo5iNLu4Ddd3S6lXDramq7LfzWO11Iu9XWcDHwqSnusTSm1Ch1vopzEHBkRh9VuK1ESCxOFHjqE0qhko8y8rFbXfSUids3MWyNiOjCj7gQ6PhPlFp1NKNWHn67dl6dktA/UKtt3UK5NNrmNUl35dJoP7rfU8sEa9vya8uPcM0rjvP+gXM/aePjZfoxvAb+uO7LfUDLi7jYvRwH7R8SOmXldRKwOvCgzvxcRL6UcJK+k3PLzEOVM+TGitCheknK2MzVKQ8h/1Sz7LOCgiHgz8ANK25DLM7NzNrxO16jeRlm+b8zM+RHxyhrD5cCylB31XB65O+Jk4GMR0WksuDv1WjGlWnw+sFutiXhv7f7TIZbXdhFxMuVs5LPAt2ocg/V7JvCJKLd+XkSpjp8HXFLL/1WnfQTlOvILeKRK+4/ABvWg9BcGv/7ZcQuwakSsmJl3A0TEdpTGX7d1nR09Zt1ExOvr+K+hrMP59e83wD11Z3wo8CAloVwmM387wvlcVOcAB9TLXJ3ryhtRaoT+TGkLsHVEHEc5c+w0+m3rNGDvKA3LvkNtb5OPvT12ecqJxF1RGs11fvdExJMp+4Of1H7uoy7niHgLcGm97DCXsv3NbzFfIzGS9TSYzn6mrS9TDoZTurpdBuxVawLu5rF3k410Go8RpcHobZTt635KW68NKJckb46IcykH0Q9R1sEL6+XU0yk1nqdT9gmfr8Nc2zCpRV43WW51/BKlJune2u3uiPgMcFhE3EPZTqZTGhrfQFdNRd2/Xkxp7/Pjmmh27Ei5Y+yTXd2mA7+NiA0z809DxdZhY8YRyszbKAeRzm0te1EOyr+qK/QCHv3cgr9Tfuw3UXYwu9TqRYAPAp+NiHspB7mhWsiSmf+gNA76ZUTcFRGbDtLbfsBJtfytA4Z/kHJgeS3lIHkksENXPK1l5hXAhyg/qJvrPN7Q1ctXKQ2gzq/z9yvKjhHKmcO3KAeYP1OuCzY9hGcfyo70E8B29fM+NYbbKA359q/T34RyjZXMnJeZf+/8UXZED3X9iFai/LDuphzs1gVek4+0qD+a0tjrT5QaoB/Ubp3luBUlK7+L0khxq9q9ySmUuwT+TmmktFtTj5mZdV4Po6ynLSkNxx6s09ia0qhvLiUB+k7XsFdREpELKHeTND6EqK73MyjXp++ql2teA1xRL698FXh7Dn6XwXp1GvdRzn6OzMwLa03NlpSd5N9q/MdRDqat57Mp5jZqbcerKNvCTZRl/kXKpT0olwMepByITqLrOSQtx3898J+U5PFOygFvsGr1Qyg1WZ22Ht13xCxRh7+pjmMLyv4ASuL+67oOvg98ODP/1mK+RjIPrddTgy8A+9Tt5uMtpncP5S6RVbq6/ZhyO3OnndLAS1xfpTTUmxsRg94108I9lLPv6ym/1QMpd9Z0fhedNlN/odQkfqTG9hPKPv7blP3bM6j7lob5G+26+SoDEvIsjUT3ptRg30M50ZsDvDxLm7duJ1FqQLobMXYaqR7SvS/M0oblR3TVxg5nysKFo63NVpPwgTai/YNjJGkiskZBkiQ1MlGQJEmNvPQgSZIaWaMgSZIaeXvkALNmzZpGaXV8Mw237EmSNIlMpTx35rczZ84ceEeFicIgNmbwp8hJkjSZbc4gt1SbKDzWzQDrr78+Sy896rcGS5I0oT344INcddVV8Mj7Rh7FROGx5gMsvfTSTJs24ueYSJK0uBr0cruNGSVJUiMTBUmS1MhEQZIkNTJRkCRJjUwUJElSIxMFSZLUyERBkiQ1MlGQJEmNTBQkSVIjEwVJktTIREGSJDUyUZAkSY1MFCRJUiMTBUnqoXkPDfpCPlUun4lv3F4zHRHfBZ4GLADuA/4rMy+LiPWBk4BVgTuAHTLzr3WYMS+TpPE0bampzNzj5H6HMWHNOmiHfoegYYxnjcKOmfmczHwucDBwQu1+FHBEZq4PHAEc3TVML8okSVJL41ajkJl3d31dEVgQEU8Cnge8snY/Azg8IlYHpox1WWbe1pOZkyRpkhrXNgoRcVxEXA/sD+wIrAncmJnzAer/m2r3XpRJkqQRGLcaBYDMfA9ARGwPHATsO57TH4nZs2f3OwRJk8DMmTP7HcKEN2vWrH6HoCGMa6LQkZmnRMQxwA3A9IiYmpnzI2IqsAYwh3IJYazLWpsxYwbTpk0bq1mWJDUwmeqvefPmDXlyPC6XHiJiuYhYs+v7lsCdwK3AZcC2tWhb4A+ZeVtmjnlZ7+ZQkqTJabxqFJYFvhkRywLzKUnClpm5MCJ2AU6KiE8Bc4Hue2V6USZJkloal0QhM28BNm0o+wuwyXiVSZKk9nwyoyRJamSiIEmSGpkoSJKkRiYKkiSpkYmCJElqZKIgSZIamShIkqRGJgqSJKmRiYIkSWpkoiBJkhqZKEiSpEYmCpIkqZGJgiRJamSiIEmSGpkoSJKkRiYKkiSpkYmCJElqZKIgSZIamShIkqRGJgqSJKmRiYIkSWpkoiBJkhqZKEiSpEYmCpIkqZGJgiRJamSiIEmSGpkoSJKkRiYKkiSpkYmCJElqZKIgSZIamShIkqRGJgqSJKmRiYIkSWpkoiBJkhqZKEiSpEYmCpIkqZGJgiRJamSiIEmSGpkoSJKkRiYKkiSpkYmCJElqZKIgSZIaLTkeE4mIVYFTgGcA84Crgfdn5m0RsRD4E7Cg9r59Zv6pDrclcFCNcxbwrsz8x2jKJElSe+NVo7AQODAzIzOfDVwDHNBV/qLM3Kj+dZKE5YBjgS0zc13gXuDjoymTJEkjMy6JQmbemZkXdnX6FbD2MIO9FvhdZv61fj8KeNsoyyRJ0giMy6WHbhGxBPAB4PtdnS+MiCWBc4H9MnMesBZwXVc/1wNr1s+LWiZJkkZg3BMF4DDgPuDw+n2tzJwTEStQ2jHsC+zTh7geZfbs2f0OQdIkMHPmzH6HMOHNmjWr3yFoCOOaKETEwcB6lPYDCwAyc079f09EHAd8rPZ+PfDSrsHXAuaMsqy1GTNmMG3atJEOJkkaIZOp/po3b96QJ8fjdntkROwPzAS2qpcWiIiVI2KZ+nlJYBvgsjrIj4CNI2K9+n0X4MxRlkmSpBEYl0QhIjYA9gbWAC6JiMsi4izgmcCvI+KPwOXAQ5RLD2TmvcD7gHMi4mpgReDg0ZRJkqSRGZdLD5l5BTClofjZQwz3PeB7Y1kmSZLa88mMkiSpkYmCJElqZKIgSZIamShIkqRGJgqSJKmRiYIkSWpkoiBJkhqZKEiSpEYmCpIkqZGJgiRJamSiIEmSGpkoSJKkRiYKkiSpkYmCJElqZKIgSZIamShIj2PzHprf7xAmNJePBEv2OwBJ/TNtqanM3OPkfocxYc06aId+hyD1nTUKkiSpkYmCJElqZKIgSZIamShIkqRGJgqSJKmRiYIkSWpkoiBJkhqZKEiSFns+HKvZaJeND1ySJC32fHhYs9E+OMwaBUmS1KhVjUJEPAu4IzNviYjlgD2A+cDBmfmPXgYoSZL6p22NwunASvXzwcCLgRcCR/ciKEmSNDG0baOwTmZmREwB3gRsAPwT+FvPIpMkSX3XtkZhXkQsD7wAmJOZtwPzgCf0LDJJktR3bWsUTgd+CiwPHF67PQ9rFCRJmtRaJQqZ+dGIeBXwUGb+rHZeAHy0Z5FJkqS+a3XpISIOzczzu5IEMvN3wBt6FpkkSeq7tm0Udmrovv0YxSFJkiagIS89RMTOnf66Pnc8Hbi9J1FJkqQJYbg2Cp0ag6V5dO3BQuAWYMdeBCVJkiaGIROFzHwpQER8LjP3GZ+QJEnSRNGYKETElMxcWL9+KiIGbc+QmQt6EpkkSeq7oWoU7gZWqJ//Rbnc0G1K7Ta1B3FJkqQJYKhEYYOuz0/rdSCSJGniaUwUMnNO1+frxiccSZI0kbR9zfQpPPbSAwCZuUOL4VcFTgGeQXlHxNXA+zPztojYlPIWymWAa4HtMvPWOtyYl0mSpPbaPnDpauCarr/7gdcCd7YcfiFwYGZGZj67juOA+jbKU4EPZeb6wEXAAVAaU451mSRJGpm273r4zMBuEXE88OmWw98JXNjV6VfAB4DnAw9k5sW1+1GUGoCde1QmSZJGoG2NwmAuA7YY6UD1NssPAN8H1gIebv9QX1+9RESs0qMySZI0Am3bKLxsQKcnAm8HrlyEaR4G3Ed5XfWbFmH4cTF79ux+hyD13MyZM/sdwoQ3a9asUQ3vMh7eaJcxuJyHM5pl3CpRAI4f8P1+So3CtiOZWEQcDKwHbJmZCyLiemDtrvLVgIWZeWcvykYS64wZM5g2bdpIBpE0CXkA6j2Xce8NtYznzZs35Mlx2zYKo36OQkTsD8wEXpeZ82rnWcAyEbFZbVOwC3BmD8skSdIIDPUI51btF9o8wjkiNgD2Bq4CLokIgL9l5psiYnvg6Ih4AvVWxs54x7pMkiSNzFA1CoM9tnkwwz7COTOvoDzyebCyS4ANx6tMkiS1N1Si0H254XXANsAXKHcUrA3sBXy7d6FJkqR+G+oRzg/fYhgRHwOen5l31U5XRcTvgN8BX+ttiJIkqV/aPkdhRcotkd2eWLtLkqRJqu3tkScBF0TEIcAcYE1gt9pdkiRNUm0ThT0p73t4G7AGcDPlgUnH9iguSZI0AbR9jsICyjsTjuptOJIkaSJpW6NARLwL2B6YDtwInJKZX+9VYJIkqf9aNWaMiE8CnwC+QWmb8A1gz9pdkiRNUm1rFN4DvGTALZPnARcB+/ciMEmS1H9tb49cFrhtQLc7gGXGNhxJkjSRtK1R+BFwWkR8Aui8nXF/4LxeBSZJkvqvbY3CrsC9wB+B+yivmL4f+K8exSVJkiaAYWsU6lsknw+8F9gJWA24vc1bIyVJ0uJt2EShvrb5e5m5fO10a49jkiRJE0TbSw8XRcSmPY1EkiRNOG0bM14HnBsR36O862FhpyAzP9WLwCRJUv+1TRSWAb5bP/9bV/eFg/QrSZImibbvenhXrwORJEkTz0je9bAe8FbK2yNvAs7MzL/2KjBJktR/bd/18A7gD8CzKc9P2BD4fe0uSZImqbY1Cp8D/jMzL+p0iIjNgVOA03sRmCRJ6r+2t0cuD1w6oNuvKO+AkCRJk1TbROHLwOcj4gkAEbEM5V0PX+5VYJIkqf/aXnr4IPAU4MMRMRdYGZgC3BwRH+j0lJlrjX2IkiSpX9omCtv1NApJkjQhtX2Ows97HYgkSZp42rZRkCRJj0MmCpIkqZGJgiRJajSiRCEiloiIp/YqGEmSNLG0fYTzShFxOvAAcHXt9oaI+Fwvg5MkSf3VtkbhKOBuYG3gwdrtUuBtvQhKkiRNDG0ThZcDu2XmzcBCgMy8DXhSrwKTJEn91zZRuBtYrbtDRKwF3DzmEUmSpAmjbaJwHPDtiHgpsEREvBA4iXJJQpIkTVJtH+H8RUpDxiOApYATgKOBr/YoLkmSNAG0fYTzQuCQ+idJkh4nGhOFiHhZmxFk5k/HLhxJkjSRDFWjcPyA79MpdzzcAaxKec30DcDTexOaJEnqt8ZEITOf1vkcEXtTkoN9M/MfEfFE4LOUpEGSJE1Sbe96+Cjwicz8B0D9/9/Ax3oVmCRJ6r+2icL9wAsGdNsY+MfYhiNJkiaStrdH7gv8KCLOBuYAawKvBz7UdkIRcTDwZmAdYMPMnF27X0u59fKB2utemXleLduUchvmMsC1wHaZeetoyiRJUnutahQy8xRgE+DPwArAX4BNa/e2vgu8GLhukLJtMnOj+tdJEqYApwIfysz1gYuAA0ZTJkmSRqZtjQKZeSVw5aJOKDMvBoiItoM8H3igMxzlKZDXAjuPokySJI1A2zYKvXZaRFweEUdGxEq121p01T5k5u2Ux0evMooySZI0Aq1rFHpo88ycExHTKE9+PBzYrs8xMXv27H6HIPXczJkz+x3ChDdr1qxRDe8yHt5olzG4nIczmmXc90QhM+fU//Mi4kjg+7XoemDtTn8RsRqwMDPvjIhFKhtJXDNmzGDatGmLOluSJgkPQL3nMu69oZbxvHnzhjw5HtGlh4hYs95RMCYiYtmIWLF+ngK8HbisFs8ClomIzer3XYAzR1kmSZJGoFWNQkSsBZwBbER5jPNyEbEN8JrMfE/LcRwKbA08BbggIu4AtqS8vnoqMJXSWPKDAJm5ICK2B46OiCdQb3McTZkkSRqZtpcejgZ+AGzOI49t/jHwpbYTyszdgN0GKXruEMNcAmw4lmWSJKm9tpceXgAckJkLKDUKZObdwIq9Cmyim/fQ/H6HMGG5bCRp8mhbo3ALsC5wVadDRDyL0uDwcWnaUlOZucfJ/Q5jQpp10A79DkGSNEba1igcDJwTEe8CloyIbYH/Bb7Ys8gkSVLftX2E8wnAnsBbKO962IHyyunTehibJEnqs2EvPdQ7Ej4N7J+Z3+19SFIx76H5TFtqar/DmJBcNpLGy7CJQmbOj4gPAfv1PhzpEbYDaWY7EEnjpW0bhZMoDy6SJEmPI23vengB8F8RsSeljcLCTkFmvrgXgUmSpP5rmygcW/8kSdLjSKtEITNP6nUgkp6gNo8AABCFSURBVCRp4mn99siI2BnYFlgDuAn4BnBCZi4cckBJkrTYavtSqAOBNwKHANcBawEfB4LyfAVJkjQJta1R2Al4Xmbe0OkQET8Afo+JgiRJk1bb2yPvrX8Du90ztuFIkqSJpLFGISKe3vX1EOA7EXEAcAOwJrAH8JXehidJkvppqEsPV1OelzClq9tLB/TzMuDwsQ5KkiRNDI2JQma2vSwhSZImKZMBSZLUqO3tkWtR3iD5XGC57rLMXL8HcUmSpAmg7e2R3wT+AnwK+GfvwpEkSRNJ20ThmcALM3NBL4ORJEkTS9s2CmcDW/QyEEmSNPG0rVHYDbgkIq4BbukuyMydxzwqSZI0IbRNFL4OzAf+jG0UJEl63GibKLwMWCMzBz7GWZIkTWJt2yhcDqzay0AkSdLE07ZG4afA+RHxdR7bRuGEMY9KkiRNCG0Thc2AG4FXDei+EDBRkCRpkmqVKGTmwJdBSZKkx4G2NQpExMrAlsB0Su3C2Zk5t1eBSZKk/mvVmDEiXghcA+wCPBt4P3BN7S5JkiaptjUKhwAfzMxvdDpExNuAQ4GNexGYJEnqv7a3R64PnDmg27eAdcc2HEmSNJG0TRT+Crx9QLe3UC5HSJKkSartpYePAOdExG7AdcA6wHrA63sUlyRJmgBa1Shk5iXAM4DDgVnAYcC6tbskSZqkWt8eWW+FPLWHsUiSpAlmyEQhIn5Gefpik4WZ+fKxDUmSJE0Uw9UoNNUgTAd2A544tuFIkqSJZMhEITOP7/4eEasC/w28F/hf4LO9C02SJPVbqzYKEbECsAewK3AO8LzM9NZISZImueHaKCxDuTVyd+BCYLPMvGIc4pIkSRPAcDUKfwOmAgcCvwOeHBFP7u4hM3863EQi4mDgzZTnL2yYmbNr9/WBk4BVgTuAHTLzr70qkyRJIzPccxQeAO4DPgAcP8jfcS2n813gxZSHNXU7CjgiM9cHjgCO7nGZJEkageEaM64zFhPJzIsBIuLhbhHxJOB5wCtrpzOAwyNidWDKWJdl5m1jMS+SJD2etH3XQy+sCdyYmfMB6v+bavdelEmSpBFq/WTGx5vZs2cPWT5z5sxximTxNGvWrFGPw2U8NJfx+BjtcnYZD89tufdGs4z7mSjMAaZHxNTMnB8RU4E1avcpPSgbkRkzZjBt2rQxmdHHI3+0vecyHh8u595zGffeUMt43rx5Q54c9+3SQ2beClwGbFs7bQv8ITNv60VZ7+dIkqTJZ1xqFCLiUGBr4CnABRFxR2ZuAOwCnBQRnwLmAjt0DdaLMkmSNALjkihk5m6Ud0MM7P4XYJOGYca8TJIkjUw/73qQJEkTnImCJElqZKIgSZIamShIkqRGJgqSJKmRiYIkSWpkoiBJkhqZKEiSpEYmCpIkqZGJgiRJamSiIEmSGpkoSJKkRiYKkiSpkYmCJElqZKIgSZIamShIkqRGJgqSJKmRiYIkSWpkoiBJkhqZKEiSpEYmCpIkqZGJgiRJamSiIEmSGpkoSJKkRiYKkiSpkYmCJElqZKIgSZIamShIkqRGJgqSJKmRiYIkSWpkoiBJkhqZKEiSpEYmCpIkqZGJgiRJamSiIEmSGpkoSJKkRiYKkiSpkYmCJElqZKIgSZIamShIkqRGJgqSJKmRiYIkSWq0ZL8DAIiIa4EH6h/AXpl5XkRsChwNLANcC2yXmbfWYRapTJIktTeRahS2ycyN6t95ETEFOBX4UGauD1wEHACwqGWSJGlkJlKiMNDzgQcy8+L6/SjgraMskyRJIzAhLj1Up9XagIuBvYG1gOs6hZl5e0QsERGrLGpZZt7ZNpjZs2cPWT5z5sy2o3pcmjVr1qjH4TIemst4fIx2ObuMh+e23HujWcYTJVHYPDPnRMQ04BDgcOCsfgY0Y8YMpk2b1s8QFmv+aHvPZTw+XM695zLuvaGW8bx584Y8OZ4Qlx4yc079Pw84EvgP4Hpg7U4/EbEasLDWCixqmSRJGoG+JwoRsWxErFg/TwHeDlwGzAKWiYjNaq+7AGfWz4taJkmSRqDviQLwZODCiLgcmA2sD3wwMxcA2wNfi4i/AlsAnwBY1DJJkjQyfW+jkJn/Bzy3oewSYMOxLJMkSe1NhBoFSZI0QZkoSJKkRiYKkiSpkYmCJElqZKIgSZIamShIkqRGJgqSJKmRiYIkSWpkoiBJkhqZKEiSpEYmCpIkqZGJgiRJamSiIEmSGpkoSJKkRiYKkiSpkYmCJElqZKIgSZIamShIkqRGJgqSJKmRiYIkSWpkoiBJkhqZKEiSpEYmCpIkqZGJgiRJamSiIEmSGpkoSJKkRiYKkiSpkYmCJElqZKIgSZIamShIkqRGJgqSJKmRiYIkSWpkoiBJkhqZKEiSpEYmCpIkqZGJgiRJamSiIEmSGpkoSJKkRiYKkiSpkYmCJElqZKIgSZIaLdnvAHolItYHTgJWBe4AdsjMv/Y3KkmSFi+TuUbhKOCIzFwfOAI4us/xSJK02JmUiUJEPAl4HnBG7XQG8LyIWL1/UUmStPiZrJce1gRuzMz5AJk5PyJuqt1vG2bYqQAPPvjgsBNZddmlRhnm5DRv3rwxG5fLeHAu4/ExVsvZZdzMbbn3hlvGXce7qYOVT1m4cOEYh9R/ETETODkzN+jqdiWwXWb+fqhhZ82atRnwix6HKEnSRLP5zJkzLx7YcbLWKMwBpkfE1FqbMBVYo3Yfzm+BzYGbgfk9jFGSpIlgKvBUyvHvMSZljQJARFwIHJeZp0bEdsC7M/OlfQ5LkqTFymROFJ5JuT1yZWAu5fbI7G9UkiQtXiZtoiBJkkZvUt4eKUmSxoaJgiRJamSiIEmSGpkoSJKkRiYKkiSpkYmCNISIWBgRy/U7DmmsuW2rLRMFSZLUaLI+wvlxIyJOAwKYBlwN7JyZc/sb1aTz8Yh4FbAqsHdmfrvfAU02EfFC4CBg+dppj8w8v48hTToRsTXweeBO4Id9DmfSiohNgAOAFWqnT2XmD/oY0qhZo7D4+3BmPj8zNwSuAPbqd0CT0ILMfBHwBuCY+hpzjZGIWAU4C9gzM59DeUX8oM+c16Kp2+yxwBvrtjz863E1YhGxEnAU8I7MnAm8Hji6dl9sWaOw+NshIt4JLA0sC1zV53gmo+MBMjMj4vfApsD3+xvSpPJC4MrMvATKa+Epj13X2NkU+H3XY+yPAb7Yx3gmqxcBTwPOjYhOt4XAusDv+hXUaJkoLMYiYnPgA8CLMvO2iHgH8L4+hzXZTaH88DV2pvQ7gMcBl/H4mAJcnpkv7ncgY8lLD4u3lYC7gTsiYhqwc5/jmazeBRAR6wEbAb/ubziTziXAs2o7BSJiakSs3OeYJptLgefWbRjgPf0MZhK7BFgvIh5+U3FEbBwRi3WiZqKweDsXuAb4S/38+/6GM2nNi4hfAucA78/MW/sd0GSSmXcCWwNfjojLgVnAzP5GNbnUbfZ9wNkRcQnwrz6HNCnVhuRvAD4dEX+MiD8D+7GY1+j49khJktTIGgVJktTIREGSJDUyUZAkSY1MFCRJUiMTBUmS1MhEQZqAImK/iDi133E8nkTEFRHxkn7HIU00PplR6pP6JM2PAc8E7gUuA/bPzIv7Gtg4iogLgVMz87iG8nWAvwE/zMzXdXU/Fbg6M/drMY0TgRsyc5+h+svMDYYZz7uBPYDpwD8oj+R9e2beO1wM0uLMGgWpDyLiY8AhlLf5PRlYCzgSeGM/45rANo2I/+jFiCNi2BOmiNiCsq62zczlgX8HzhzvOKR+cMOUxllErAh8FnhXZn6nq+js+jfYMN8ENgeWAf4IfCAzr6hl/wkcDKwJ3AN8JTMPjojVgBOBzYAFlLeLbpGZCyJiDeAw4MXAfXWYQ+v49gM2AOZREpdrgTfXv4/W7u/uvAa6zs+Xgf+s0/k68OnMnB8RO1EeF/wr4N3AXcAHM/PciNi/ztOmEXEIcGJm7tqw2A4EPge8dGBBZxqZuVlXt4XAesDLgHcCCyPiI8DPMnPLiLgW+Foti4hYlvKa9vdk5gWDTH9j4NLM/AM8/DTJk7qmt0yNbxvKo9X/BLwyM/8ZEW8AvkCpibiMsu7+XIcbLI4n0bBupH6wRkEafy8EnkB5tXJb51IOfE+iPKr7tK6y4ymPll4emAH8tHbfHbgBWJ1Sa7E35YC5BCUh+SPl4PVy4CMR8equcW4JnAKsDPwBOI+yv5hOSXKO7ur3JMojgdcFngu8ike/S2ATIIHVKAf84yNiSmZ+EvgFsGtmLjdEkgBwBLB+RLxiiH4eIzOPoSyrA+s0tuwq3hZ4HbBSZg73SONfA6+OiM9ExH/Ud6t0O5jy2OkXAasAewILImJ94AzgI5T18EPKY5SXHiwOSqI13LqRxpU1CtL4WxW4vcXB6WGZeULncz3jnxsRK2bm3cBDlJcq/bE+a77ziuaHgKcCa2fm1ZSDMhHxAmD1zPxs7e//IuJY4O2UhADgF5l5Xu3/m5R3MRxQawm+ARwTESsB04DXUg62/wTuj4ivUN4r0EkmrsvMY+u4TqJcYnky8Pe28w88AOxPOWsf7Ix/URyamXPa9JiZv4iIrYEPAh8GloyIYyhtFhZSXsi2aWbeWAe5BCAi3gb8IDN/XL8fXId/EXDhwDgiYhOGXzfSuDJRkMbfHcBqEbFkm2QhIqZSDpJvoZyVLqhFq1HeHvpmYB/ggPpSpU9k5qXAQZQX0pwfEQDHZOYBwNrAGhFxV9dkplITieqWrs//pCQ287u+AywHrAEsBdxcpwGl5qH7APxwQpCZ/6j9LTfcfA/iWGCPiNhy2D7baUwSIuK+rq/PyszrM/Nc4NxaI/NS4JuUmpKzKDVE1wwyqjWA6zpf6mWfOZTagsHiaLNupHFloiCNv0spZ8hbAd9q0f87KG0FXkFpL7AipdZgCkBm/hZ4Y0QsBexKaWS3Zm2Nvzuwe0RsAPwsIn5LOTD9LTPXGzihRTCH0mZhtZHUkHRp/Va6zHwoIj4D/A+lvUXH/cATO18i4iktp9E47cxsTGQycwHwk4j4KeVSz7GU9fkMyiWDbjcBG3bFNoXSluTGrn664xjLdSONCRMFaZxl5t0R8SngiIj4F3A+5TLBK4CXZuaeAwZZnnIwvoNyQPx8p6Be634LcE4d7z3A/Fr2esoryK+hNHKcX/9+A9wTEXsBhwIPUlrxL1OTjpHMy80RcT7wpYjYl9L47mnAv2Xmz1uM4hbg6SOY5CnAXsBrgL/Wbn8ENoiIjSjzu98op/EYEfFGSkPS8ygNMjcGtgA+UmsJTqC8Jnv7Or0XUNqSnAl8IiJeDlxEuewwj3ppYhBjtm6ksWJjRqkPMvPLlGco7APcRjmT3BX47iC9n0ypvr4RuJJyB0G37YFra5KwC7Bd7b4e5Xr+fZRajCMz88J6CWFLYCPKMwpuB46j1FQsih2ApWtscym1JE9tOexXgW0iYm5EDNuyv8b+aUqDwU63qygNLC+gJA8Dn0NxPKUNx10RMdjybWMu8N46/nuAU4GDMrPTqPTjlDsdfgvcCXwRWCIzk7I+DqMs5y2BLTPzwSHmbyzXjTRqUxYubF3zJ0mSHmesUZAkSY1MFCRJUiMTBUmS1MhEQZIkNTJRkCRJjUwUJElSIxMFSZLUyERBkiQ1MlGQJEmN/h/SbtZFmZPawgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "nombre_produits = open_food_facts_df['nutriscore_grade'].count()\n",
    "plt.rcParams[\"figure.figsize\"] = (8, 6)\n",
    "sns.countplot(x='nutriscore_grade', data=open_food_facts_df, color='tab:blue', order=['a', 'b', 'c', 'd', 'e'])\n",
    "plt.title('Répartition des {} produits selon leur classement Nutri-Score'.format(nombre_produits))\n",
    "plt.xlabel('Classement Nutri-Score')\n",
    "plt.ylabel('Nombre de produits')\n",
    "\n",
    "#for variable, pourcentage in enumerate(donnees_manquantes_df.pourcentage):\n",
    " #   plt.text(variable, pourcentage, str(pourcentage)+\"\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "La plupart des produits ont un Nutri-Score moyen ou mauvais (environ 70%)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## *nutriscore_score* : score final Nutri-Score des produits"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    104300.000000\n",
       "mean          9.488926\n",
       "std           8.933405\n",
       "min         -14.000000\n",
       "25%           2.000000\n",
       "50%          10.000000\n",
       "75%          16.000000\n",
       "max          40.000000\n",
       "Name: nutriscore_score, dtype: float64"
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df['nutriscore_score'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(-3.8249999999999957, 0.5, 'Nombre de produits')"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Distribution du score final Nutri-Score\n",
    "nombre_produits = open_food_facts_df['nutriscore_score'].count()\n",
    "sns.displot(x='nutriscore_score', data=open_food_facts_df, discrete=True)\n",
    "plt.title('Répartition des {} produits selon leur score final Nutri-Score'.format(nombre_produits))\n",
    "plt.xlabel('Score final Nutri-Score')\n",
    "plt.ylabel('Nombre de produits')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Score final Nutri-Score')"
      ]
     },
     "execution_count": 115,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Dispersion du score final Nutri-Score\n",
    "sns.boxplot(x='nutriscore_score', data=open_food_facts_df)\n",
    "plt.title('Dispersion du score final Nutri-Score parmi les produits')\n",
    "plt.xlabel('Score final Nutri-Score')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Voyons quels produits sont les outliers sur leur score final du Nutri-Score."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>brands_tags</th>\n",
       "      <th>labels_en</th>\n",
       "      <th>additives_tags</th>\n",
       "      <th>nutriscore_grade</th>\n",
       "      <th>pnns_groups_1</th>\n",
       "      <th>pnns_groups_2</th>\n",
       "      <th>additives_n</th>\n",
       "      <th>nutriscore_score</th>\n",
       "      <th>nova_group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>8012</td>\n",
       "      <td>cynxshno</td>\n",
       "      <td>NaN</td>\n",
       "      <td>en:e150d</td>\n",
       "      <td>e</td>\n",
       "      <td>Beverages</td>\n",
       "      <td>Unsweetened beverages</td>\n",
       "      <td>1.0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>90646</td>\n",
       "      <td>body-attack</td>\n",
       "      <td>NaN</td>\n",
       "      <td>en:e551,en:e950,en:e955</td>\n",
       "      <td>e</td>\n",
       "      <td>unknown</td>\n",
       "      <td>unknown</td>\n",
       "      <td>3.0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       brands_tags labels_en           additives_tags nutriscore_grade  \\\n",
       "8012      cynxshno       NaN                 en:e150d                e   \n",
       "90646  body-attack       NaN  en:e551,en:e950,en:e955                e   \n",
       "\n",
       "      pnns_groups_1          pnns_groups_2  additives_n  nutriscore_score  \\\n",
       "8012      Beverages  Unsweetened beverages          1.0              40.0   \n",
       "90646       unknown                unknown          3.0              40.0   \n",
       "\n",
       "      nova_group  \n",
       "8012           4  \n",
       "90646          4  "
      ]
     },
     "execution_count": 116,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df[open_food_facts_df['nutriscore_score'] == 40]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Le premier produit est une boisson de Pologne. Le deuxième un complément alimentaire de protéines."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## *additives_n* : nombre d'additifs des produits"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    104300.000000\n",
       "mean          2.140489\n",
       "std           2.476332\n",
       "min           0.000000\n",
       "25%           0.000000\n",
       "50%           1.000000\n",
       "75%           3.000000\n",
       "max          30.000000\n",
       "Name: additives_n, dtype: float64"
      ]
     },
     "execution_count": 117,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df['additives_n'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(-10.824999999999996, 0.5, 'Nombre de produits')"
      ]
     },
     "execution_count": 118,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Distribution du nombre d'additifs des produits\n",
    "nombre_produits = open_food_facts_df['additives_n'].count()\n",
    "sns.displot(x='additives_n', data=open_food_facts_df, discrete=True)\n",
    "plt.title('Répartition des {} produits selon leur nombre d\\'additifs'.format(nombre_produits))\n",
    "plt.xlabel('Nombre d\\'additifs')\n",
    "plt.ylabel('Nombre de produits')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "La plupart des produits n'ont aucun additif ou en ont très peu. Pour 75% d'entre eux, il y en a 3 au maximum."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, \"Nombre d'additifs\")"
      ]
     },
     "execution_count": 119,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Dispersion du nombre d'additifs des produits\n",
    "sns.boxplot(x='additives_n', data=open_food_facts_df)\n",
    "plt.title('Dispersion du nombre d\\'additifs parmi les produits')\n",
    "plt.xlabel('Nombre d\\'additifs')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Voyons quels produits sont les outliers sur leur nombre d'additifs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>brands_tags</th>\n",
       "      <th>labels_en</th>\n",
       "      <th>additives_tags</th>\n",
       "      <th>nutriscore_grade</th>\n",
       "      <th>pnns_groups_1</th>\n",
       "      <th>pnns_groups_2</th>\n",
       "      <th>additives_n</th>\n",
       "      <th>nutriscore_score</th>\n",
       "      <th>nova_group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>4737</td>\n",
       "      <td>chef-select-to-go</td>\n",
       "      <td>NaN</td>\n",
       "      <td>en:e14xx,en:e160c,en:e161b,en:e202,en:e224,en:...</td>\n",
       "      <td>d</td>\n",
       "      <td>Composite foods</td>\n",
       "      <td>Sandwiches</td>\n",
       "      <td>14.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>37914</td>\n",
       "      <td>leader-price</td>\n",
       "      <td>NaN</td>\n",
       "      <td>en:e14xx,en:e160c,en:e161b,en:e202,en:e260,en:...</td>\n",
       "      <td>e</td>\n",
       "      <td>Fat and sauces</td>\n",
       "      <td>Dressings and sauces</td>\n",
       "      <td>8.0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>78633</td>\n",
       "      <td>poppies-berlidon</td>\n",
       "      <td>Made in France</td>\n",
       "      <td>en:e14xx,en:e160a,en:e160c,en:e171,en:e202,en:...</td>\n",
       "      <td>e</td>\n",
       "      <td>Sugary snacks</td>\n",
       "      <td>Biscuits and cakes</td>\n",
       "      <td>14.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14029</td>\n",
       "      <td>haribo</td>\n",
       "      <td>NaN</td>\n",
       "      <td>en:e100,en:e131,en:e141,en:e141ii,en:e153,en:e...</td>\n",
       "      <td>d</td>\n",
       "      <td>Sugary snacks</td>\n",
       "      <td>Sweets</td>\n",
       "      <td>15.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>93069</td>\n",
       "      <td>winny</td>\n",
       "      <td>Green Dot,Made in France,fr:eco-emballages</td>\n",
       "      <td>en:e14xx,en:e160,en:e262,en:e300,en:e322,en:e3...</td>\n",
       "      <td>c</td>\n",
       "      <td>Composite foods</td>\n",
       "      <td>One-dish meals</td>\n",
       "      <td>9.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9656</td>\n",
       "      <td>la-laitiere,nestle,secret-de-mousse</td>\n",
       "      <td>Green Dot,New,fr:eco-emballages,fr:nutritional...</td>\n",
       "      <td>en:e100,en:e14xx,en:e160a,en:e160ai,en:e171,en...</td>\n",
       "      <td>c</td>\n",
       "      <td>Milk and dairy products</td>\n",
       "      <td>Dairy desserts</td>\n",
       "      <td>14.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>81954</td>\n",
       "      <td>les-delices-des-7-vallees</td>\n",
       "      <td>NaN</td>\n",
       "      <td>en:e14xx,en:e160a,en:e160b,en:e202,en:e300,en:...</td>\n",
       "      <td>d</td>\n",
       "      <td>sugary-snacks</td>\n",
       "      <td>pastries</td>\n",
       "      <td>9.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>102760</td>\n",
       "      <td>look-o-look</td>\n",
       "      <td>FSC,FSC Mix,fr:fsc-c021442</td>\n",
       "      <td>en:e100,en:e120,en:e133,en:e150d,en:e153,en:e1...</td>\n",
       "      <td>d</td>\n",
       "      <td>Sugary snacks</td>\n",
       "      <td>Sweets</td>\n",
       "      <td>17.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>52734</td>\n",
       "      <td>monoprix</td>\n",
       "      <td>Green Dot</td>\n",
       "      <td>en:e14xx,en:e160a,en:e250,en:e316,en:e331,en:e...</td>\n",
       "      <td>d</td>\n",
       "      <td>Composite foods</td>\n",
       "      <td>One-dish meals</td>\n",
       "      <td>8.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>21445</td>\n",
       "      <td>casino</td>\n",
       "      <td>NaN</td>\n",
       "      <td>en:e160,en:e160b,en:e322,en:e322i,en:e330,en:e...</td>\n",
       "      <td>c</td>\n",
       "      <td>Milk and dairy products</td>\n",
       "      <td>Ice cream</td>\n",
       "      <td>8.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>41774</td>\n",
       "      <td>le-gaulois</td>\n",
       "      <td>French meat,No preservatives,Palm-oil-free,fr:...</td>\n",
       "      <td>en:e100,en:e14xx,en:e160c,en:e331,en:e407,en:e...</td>\n",
       "      <td>c</td>\n",
       "      <td>Fish Meat Eggs</td>\n",
       "      <td>Meat</td>\n",
       "      <td>8.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>59881</td>\n",
       "      <td>rochambeau</td>\n",
       "      <td>NaN</td>\n",
       "      <td>en:e127,en:e202,en:e220,en:e330,en:e415,en:e42...</td>\n",
       "      <td>e</td>\n",
       "      <td>Sugary snacks</td>\n",
       "      <td>Biscuits and cakes</td>\n",
       "      <td>8.0</td>\n",
       "      <td>22.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>26183</td>\n",
       "      <td>claude-leger</td>\n",
       "      <td>Nutriscore</td>\n",
       "      <td>en:e14xx,en:e202,en:e250,en:e330,en:e450,en:e4...</td>\n",
       "      <td>b</td>\n",
       "      <td>Composite foods</td>\n",
       "      <td>One-dish meals</td>\n",
       "      <td>9.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>102959</td>\n",
       "      <td>carte-d-or</td>\n",
       "      <td>Green Dot</td>\n",
       "      <td>en:e14xx,en:e160,en:e171,en:e172,en:e322,en:e3...</td>\n",
       "      <td>c</td>\n",
       "      <td>unknown</td>\n",
       "      <td>unknown</td>\n",
       "      <td>10.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>54714</td>\n",
       "      <td>collection-saveurs-d-asie,carrefour</td>\n",
       "      <td>NaN</td>\n",
       "      <td>en:e330,en:e407,en:e412,en:e415,en:e450,en:e45...</td>\n",
       "      <td>c</td>\n",
       "      <td>Salty snacks</td>\n",
       "      <td>Appetizers</td>\n",
       "      <td>8.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>61504</td>\n",
       "      <td>mamie-nova</td>\n",
       "      <td>NaN</td>\n",
       "      <td>en:e14xx,en:e163,en:e270,en:e330,en:e331,en:e3...</td>\n",
       "      <td>e</td>\n",
       "      <td>Milk and dairy products</td>\n",
       "      <td>Dairy desserts</td>\n",
       "      <td>11.0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>50111</td>\n",
       "      <td>charculor</td>\n",
       "      <td>NaN</td>\n",
       "      <td>en:e160b,en:e250,en:e262,en:e262i,en:e301,en:e...</td>\n",
       "      <td>e</td>\n",
       "      <td>Fish Meat Eggs</td>\n",
       "      <td>Processed meat</td>\n",
       "      <td>8.0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>94487</td>\n",
       "      <td>nawhal-s</td>\n",
       "      <td>Halal</td>\n",
       "      <td>en:e14xx,en:e160a,en:e160ai,en:e160c,en:e163,e...</td>\n",
       "      <td>d</td>\n",
       "      <td>Fat and sauces</td>\n",
       "      <td>Dressings and sauces</td>\n",
       "      <td>8.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>13895</td>\n",
       "      <td>haribo</td>\n",
       "      <td>NaN</td>\n",
       "      <td>en:e120,en:e131,en:e161b,en:e296,en:e330,en:e3...</td>\n",
       "      <td>d</td>\n",
       "      <td>Sugary snacks</td>\n",
       "      <td>Sweets</td>\n",
       "      <td>8.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>13943</td>\n",
       "      <td>haribo</td>\n",
       "      <td>NaN</td>\n",
       "      <td>en:e100,en:e131,en:e153,en:e163,en:e330,en:e41...</td>\n",
       "      <td>d</td>\n",
       "      <td>Sugary snacks</td>\n",
       "      <td>Sweets</td>\n",
       "      <td>8.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>61526</td>\n",
       "      <td>nova</td>\n",
       "      <td>NaN</td>\n",
       "      <td>en:e14xx,en:e150a,en:e150b,en:e401,en:e407,en:...</td>\n",
       "      <td>c</td>\n",
       "      <td>Milk and dairy products</td>\n",
       "      <td>Dairy desserts</td>\n",
       "      <td>8.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>71332</td>\n",
       "      <td>marque-repere</td>\n",
       "      <td>Nutriscore,Nutriscore Grade C</td>\n",
       "      <td>en:e14xx,en:e224,en:e250,en:e300,en:e316,en:e3...</td>\n",
       "      <td>d</td>\n",
       "      <td>Composite foods</td>\n",
       "      <td>Sandwiches</td>\n",
       "      <td>11.0</td>\n",
       "      <td>16.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>45911</td>\n",
       "      <td>vedere</td>\n",
       "      <td>Green Dot,Pure butter</td>\n",
       "      <td>en:e160,en:e270,en:e327,en:e330,en:e401,en:e41...</td>\n",
       "      <td>e</td>\n",
       "      <td>Sugary snacks</td>\n",
       "      <td>Biscuits and cakes</td>\n",
       "      <td>10.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18200</td>\n",
       "      <td>charal</td>\n",
       "      <td>French meat,Kosher,Orthodox Union Kosher,fr:Vi...</td>\n",
       "      <td>en:e14xx,en:e160a,en:e160c,en:e202,en:e224,en:...</td>\n",
       "      <td>d</td>\n",
       "      <td>Composite foods</td>\n",
       "      <td>Sandwiches</td>\n",
       "      <td>16.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>93068</td>\n",
       "      <td>winny</td>\n",
       "      <td>Green Dot,Made in France,fr:eco-emballages</td>\n",
       "      <td>en:e160,en:e250,en:e301,en:e322,en:e322i,en:e3...</td>\n",
       "      <td>e</td>\n",
       "      <td>Composite foods</td>\n",
       "      <td>One-dish meals</td>\n",
       "      <td>9.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>90373</td>\n",
       "      <td>marque-blanche</td>\n",
       "      <td>Halal</td>\n",
       "      <td>en:e14xx,en:e150c,en:e250,en:e262,en:e301,en:e...</td>\n",
       "      <td>e</td>\n",
       "      <td>Fish Meat Eggs</td>\n",
       "      <td>Processed meat</td>\n",
       "      <td>10.0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>72233</td>\n",
       "      <td>hafner</td>\n",
       "      <td>NaN</td>\n",
       "      <td>en:e14xx,en:e160a,en:e322,en:e331,en:e339,en:e...</td>\n",
       "      <td>d</td>\n",
       "      <td>Sugary snacks</td>\n",
       "      <td>Biscuits and cakes</td>\n",
       "      <td>18.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>39598</td>\n",
       "      <td>leader-price</td>\n",
       "      <td>NaN</td>\n",
       "      <td>en:e14xx,en:e250,en:e282,en:e301,en:e322,en:e3...</td>\n",
       "      <td>b</td>\n",
       "      <td>Composite foods</td>\n",
       "      <td>Sandwiches</td>\n",
       "      <td>10.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3141</td>\n",
       "      <td>leclerc</td>\n",
       "      <td>NaN</td>\n",
       "      <td>en:e101,en:e101i,en:e14xx,en:e160a,en:e211,en:...</td>\n",
       "      <td>d</td>\n",
       "      <td>Sugary snacks</td>\n",
       "      <td>Biscuits and cakes</td>\n",
       "      <td>19.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14013</td>\n",
       "      <td>haribo</td>\n",
       "      <td>NaN</td>\n",
       "      <td>en:e100,en:e131,en:e141,en:e141ii,en:e160a,en:...</td>\n",
       "      <td>d</td>\n",
       "      <td>Sugary snacks</td>\n",
       "      <td>Sweets</td>\n",
       "      <td>14.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                brands_tags  \\\n",
       "4737                      chef-select-to-go   \n",
       "37914                          leader-price   \n",
       "78633                      poppies-berlidon   \n",
       "14029                                haribo   \n",
       "93069                                 winny   \n",
       "9656    la-laitiere,nestle,secret-de-mousse   \n",
       "81954             les-delices-des-7-vallees   \n",
       "102760                          look-o-look   \n",
       "52734                              monoprix   \n",
       "21445                                casino   \n",
       "41774                            le-gaulois   \n",
       "59881                            rochambeau   \n",
       "26183                          claude-leger   \n",
       "102959                           carte-d-or   \n",
       "54714   collection-saveurs-d-asie,carrefour   \n",
       "61504                            mamie-nova   \n",
       "50111                             charculor   \n",
       "94487                              nawhal-s   \n",
       "13895                                haribo   \n",
       "13943                                haribo   \n",
       "61526                                  nova   \n",
       "71332                         marque-repere   \n",
       "45911                                vedere   \n",
       "18200                                charal   \n",
       "93068                                 winny   \n",
       "90373                        marque-blanche   \n",
       "72233                                hafner   \n",
       "39598                          leader-price   \n",
       "3141                                leclerc   \n",
       "14013                                haribo   \n",
       "\n",
       "                                                labels_en  \\\n",
       "4737                                                  NaN   \n",
       "37914                                                 NaN   \n",
       "78633                                      Made in France   \n",
       "14029                                                 NaN   \n",
       "93069          Green Dot,Made in France,fr:eco-emballages   \n",
       "9656    Green Dot,New,fr:eco-emballages,fr:nutritional...   \n",
       "81954                                                 NaN   \n",
       "102760                         FSC,FSC Mix,fr:fsc-c021442   \n",
       "52734                                           Green Dot   \n",
       "21445                                                 NaN   \n",
       "41774   French meat,No preservatives,Palm-oil-free,fr:...   \n",
       "59881                                                 NaN   \n",
       "26183                                          Nutriscore   \n",
       "102959                                          Green Dot   \n",
       "54714                                                 NaN   \n",
       "61504                                                 NaN   \n",
       "50111                                                 NaN   \n",
       "94487                                               Halal   \n",
       "13895                                                 NaN   \n",
       "13943                                                 NaN   \n",
       "61526                                                 NaN   \n",
       "71332                       Nutriscore,Nutriscore Grade C   \n",
       "45911                               Green Dot,Pure butter   \n",
       "18200   French meat,Kosher,Orthodox Union Kosher,fr:Vi...   \n",
       "93068          Green Dot,Made in France,fr:eco-emballages   \n",
       "90373                                               Halal   \n",
       "72233                                                 NaN   \n",
       "39598                                                 NaN   \n",
       "3141                                                  NaN   \n",
       "14013                                                 NaN   \n",
       "\n",
       "                                           additives_tags nutriscore_grade  \\\n",
       "4737    en:e14xx,en:e160c,en:e161b,en:e202,en:e224,en:...                d   \n",
       "37914   en:e14xx,en:e160c,en:e161b,en:e202,en:e260,en:...                e   \n",
       "78633   en:e14xx,en:e160a,en:e160c,en:e171,en:e202,en:...                e   \n",
       "14029   en:e100,en:e131,en:e141,en:e141ii,en:e153,en:e...                d   \n",
       "93069   en:e14xx,en:e160,en:e262,en:e300,en:e322,en:e3...                c   \n",
       "9656    en:e100,en:e14xx,en:e160a,en:e160ai,en:e171,en...                c   \n",
       "81954   en:e14xx,en:e160a,en:e160b,en:e202,en:e300,en:...                d   \n",
       "102760  en:e100,en:e120,en:e133,en:e150d,en:e153,en:e1...                d   \n",
       "52734   en:e14xx,en:e160a,en:e250,en:e316,en:e331,en:e...                d   \n",
       "21445   en:e160,en:e160b,en:e322,en:e322i,en:e330,en:e...                c   \n",
       "41774   en:e100,en:e14xx,en:e160c,en:e331,en:e407,en:e...                c   \n",
       "59881   en:e127,en:e202,en:e220,en:e330,en:e415,en:e42...                e   \n",
       "26183   en:e14xx,en:e202,en:e250,en:e330,en:e450,en:e4...                b   \n",
       "102959  en:e14xx,en:e160,en:e171,en:e172,en:e322,en:e3...                c   \n",
       "54714   en:e330,en:e407,en:e412,en:e415,en:e450,en:e45...                c   \n",
       "61504   en:e14xx,en:e163,en:e270,en:e330,en:e331,en:e3...                e   \n",
       "50111   en:e160b,en:e250,en:e262,en:e262i,en:e301,en:e...                e   \n",
       "94487   en:e14xx,en:e160a,en:e160ai,en:e160c,en:e163,e...                d   \n",
       "13895   en:e120,en:e131,en:e161b,en:e296,en:e330,en:e3...                d   \n",
       "13943   en:e100,en:e131,en:e153,en:e163,en:e330,en:e41...                d   \n",
       "61526   en:e14xx,en:e150a,en:e150b,en:e401,en:e407,en:...                c   \n",
       "71332   en:e14xx,en:e224,en:e250,en:e300,en:e316,en:e3...                d   \n",
       "45911   en:e160,en:e270,en:e327,en:e330,en:e401,en:e41...                e   \n",
       "18200   en:e14xx,en:e160a,en:e160c,en:e202,en:e224,en:...                d   \n",
       "93068   en:e160,en:e250,en:e301,en:e322,en:e322i,en:e3...                e   \n",
       "90373   en:e14xx,en:e150c,en:e250,en:e262,en:e301,en:e...                e   \n",
       "72233   en:e14xx,en:e160a,en:e322,en:e331,en:e339,en:e...                d   \n",
       "39598   en:e14xx,en:e250,en:e282,en:e301,en:e322,en:e3...                b   \n",
       "3141    en:e101,en:e101i,en:e14xx,en:e160a,en:e211,en:...                d   \n",
       "14013   en:e100,en:e131,en:e141,en:e141ii,en:e160a,en:...                d   \n",
       "\n",
       "                  pnns_groups_1         pnns_groups_2  additives_n  \\\n",
       "4737            Composite foods            Sandwiches         14.0   \n",
       "37914            Fat and sauces  Dressings and sauces          8.0   \n",
       "78633             Sugary snacks    Biscuits and cakes         14.0   \n",
       "14029             Sugary snacks                Sweets         15.0   \n",
       "93069           Composite foods        One-dish meals          9.0   \n",
       "9656    Milk and dairy products        Dairy desserts         14.0   \n",
       "81954             sugary-snacks              pastries          9.0   \n",
       "102760            Sugary snacks                Sweets         17.0   \n",
       "52734           Composite foods        One-dish meals          8.0   \n",
       "21445   Milk and dairy products             Ice cream          8.0   \n",
       "41774            Fish Meat Eggs                  Meat          8.0   \n",
       "59881             Sugary snacks    Biscuits and cakes          8.0   \n",
       "26183           Composite foods        One-dish meals          9.0   \n",
       "102959                  unknown               unknown         10.0   \n",
       "54714              Salty snacks            Appetizers          8.0   \n",
       "61504   Milk and dairy products        Dairy desserts         11.0   \n",
       "50111            Fish Meat Eggs        Processed meat          8.0   \n",
       "94487            Fat and sauces  Dressings and sauces          8.0   \n",
       "13895             Sugary snacks                Sweets          8.0   \n",
       "13943             Sugary snacks                Sweets          8.0   \n",
       "61526   Milk and dairy products        Dairy desserts          8.0   \n",
       "71332           Composite foods            Sandwiches         11.0   \n",
       "45911             Sugary snacks    Biscuits and cakes         10.0   \n",
       "18200           Composite foods            Sandwiches         16.0   \n",
       "93068           Composite foods        One-dish meals          9.0   \n",
       "90373            Fish Meat Eggs        Processed meat         10.0   \n",
       "72233             Sugary snacks    Biscuits and cakes         18.0   \n",
       "39598           Composite foods            Sandwiches         10.0   \n",
       "3141              Sugary snacks    Biscuits and cakes         19.0   \n",
       "14013             Sugary snacks                Sweets         14.0   \n",
       "\n",
       "        nutriscore_score nova_group  \n",
       "4737                11.0          4  \n",
       "37914               19.0          4  \n",
       "78633               20.0          4  \n",
       "14029               14.0          4  \n",
       "93069                5.0          4  \n",
       "9656                 7.0          4  \n",
       "81954               17.0          4  \n",
       "102760              14.0          4  \n",
       "52734               12.0          4  \n",
       "21445               10.0          4  \n",
       "41774                8.0          4  \n",
       "59881               22.0          4  \n",
       "26183                1.0          4  \n",
       "102959               6.0          4  \n",
       "54714                6.0          4  \n",
       "61504               19.0          4  \n",
       "50111               19.0          4  \n",
       "94487               17.0          4  \n",
       "13895               14.0          4  \n",
       "13943               14.0          4  \n",
       "61526                4.0          4  \n",
       "71332               16.0          4  \n",
       "45911               24.0          4  \n",
       "18200               12.0          4  \n",
       "93068               20.0          4  \n",
       "90373               19.0          4  \n",
       "72233               18.0          4  \n",
       "39598                1.0          4  \n",
       "3141                13.0          4  \n",
       "14013               14.0          4  "
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filtered_df = open_food_facts_df[open_food_facts_df['additives_n'] > 7]\n",
    "filtered_df.sample(30)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.9621323603043257"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "st.skew(open_food_facts_df['additives_n'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "La distribution du nombre d'additifs est totalement asymétrique, le coefficient de skewness valant 1.96 le confirme car il est positif. Il y aura besoin de transformer cette variable avec le log avant d'utiliser le dataset pour entraîner un modèle.\n",
    "\n",
    "Voyons les stats des produits sans les outliers :"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    99929.000000\n",
       "mean         1.799508\n",
       "std          1.833887\n",
       "min          0.000000\n",
       "25%          0.000000\n",
       "50%          1.000000\n",
       "75%          3.000000\n",
       "max          7.000000\n",
       "Name: additives_n, dtype: float64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filtered_df = open_food_facts_df[open_food_facts_df['additives_n'] < 8]\n",
    "filtered_df['additives_n'].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## *pnns_groups_1* : catégorie principale des produits"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Sugary snacks              22711\n",
       "Composite foods            13098\n",
       "Fish Meat Eggs             12689\n",
       "Milk and dairy products    12460\n",
       "Cereals and potatoes        8721\n",
       "Beverages                   7807\n",
       "Salty snacks                6897\n",
       "unknown                     6141\n",
       "Fruits and vegetables       6140\n",
       "Fat and sauces              5396\n",
       "sugary-snacks               1383\n",
       "fruits-and-vegetables        712\n",
       "cereals-and-potatoes           4\n",
       "Name: pnns_groups_1, dtype: int64"
      ]
     },
     "execution_count": 134,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df['pnns_groups_1'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Nombre de produits')"
      ]
     },
     "execution_count": 136,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ordre_categorie = open_food_facts_df['pnns_groups_1'].value_counts().index\n",
    "nombre_produits = open_food_facts_df['pnns_groups_1'].count()\n",
    "plt.rcParams[\"figure.figsize\"] = (12, 6)\n",
    "sns.countplot(y='pnns_groups_1', data=open_food_facts_df, color='tab:blue', order=ordre_categorie)\n",
    "plt.title('Répartition des {} produits selon leur catégorie principale'.format(nombre_produits))\n",
    "plt.ylabel('Catégorie principale')\n",
    "plt.xlabel('Nombre de produits')\n",
    "\n",
    "#for variable, pourcentage in enumerate(donnees_manquantes_df.pourcentage):\n",
    " #   plt.text(variable, pourcentage, str(pourcentage)+\"\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Les deux principales catégories représentées parmi les produits sont les snacks sucrés et les plats préparés. On peut remarquer que trois catégories sont doublonnées : *sugary-snacks*, *fruits-and-vegetables* et *cereals-and-potatoes*. Ces catégories devront être regroupées avec leur doublon.\n",
    "\n",
    "On note également que les produits de bonne qualité nutritionnelle tels que les fruits, légumes et céréales sont peu représentés dans notre jeu de données."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Regroupons les catégories concernées."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "open_food_facts_df['pnns_groups_1'].replace(['sugary-snacks', 'fruits-and-vegetables', 'cereals-and-potatoes'],\n",
    "                                           ['Sugary snacks', 'Fruits and vegetables', 'Cereals and potatoes'],\n",
    "                                           inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Sugary snacks              24094\n",
       "Composite foods            13098\n",
       "Fish Meat Eggs             12689\n",
       "Milk and dairy products    12460\n",
       "Cereals and potatoes        8725\n",
       "Beverages                   7807\n",
       "Salty snacks                6897\n",
       "Fruits and vegetables       6852\n",
       "unknown                     6141\n",
       "Fat and sauces              5396\n",
       "Name: pnns_groups_1, dtype: int64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df['pnns_groups_1'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Nombre de produits')"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ordre_categorie = open_food_facts_df['pnns_groups_1'].value_counts().index\n",
    "nombre_produits = open_food_facts_df['pnns_groups_1'].count()\n",
    "plt.rcParams[\"figure.figsize\"] = (12, 6)\n",
    "sns.countplot(y='pnns_groups_1', data=open_food_facts_df, color='tab:blue', order=ordre_categorie)\n",
    "plt.title('Répartition des {} produits selon leur catégorie principale'.format(nombre_produits))\n",
    "plt.ylabel('Catégorie principale')\n",
    "plt.xlabel('Nombre de produits')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## *pnns_groups_2* : catégorie secondaire des produits"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Biscuits and cakes                  11070\n",
       "One-dish meals                      10693\n",
       "Sweets                               8043\n",
       "Processed meat                       6718\n",
       "unknown                              6141\n",
       "Dressings and sauces                 4356\n",
       "Cheese                               4250\n",
       "Milk and yogurt                      4144\n",
       "Salty and fatty products             3621\n",
       "Chocolate products                   3598\n",
       "Cereals                              3502\n",
       "Sweetened beverages                  3284\n",
       "Meat                                 2775\n",
       "Appetizers                           2700\n",
       "Fish and seafood                     2647\n",
       "Bread                                2322\n",
       "Vegetables                           2291\n",
       "Dairy desserts                       2215\n",
       "Fruits                               2155\n",
       "Ice cream                            1851\n",
       "Breakfast cereals                    1512\n",
       "Fruit juices                         1449\n",
       "pastries                             1383\n",
       "Pizza pies and quiche                1294\n",
       "Sandwiches                           1111\n",
       "Unsweetened beverages                1105\n",
       "Artificially sweetened beverages     1102\n",
       "Fats                                 1040\n",
       "Dried fruits                          847\n",
       "Soups                                 847\n",
       "Legumes                               715\n",
       "vegetables                            672\n",
       "Potatoes                              670\n",
       "Nuts                                  576\n",
       "Plant-based milk substitutes          572\n",
       "Offals                                342\n",
       "Eggs                                  207\n",
       "Waters and flavored waters            160\n",
       "Pizza pies and quiches                139\n",
       "Fruit nectars                          92\n",
       "Teas and herbal teas and coffees       43\n",
       "fruits                                 40\n",
       "cereals                                 3\n",
       "Alcoholic beverages                     2\n",
       "legumes                                 1\n",
       "Name: pnns_groups_2, dtype: int64"
      ]
     },
     "execution_count": 137,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df['pnns_groups_2'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Nombre de produits')"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ordre_categorie = open_food_facts_df['pnns_groups_2'].value_counts().index\n",
    "nombre_produits = open_food_facts_df['pnns_groups_2'].count()\n",
    "plt.rcParams[\"figure.figsize\"] = (12, 10)\n",
    "sns.countplot(y='pnns_groups_2', data=open_food_facts_df, color='tab:blue', order=ordre_categorie)\n",
    "plt.title('Répartition des {} produits selon leur catégorie secondaire'.format(nombre_produits))\n",
    "plt.ylabel('Catégorie secondaire')\n",
    "plt.xlabel('Nombre de produits')\n",
    "\n",
    "#for variable, pourcentage in enumerate(donnees_manquantes_df.pourcentage):\n",
    " #   plt.text(variable, pourcentage, str(pourcentage)+\"\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Les principales sous-catégories représentées parmi les produits sont les biscuits, gâteaux, plats préparés, bonbons et la viande transformée.\n",
    "\n",
    "Les catégories qui pourront être rassemblées en une seule sont :\n",
    "- *Vegetables*, *vegetables*, *legumes* et *Legumes* \n",
    "- *Fruits* et *fruits*\n",
    "- *Cereals* et *cereals*\n",
    "- *Pizza pies and quiche* et *Pizza pies and quiches*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Regroupons ces sous-catégories."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "open_food_facts_df['pnns_groups_2'].replace(\n",
    "                        ['fruits', 'legumes', 'Legumes', 'vegetables', 'cereals', 'Pizza pies and quiches'],\n",
    "                        ['Fruits', 'Vegetables', 'Vegetables', 'Vegetables', 'Cereals', 'Pizza pies and quiche'],\n",
    "                        inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Biscuits and cakes                  11070\n",
       "One-dish meals                      10693\n",
       "Sweets                               8043\n",
       "Processed meat                       6718\n",
       "unknown                              6141\n",
       "Dressings and sauces                 4356\n",
       "Cheese                               4250\n",
       "Milk and yogurt                      4144\n",
       "Vegetables                           3679\n",
       "Salty and fatty products             3621\n",
       "Chocolate products                   3598\n",
       "Cereals                              3505\n",
       "Sweetened beverages                  3284\n",
       "Meat                                 2775\n",
       "Appetizers                           2700\n",
       "Fish and seafood                     2647\n",
       "Bread                                2322\n",
       "Dairy desserts                       2215\n",
       "Fruits                               2195\n",
       "Ice cream                            1851\n",
       "Breakfast cereals                    1512\n",
       "Fruit juices                         1449\n",
       "Pizza pies and quiche                1433\n",
       "pastries                             1383\n",
       "Sandwiches                           1111\n",
       "Unsweetened beverages                1105\n",
       "Artificially sweetened beverages     1102\n",
       "Fats                                 1040\n",
       "Dried fruits                          847\n",
       "Soups                                 847\n",
       "Potatoes                              670\n",
       "Nuts                                  576\n",
       "Plant-based milk substitutes          572\n",
       "Offals                                342\n",
       "Eggs                                  207\n",
       "Waters and flavored waters            160\n",
       "Fruit nectars                          92\n",
       "Teas and herbal teas and coffees       43\n",
       "Alcoholic beverages                     2\n",
       "Name: pnns_groups_2, dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df['pnns_groups_2'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Nombre de produits')"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ordre_categorie = open_food_facts_df['pnns_groups_2'].value_counts().index\n",
    "nombre_produits = open_food_facts_df['pnns_groups_2'].count()\n",
    "plt.rcParams[\"figure.figsize\"] = (12, 10)\n",
    "sns.countplot(y='pnns_groups_2', data=open_food_facts_df, color='tab:blue', order=ordre_categorie)\n",
    "plt.title('Répartition des {} produits selon leur catégorie secondaire'.format(nombre_produits))\n",
    "plt.ylabel('Catégorie secondaire')\n",
    "plt.xlabel('Nombre de produits')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## *brands_tags* : marque des produits"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "u                            3576\n",
       "auchan                       3107\n",
       "carrefour                    2965\n",
       "casino                       2035\n",
       "leader-price                 1688\n",
       "picard                       1325\n",
       "cora                         1116\n",
       "fleury-michon                 799\n",
       "netto                         782\n",
       "monoprix                      721\n",
       "belle-france                  716\n",
       "la-vie-claire                 669\n",
       "franprix                      631\n",
       "le-gaulois                    532\n",
       "thiriet                       506\n",
       "bio-village,marque-repere     446\n",
       "u-saveurs,u                   397\n",
       "u-bio,u                       383\n",
       "nestle                        369\n",
       "dia                           330\n",
       "monique-ranou                 327\n",
       "sans-marque                   322\n",
       "bjorg                         321\n",
       "casino-bio,casino             285\n",
       "yoplait                       284\n",
       "haribo                        278\n",
       "andros                        276\n",
       "paturages                     271\n",
       "knorr                         264\n",
       "bonduelle                     262\n",
       "Name: brands_tags, dtype: int64"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df['brands_tags'].value_counts().head(30)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "La plupart des marques représentées sont des chaînes de la grande distribution ou des entreprises de l'agro-alimentaire. La variable peut donner aussi une indication sur le lieu d'achat du produit."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## *labels_en* : label portés par les produits"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Green Dot                                                                       4116\n",
       "Made in France                                                                  3010\n",
       "Organic                                                                         2244\n",
       "Organic,EU Organic,AB Agriculture Biologique                                    1700\n",
       "No preservatives                                                                1037\n",
       "No colorings,No preservatives                                                    823\n",
       "PDO                                                                              763\n",
       "Gluten-free                                                                      750\n",
       "Green Dot,Made in France                                                         718\n",
       "No colorings                                                                     525\n",
       "Green Dot,fr:eco-emballages                                                      501\n",
       "Palm-oil-free                                                                    460\n",
       "Green Dot,fr:Triman                                                              454\n",
       "Organic,EU Organic,FR-BIO-01,AB Agriculture Biologique                           442\n",
       "Made in Italy                                                                    441\n",
       "PGI                                                                              382\n",
       "fr:viande-francaise,fr:Volaille Française                                        377\n",
       "Distributor labels,fr:Sélection Intermarché                                      374\n",
       "Produced in Bretagne                                                             371\n",
       "Organic,EU Organic,AB Agriculture Biologique,FR-BIO-01                           369\n",
       "Organic,EU Organic,FR-BIO-01                                                     366\n",
       "Transformed in France                                                            354\n",
       "No added sugar                                                                   354\n",
       "French meat,French pork                                                          339\n",
       "Nutriscore,Nutriscore Grade A                                                    339\n",
       "Lait-francais-french-milk                                                        337\n",
       "Nutriscore,Nutriscore Grade C                                                    254\n",
       "Organic,EU Organic,AB Agriculture Biologique,EU/non-EU Agriculture,FR-BIO-01     244\n",
       "Halal                                                                            240\n",
       "Made in Belgium                                                                  235\n",
       "Name: labels_en, dtype: int64"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df['labels_en'].value_counts().head(30)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    71866.000000\n",
       "mean         2.875059\n",
       "std         15.074944\n",
       "min         -1.000000\n",
       "25%         -1.000000\n",
       "50%         -1.000000\n",
       "75%         -1.000000\n",
       "max        283.000000\n",
       "Name: labels_en, dtype: float64"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df['labels_en'].str.find(\"preservatives\").describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>brands_tags</th>\n",
       "      <th>labels_en</th>\n",
       "      <th>additives_tags</th>\n",
       "      <th>nutriscore_grade</th>\n",
       "      <th>pnns_groups_1</th>\n",
       "      <th>pnns_groups_2</th>\n",
       "      <th>additives_n</th>\n",
       "      <th>nutriscore_score</th>\n",
       "      <th>nova_group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>10</td>\n",
       "      <td>heinz</td>\n",
       "      <td>Vegetarian,No artificial colors,No flavors,No ...</td>\n",
       "      <td>en:e101,en:e101i</td>\n",
       "      <td>d</td>\n",
       "      <td>Fat and sauces</td>\n",
       "      <td>Dressings and sauces</td>\n",
       "      <td>1.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>heinz</td>\n",
       "      <td>No preservatives,No thickening agent</td>\n",
       "      <td>NaN</td>\n",
       "      <td>d</td>\n",
       "      <td>Fat and sauces</td>\n",
       "      <td>Dressings and sauces</td>\n",
       "      <td>0.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>31</td>\n",
       "      <td>genius</td>\n",
       "      <td>Gluten-free,No preservatives</td>\n",
       "      <td>en:e422,en:e464</td>\n",
       "      <td>b</td>\n",
       "      <td>Cereals and potatoes</td>\n",
       "      <td>Bread</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>99</td>\n",
       "      <td>liebig</td>\n",
       "      <td>Cooked in France,No colorings,No preservatives</td>\n",
       "      <td>en:e14xx</td>\n",
       "      <td>c</td>\n",
       "      <td>Fruits and vegetables</td>\n",
       "      <td>Soups</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>201</td>\n",
       "      <td>koon-chun</td>\n",
       "      <td>No preservatives</td>\n",
       "      <td>en:e401</td>\n",
       "      <td>c</td>\n",
       "      <td>Fat and sauces</td>\n",
       "      <td>Dressings and sauces</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104270</td>\n",
       "      <td>yeo-s</td>\n",
       "      <td>No preservatives</td>\n",
       "      <td>en:e407,en:e452,en:e471,en:e500</td>\n",
       "      <td>b</td>\n",
       "      <td>Beverages</td>\n",
       "      <td>Plant-based milk substitutes</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104275</td>\n",
       "      <td>pa-food</td>\n",
       "      <td>Vegetarian,Halal,Vegan,No cholesterol,No prese...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>b</td>\n",
       "      <td>Cereals and potatoes</td>\n",
       "      <td>Bread</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104279</td>\n",
       "      <td>kawan</td>\n",
       "      <td>No preservatives</td>\n",
       "      <td>en:e500</td>\n",
       "      <td>c</td>\n",
       "      <td>Cereals and potatoes</td>\n",
       "      <td>Bread</td>\n",
       "      <td>1.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104280</td>\n",
       "      <td>marigold</td>\n",
       "      <td>FSC,No-preservativeshalal</td>\n",
       "      <td>en:e296,en:e330,en:e466</td>\n",
       "      <td>b</td>\n",
       "      <td>unknown</td>\n",
       "      <td>unknown</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104288</td>\n",
       "      <td>giovanni-rana</td>\n",
       "      <td>No preservatives,Palm-oil-free</td>\n",
       "      <td>NaN</td>\n",
       "      <td>c</td>\n",
       "      <td>Cereals and potatoes</td>\n",
       "      <td>Potatoes</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8001 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          brands_tags                                          labels_en  \\\n",
       "10              heinz  Vegetarian,No artificial colors,No flavors,No ...   \n",
       "18              heinz               No preservatives,No thickening agent   \n",
       "31             genius                       Gluten-free,No preservatives   \n",
       "99             liebig     Cooked in France,No colorings,No preservatives   \n",
       "201         koon-chun                                   No preservatives   \n",
       "...               ...                                                ...   \n",
       "104270          yeo-s                                   No preservatives   \n",
       "104275        pa-food  Vegetarian,Halal,Vegan,No cholesterol,No prese...   \n",
       "104279          kawan                                   No preservatives   \n",
       "104280       marigold                          FSC,No-preservativeshalal   \n",
       "104288  giovanni-rana                     No preservatives,Palm-oil-free   \n",
       "\n",
       "                         additives_tags nutriscore_grade  \\\n",
       "10                     en:e101,en:e101i                d   \n",
       "18                                  NaN                d   \n",
       "31                      en:e422,en:e464                b   \n",
       "99                             en:e14xx                c   \n",
       "201                             en:e401                c   \n",
       "...                                 ...              ...   \n",
       "104270  en:e407,en:e452,en:e471,en:e500                b   \n",
       "104275                              NaN                b   \n",
       "104279                          en:e500                c   \n",
       "104280          en:e296,en:e330,en:e466                b   \n",
       "104288                              NaN                c   \n",
       "\n",
       "                pnns_groups_1                 pnns_groups_2  additives_n  \\\n",
       "10             Fat and sauces          Dressings and sauces          1.0   \n",
       "18             Fat and sauces          Dressings and sauces          0.0   \n",
       "31       Cereals and potatoes                         Bread          2.0   \n",
       "99      Fruits and vegetables                         Soups          1.0   \n",
       "201            Fat and sauces          Dressings and sauces          1.0   \n",
       "...                       ...                           ...          ...   \n",
       "104270              Beverages  Plant-based milk substitutes          4.0   \n",
       "104275   Cereals and potatoes                         Bread          0.0   \n",
       "104279   Cereals and potatoes                         Bread          1.0   \n",
       "104280                unknown                       unknown          3.0   \n",
       "104288   Cereals and potatoes                      Potatoes          0.0   \n",
       "\n",
       "        nutriscore_score nova_group  \n",
       "10                  14.0          4  \n",
       "18                  13.0          3  \n",
       "31                   0.0          4  \n",
       "99                   3.0          4  \n",
       "201                  6.0          4  \n",
       "...                  ...        ...  \n",
       "104270               0.0          4  \n",
       "104275               2.0          3  \n",
       "104279              10.0          3  \n",
       "104280               2.0          4  \n",
       "104288               4.0          4  \n",
       "\n",
       "[8001 rows x 9 columns]"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df[open_food_facts_df['labels_en'].str.find(\"preservatives\") > -1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    71866.000000\n",
       "mean        -0.503326\n",
       "std          5.985629\n",
       "min         -1.000000\n",
       "25%         -1.000000\n",
       "50%         -1.000000\n",
       "75%         -1.000000\n",
       "max        222.000000\n",
       "Name: labels_en, dtype: float64"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df['labels_en'].str.find(\"additives\").describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>brands_tags</th>\n",
       "      <th>labels_en</th>\n",
       "      <th>additives_tags</th>\n",
       "      <th>nutriscore_grade</th>\n",
       "      <th>pnns_groups_1</th>\n",
       "      <th>pnns_groups_2</th>\n",
       "      <th>additives_n</th>\n",
       "      <th>nutriscore_score</th>\n",
       "      <th>nova_group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>125</td>\n",
       "      <td>thai-kitchen</td>\n",
       "      <td>No additives</td>\n",
       "      <td>NaN</td>\n",
       "      <td>d</td>\n",
       "      <td>Beverages</td>\n",
       "      <td>Plant-based milk substitutes</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>261</td>\n",
       "      <td>lily-of-the-desert</td>\n",
       "      <td>Organic,No additives,No preservatives</td>\n",
       "      <td>en:e330</td>\n",
       "      <td>b</td>\n",
       "      <td>Beverages</td>\n",
       "      <td>Unsweetened beverages</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>669</td>\n",
       "      <td>old-dominion</td>\n",
       "      <td>Gluten-free,No additives,No preservatives</td>\n",
       "      <td>NaN</td>\n",
       "      <td>c</td>\n",
       "      <td>Salty snacks</td>\n",
       "      <td>Nuts</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>877</td>\n",
       "      <td>thomas-brioche</td>\n",
       "      <td>No colorings,No preservatives,No additives</td>\n",
       "      <td>NaN</td>\n",
       "      <td>d</td>\n",
       "      <td>Sugary snacks</td>\n",
       "      <td>pastries</td>\n",
       "      <td>0.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>978</td>\n",
       "      <td>carrefour</td>\n",
       "      <td>No additives</td>\n",
       "      <td>NaN</td>\n",
       "      <td>e</td>\n",
       "      <td>Sugary snacks</td>\n",
       "      <td>Biscuits and cakes</td>\n",
       "      <td>0.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>103258</td>\n",
       "      <td>vivera</td>\n",
       "      <td>No additives,With Sunflower oil</td>\n",
       "      <td>NaN</td>\n",
       "      <td>a</td>\n",
       "      <td>unknown</td>\n",
       "      <td>unknown</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-5.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>103336</td>\n",
       "      <td>body-fit</td>\n",
       "      <td>Low or no salt,Organic,EU Organic,EU/non-EU Ag...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>a</td>\n",
       "      <td>Sugary snacks</td>\n",
       "      <td>Sweets</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-2.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104222</td>\n",
       "      <td>meadow-fresh</td>\n",
       "      <td>No additives,No preservatives</td>\n",
       "      <td>NaN</td>\n",
       "      <td>b</td>\n",
       "      <td>unknown</td>\n",
       "      <td>unknown</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104225</td>\n",
       "      <td>mother-earth</td>\n",
       "      <td>No additives</td>\n",
       "      <td>NaN</td>\n",
       "      <td>d</td>\n",
       "      <td>Cereals and potatoes</td>\n",
       "      <td>Vegetables</td>\n",
       "      <td>0.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104243</td>\n",
       "      <td>bio-lanka</td>\n",
       "      <td>Organic,Vegetarian,EU Organic,Gluten-free,Vega...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>a</td>\n",
       "      <td>Fruits and vegetables</td>\n",
       "      <td>Fruits</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-5.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1087 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               brands_tags                                          labels_en  \\\n",
       "125           thai-kitchen                                       No additives   \n",
       "261     lily-of-the-desert              Organic,No additives,No preservatives   \n",
       "669           old-dominion          Gluten-free,No additives,No preservatives   \n",
       "877         thomas-brioche         No colorings,No preservatives,No additives   \n",
       "978              carrefour                                       No additives   \n",
       "...                    ...                                                ...   \n",
       "103258              vivera                    No additives,With Sunflower oil   \n",
       "103336            body-fit  Low or no salt,Organic,EU Organic,EU/non-EU Ag...   \n",
       "104222        meadow-fresh                      No additives,No preservatives   \n",
       "104225        mother-earth                                       No additives   \n",
       "104243           bio-lanka  Organic,Vegetarian,EU Organic,Gluten-free,Vega...   \n",
       "\n",
       "       additives_tags nutriscore_grade          pnns_groups_1  \\\n",
       "125               NaN                d              Beverages   \n",
       "261           en:e330                b              Beverages   \n",
       "669               NaN                c           Salty snacks   \n",
       "877               NaN                d          Sugary snacks   \n",
       "978               NaN                e          Sugary snacks   \n",
       "...               ...              ...                    ...   \n",
       "103258            NaN                a                unknown   \n",
       "103336            NaN                a          Sugary snacks   \n",
       "104222            NaN                b                unknown   \n",
       "104225            NaN                d   Cereals and potatoes   \n",
       "104243            NaN                a  Fruits and vegetables   \n",
       "\n",
       "                       pnns_groups_2  additives_n  nutriscore_score nova_group  \n",
       "125     Plant-based milk substitutes          0.0              11.0          1  \n",
       "261            Unsweetened beverages          1.0               1.0        NaN  \n",
       "669                             Nuts          0.0               3.0          3  \n",
       "877                         pastries          0.0              17.0          4  \n",
       "978               Biscuits and cakes          0.0              21.0          4  \n",
       "...                              ...          ...               ...        ...  \n",
       "103258                       unknown          0.0              -5.0          4  \n",
       "103336                        Sweets          0.0              -2.0          1  \n",
       "104222                       unknown          0.0               1.0          1  \n",
       "104225                    Vegetables          0.0              14.0          4  \n",
       "104243                        Fruits          0.0              -5.0          1  \n",
       "\n",
       "[1087 rows x 9 columns]"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df[open_food_facts_df['labels_en'].str.find(\"additives\") > -1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Au vu de la grande diversité des labels présents sur les produits. Tous ne seront pas utiles pour l'analyse et l'application, il sera intéressant de travailler sur les labels de type \"No additives\" ou \"No preservatives\"."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## *additives_tags* : liste d'additifs présents dans les produits"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "en:e322,en:e322i                      3834\n",
       "en:e330                               2802\n",
       "en:e440                               2540\n",
       "en:e14xx                              1969\n",
       "en:e300                               1870\n",
       "en:e250,en:e301                       1248\n",
       "en:e300,en:e330                        912\n",
       "en:e250                                818\n",
       "en:e407                                812\n",
       "en:e250,en:e316                        748\n",
       "en:e330,en:e440                        745\n",
       "en:e415                                593\n",
       "en:e252                                571\n",
       "en:e412                                502\n",
       "en:e202                                493\n",
       "en:e471                                487\n",
       "en:e412,en:e415                        458\n",
       "en:e14xx,en:e415                       393\n",
       "en:e160c                               357\n",
       "en:e160b                               330\n",
       "en:e500,en:e500ii                      323\n",
       "en:e224,en:e330                        306\n",
       "en:e270                                284\n",
       "en:e250,en:e252                        273\n",
       "en:e250,en:e252,en:e301                260\n",
       "en:e392                                260\n",
       "en:e14xx,en:e330                       242\n",
       "en:e322,en:e322i,en:e500,en:e500ii     235\n",
       "en:e322                                228\n",
       "en:e428                                225\n",
       "Name: additives_tags, dtype: int64"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df['additives_tags'].value_counts().head(30)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Analyses multivariées"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 104300 entries, 0 to 104299\n",
      "Data columns (total 9 columns):\n",
      "brands_tags         103844 non-null object\n",
      "labels_en           71866 non-null object\n",
      "additives_tags      73811 non-null object\n",
      "nutriscore_grade    104300 non-null object\n",
      "pnns_groups_1       104159 non-null object\n",
      "pnns_groups_2       104300 non-null object\n",
      "additives_n         104300 non-null float64\n",
      "nutriscore_score    104300 non-null float64\n",
      "nova_group          104067 non-null object\n",
      "dtypes: float64(2), object(7)\n",
      "memory usage: 7.2+ MB\n"
     ]
    }
   ],
   "source": [
    "open_food_facts_df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Y a-t-il plus d'additifs lorsque les produits ont une faible qualité nutritionnelle ?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, \"Nombre d'additifs\")"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.scatterplot(data=open_food_facts_df, x=\"additives_n\", y=\"nutriscore_score\")\n",
    "plt.title('Répartition des produits selon leur nombre d\\'additifs et leur score de Nutriscore')\n",
    "plt.ylabel('Score Nutriscore')\n",
    "plt.xlabel('Nombre d\\'additifs')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>additives_n</th>\n",
       "      <th>nutriscore_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>additives_n</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>0.18417</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>nutriscore_score</td>\n",
       "      <td>0.18417</td>\n",
       "      <td>1.00000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  additives_n  nutriscore_score\n",
       "additives_n           1.00000           0.18417\n",
       "nutriscore_score      0.18417           1.00000"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df.corr()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Il ne semble pas avoir de corrélation entre le nombre d'additifs et le score Nutriscore.\n",
    "\n",
    "Considérons plutôt ces variables comme qualitatives pour vérifier leur indépendance entre elles. Dressons alors le tableau de contingence."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>nutriscore_score</th>\n",
       "      <th>-14.0</th>\n",
       "      <th>-13.0</th>\n",
       "      <th>-12.0</th>\n",
       "      <th>-11.0</th>\n",
       "      <th>-10.0</th>\n",
       "      <th>-9.0</th>\n",
       "      <th>-8.0</th>\n",
       "      <th>-7.0</th>\n",
       "      <th>-6.0</th>\n",
       "      <th>-5.0</th>\n",
       "      <th>...</th>\n",
       "      <th>30.0</th>\n",
       "      <th>31.0</th>\n",
       "      <th>32.0</th>\n",
       "      <th>33.0</th>\n",
       "      <th>34.0</th>\n",
       "      <th>35.0</th>\n",
       "      <th>36.0</th>\n",
       "      <th>37.0</th>\n",
       "      <th>40.0</th>\n",
       "      <th>Total</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>additives_n</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>41.0</td>\n",
       "      <td>69.0</td>\n",
       "      <td>107.0</td>\n",
       "      <td>147.0</td>\n",
       "      <td>211.0</td>\n",
       "      <td>334.0</td>\n",
       "      <td>891.0</td>\n",
       "      <td>926.0</td>\n",
       "      <td>...</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>30489</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>58.0</td>\n",
       "      <td>164.0</td>\n",
       "      <td>350.0</td>\n",
       "      <td>...</td>\n",
       "      <td>12.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>24228</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>71.0</td>\n",
       "      <td>224.0</td>\n",
       "      <td>...</td>\n",
       "      <td>12.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>16296</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>46.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>10879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>...</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7521</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>...</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5006</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>...</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3318</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2192</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1513</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>657</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>365</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>263</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>13.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>173</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>151</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>19.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>20.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>21.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>22.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>23.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>24.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>25.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>26.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>28.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>30.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Total</td>\n",
       "      <td>4.0</td>\n",
       "      <td>16.0</td>\n",
       "      <td>46.0</td>\n",
       "      <td>73.0</td>\n",
       "      <td>124.0</td>\n",
       "      <td>170.0</td>\n",
       "      <td>262.0</td>\n",
       "      <td>438.0</td>\n",
       "      <td>1179.0</td>\n",
       "      <td>1616.0</td>\n",
       "      <td>...</td>\n",
       "      <td>42.0</td>\n",
       "      <td>25.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>104300</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>30 rows × 54 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "nutriscore_score  -14.0  -13.0  -12.0  -11.0  -10.0   -9.0   -8.0   -7.0  \\\n",
       "additives_n                                                                \n",
       "0.0                 4.0   15.0   41.0   69.0  107.0  147.0  211.0  334.0   \n",
       "1.0                 NaN    1.0    3.0    4.0   13.0   15.0   24.0   58.0   \n",
       "2.0                 NaN    NaN    2.0    NaN    4.0    4.0   14.0   27.0   \n",
       "3.0                 NaN    NaN    NaN    NaN    NaN    3.0    7.0   10.0   \n",
       "4.0                 NaN    NaN    NaN    NaN    NaN    1.0    1.0    2.0   \n",
       "5.0                 NaN    NaN    NaN    NaN    NaN    NaN    1.0    2.0   \n",
       "6.0                 NaN    NaN    NaN    NaN    NaN    NaN    2.0    2.0   \n",
       "7.0                 NaN    NaN    NaN    NaN    NaN    NaN    1.0    1.0   \n",
       "8.0                 NaN    NaN    NaN    NaN    NaN    NaN    NaN    2.0   \n",
       "9.0                 NaN    NaN    NaN    NaN    NaN    NaN    NaN    NaN   \n",
       "10.0                NaN    NaN    NaN    NaN    NaN    NaN    1.0    NaN   \n",
       "11.0                NaN    NaN    NaN    NaN    NaN    NaN    NaN    NaN   \n",
       "12.0                NaN    NaN    NaN    NaN    NaN    NaN    NaN    NaN   \n",
       "13.0                NaN    NaN    NaN    NaN    NaN    NaN    NaN    NaN   \n",
       "14.0                NaN    NaN    NaN    NaN    NaN    NaN    NaN    NaN   \n",
       "15.0                NaN    NaN    NaN    NaN    NaN    NaN    NaN    NaN   \n",
       "16.0                NaN    NaN    NaN    NaN    NaN    NaN    NaN    NaN   \n",
       "17.0                NaN    NaN    NaN    NaN    NaN    NaN    NaN    NaN   \n",
       "18.0                NaN    NaN    NaN    NaN    NaN    NaN    NaN    NaN   \n",
       "19.0                NaN    NaN    NaN    NaN    NaN    NaN    NaN    NaN   \n",
       "20.0                NaN    NaN    NaN    NaN    NaN    NaN    NaN    NaN   \n",
       "21.0                NaN    NaN    NaN    NaN    NaN    NaN    NaN    NaN   \n",
       "22.0                NaN    NaN    NaN    NaN    NaN    NaN    NaN    NaN   \n",
       "23.0                NaN    NaN    NaN    NaN    NaN    NaN    NaN    NaN   \n",
       "24.0                NaN    NaN    NaN    NaN    NaN    NaN    NaN    NaN   \n",
       "25.0                NaN    NaN    NaN    NaN    NaN    NaN    NaN    NaN   \n",
       "26.0                NaN    NaN    NaN    NaN    NaN    NaN    NaN    NaN   \n",
       "28.0                NaN    NaN    NaN    NaN    NaN    NaN    NaN    NaN   \n",
       "30.0                NaN    NaN    NaN    NaN    NaN    NaN    NaN    NaN   \n",
       "Total               4.0   16.0   46.0   73.0  124.0  170.0  262.0  438.0   \n",
       "\n",
       "nutriscore_score    -6.0    -5.0  ...  30.0  31.0  32.0  33.0  34.0  35.0  \\\n",
       "additives_n                       ...                                       \n",
       "0.0                891.0   926.0  ...   6.0   6.0   4.0   4.0   1.0   1.0   \n",
       "1.0                164.0   350.0  ...  12.0   9.0   5.0   2.0   3.0   NaN   \n",
       "2.0                 71.0   224.0  ...  12.0   3.0   1.0   NaN   2.0   2.0   \n",
       "3.0                 20.0    46.0  ...   1.0   3.0   1.0   NaN   1.0   NaN   \n",
       "4.0                  8.0    38.0  ...   5.0   4.0   1.0   2.0   NaN   2.0   \n",
       "5.0                 13.0     9.0  ...   2.0   NaN   1.0   NaN   NaN   NaN   \n",
       "6.0                  5.0     8.0  ...   2.0   NaN   NaN   1.0   NaN   1.0   \n",
       "7.0                  3.0     9.0  ...   1.0   NaN   NaN   NaN   NaN   NaN   \n",
       "8.0                  2.0     1.0  ...   1.0   NaN   NaN   NaN   NaN   NaN   \n",
       "9.0                  2.0     2.0  ...   NaN   NaN   NaN   NaN   NaN   NaN   \n",
       "10.0                 NaN     1.0  ...   NaN   NaN   NaN   NaN   NaN   NaN   \n",
       "11.0                 NaN     2.0  ...   NaN   NaN   NaN   NaN   NaN   NaN   \n",
       "12.0                 NaN     NaN  ...   NaN   NaN   NaN   NaN   NaN   NaN   \n",
       "13.0                 NaN     NaN  ...   NaN   NaN   NaN   NaN   NaN   NaN   \n",
       "14.0                 NaN     NaN  ...   NaN   NaN   NaN   NaN   NaN   NaN   \n",
       "15.0                 NaN     NaN  ...   NaN   NaN   NaN   NaN   NaN   NaN   \n",
       "16.0                 NaN     NaN  ...   NaN   NaN   NaN   NaN   NaN   NaN   \n",
       "17.0                 NaN     NaN  ...   NaN   NaN   NaN   NaN   NaN   NaN   \n",
       "18.0                 NaN     NaN  ...   NaN   NaN   NaN   NaN   NaN   NaN   \n",
       "19.0                 NaN     NaN  ...   NaN   NaN   NaN   NaN   NaN   NaN   \n",
       "20.0                 NaN     NaN  ...   NaN   NaN   NaN   NaN   NaN   NaN   \n",
       "21.0                 NaN     NaN  ...   NaN   NaN   NaN   NaN   NaN   NaN   \n",
       "22.0                 NaN     NaN  ...   NaN   NaN   NaN   NaN   NaN   NaN   \n",
       "23.0                 NaN     NaN  ...   NaN   NaN   NaN   NaN   NaN   NaN   \n",
       "24.0                 NaN     NaN  ...   NaN   NaN   NaN   NaN   NaN   NaN   \n",
       "25.0                 NaN     NaN  ...   NaN   NaN   NaN   NaN   NaN   NaN   \n",
       "26.0                 NaN     NaN  ...   NaN   NaN   NaN   NaN   NaN   NaN   \n",
       "28.0                 NaN     NaN  ...   NaN   NaN   NaN   NaN   NaN   NaN   \n",
       "30.0                 NaN     NaN  ...   NaN   NaN   NaN   NaN   NaN   NaN   \n",
       "Total             1179.0  1616.0  ...  42.0  25.0  13.0   9.0   7.0   6.0   \n",
       "\n",
       "nutriscore_score  36.0  37.0  40.0   Total  \n",
       "additives_n                                 \n",
       "0.0                3.0   2.0   NaN   30489  \n",
       "1.0                3.0   NaN   1.0   24228  \n",
       "2.0                NaN   1.0   NaN   16296  \n",
       "3.0                NaN   1.0   1.0   10879  \n",
       "4.0                NaN   NaN   NaN    7521  \n",
       "5.0                NaN   NaN   NaN    5006  \n",
       "6.0                NaN   NaN   NaN    3318  \n",
       "7.0                NaN   NaN   NaN    2192  \n",
       "8.0                NaN   NaN   NaN    1513  \n",
       "9.0                NaN   NaN   NaN     999  \n",
       "10.0               NaN   NaN   NaN     657  \n",
       "11.0               NaN   NaN   NaN     365  \n",
       "12.0               NaN   NaN   NaN     263  \n",
       "13.0               NaN   NaN   NaN     173  \n",
       "14.0               NaN   NaN   NaN     151  \n",
       "15.0               NaN   NaN   NaN      96  \n",
       "16.0               NaN   NaN   NaN      52  \n",
       "17.0               NaN   NaN   NaN      37  \n",
       "18.0               NaN   NaN   NaN      22  \n",
       "19.0               NaN   NaN   NaN      13  \n",
       "20.0               NaN   NaN   NaN       8  \n",
       "21.0               NaN   NaN   NaN       5  \n",
       "22.0               NaN   NaN   NaN       7  \n",
       "23.0               NaN   NaN   NaN       1  \n",
       "24.0               NaN   NaN   NaN       2  \n",
       "25.0               NaN   NaN   NaN       1  \n",
       "26.0               NaN   NaN   NaN       1  \n",
       "28.0               NaN   NaN   NaN       2  \n",
       "30.0               NaN   NaN   NaN       3  \n",
       "Total              6.0   4.0   2.0  104300  \n",
       "\n",
       "[30 rows x 54 columns]"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = \"additives_n\"\n",
    "Y = \"nutriscore_score\"\n",
    "tableau_contingence = open_food_facts_df[[X, Y]].pivot_table(index=X, \n",
    "                                                             columns=Y, \n",
    "                                                             aggfunc=len, \n",
    "                                                             margins=True,\n",
    "                                                             margins_name=\"Total\")\n",
    "\n",
    "tableau_contingence"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Au vu du tableau de contingence, il est préférable de créer des classes pour la variable 'nombre d'additifs' et d'utiliser la classe du nutriscore au lieu du score.\n",
    "\n",
    "Création des classes du nombre d'additifs :"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "open_food_facts_df.loc[open_food_facts_df['additives_n'] == 0, 'additives_n_classes'] = 'Pas d\\'additif'\n",
    "open_food_facts_df.loc[(open_food_facts_df['additives_n'] == 1) | (open_food_facts_df['additives_n'] == 2), \n",
    "                   'additives_n_classes'] = '1 ou 2 additifs'\n",
    "open_food_facts_df.loc[(open_food_facts_df['additives_n'] >= 3) & (open_food_facts_df['additives_n'] <= 7), \n",
    "                   'additives_n_classes'] = 'Entre 3 et 7 additifs'\n",
    "open_food_facts_df.loc[open_food_facts_df['additives_n'] > 7, 'additives_n_classes'] = 'Plus de 7 additifs'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>additives_n</th>\n",
       "      <th>additives_n_classes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>17239</th>\n",
       "      <td>2.0</td>\n",
       "      <td>1 ou 2 additifs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56612</th>\n",
       "      <td>2.0</td>\n",
       "      <td>1 ou 2 additifs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>74767</th>\n",
       "      <td>5.0</td>\n",
       "      <td>Entre 3 et 7 additifs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1 ou 2 additifs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57632</th>\n",
       "      <td>0.0</td>\n",
       "      <td>Pas d'additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14486</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1 ou 2 additifs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>83192</th>\n",
       "      <td>5.0</td>\n",
       "      <td>Entre 3 et 7 additifs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29480</th>\n",
       "      <td>3.0</td>\n",
       "      <td>Entre 3 et 7 additifs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39495</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1 ou 2 additifs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2787</th>\n",
       "      <td>2.0</td>\n",
       "      <td>1 ou 2 additifs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10626</th>\n",
       "      <td>0.0</td>\n",
       "      <td>Pas d'additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>86501</th>\n",
       "      <td>0.0</td>\n",
       "      <td>Pas d'additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23524</th>\n",
       "      <td>5.0</td>\n",
       "      <td>Entre 3 et 7 additifs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32914</th>\n",
       "      <td>0.0</td>\n",
       "      <td>Pas d'additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51753</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1 ou 2 additifs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31065</th>\n",
       "      <td>2.0</td>\n",
       "      <td>1 ou 2 additifs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67640</th>\n",
       "      <td>0.0</td>\n",
       "      <td>Pas d'additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>103510</th>\n",
       "      <td>3.0</td>\n",
       "      <td>Entre 3 et 7 additifs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>88927</th>\n",
       "      <td>0.0</td>\n",
       "      <td>Pas d'additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93191</th>\n",
       "      <td>0.0</td>\n",
       "      <td>Pas d'additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35890</th>\n",
       "      <td>0.0</td>\n",
       "      <td>Pas d'additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76692</th>\n",
       "      <td>0.0</td>\n",
       "      <td>Pas d'additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22814</th>\n",
       "      <td>0.0</td>\n",
       "      <td>Pas d'additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>72888</th>\n",
       "      <td>3.0</td>\n",
       "      <td>Entre 3 et 7 additifs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>88152</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1 ou 2 additifs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24849</th>\n",
       "      <td>5.0</td>\n",
       "      <td>Entre 3 et 7 additifs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>153</th>\n",
       "      <td>4.0</td>\n",
       "      <td>Entre 3 et 7 additifs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30202</th>\n",
       "      <td>4.0</td>\n",
       "      <td>Entre 3 et 7 additifs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56399</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1 ou 2 additifs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45468</th>\n",
       "      <td>4.0</td>\n",
       "      <td>Entre 3 et 7 additifs</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        additives_n    additives_n_classes\n",
       "17239           2.0        1 ou 2 additifs\n",
       "56612           2.0        1 ou 2 additifs\n",
       "74767           5.0  Entre 3 et 7 additifs\n",
       "47              1.0        1 ou 2 additifs\n",
       "57632           0.0          Pas d'additif\n",
       "14486           1.0        1 ou 2 additifs\n",
       "83192           5.0  Entre 3 et 7 additifs\n",
       "29480           3.0  Entre 3 et 7 additifs\n",
       "39495           1.0        1 ou 2 additifs\n",
       "2787            2.0        1 ou 2 additifs\n",
       "10626           0.0          Pas d'additif\n",
       "86501           0.0          Pas d'additif\n",
       "23524           5.0  Entre 3 et 7 additifs\n",
       "32914           0.0          Pas d'additif\n",
       "51753           1.0        1 ou 2 additifs\n",
       "31065           2.0        1 ou 2 additifs\n",
       "67640           0.0          Pas d'additif\n",
       "103510          3.0  Entre 3 et 7 additifs\n",
       "88927           0.0          Pas d'additif\n",
       "93191           0.0          Pas d'additif\n",
       "35890           0.0          Pas d'additif\n",
       "76692           0.0          Pas d'additif\n",
       "22814           0.0          Pas d'additif\n",
       "72888           3.0  Entre 3 et 7 additifs\n",
       "88152           1.0        1 ou 2 additifs\n",
       "24849           5.0  Entre 3 et 7 additifs\n",
       "153             4.0  Entre 3 et 7 additifs\n",
       "30202           4.0  Entre 3 et 7 additifs\n",
       "56399           1.0        1 ou 2 additifs\n",
       "45468           4.0  Entre 3 et 7 additifs"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df[['additives_n', 'additives_n_classes']].sample(30)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0     30489\n",
       "1.0     24228\n",
       "2.0     16296\n",
       "3.0     10879\n",
       "4.0      7521\n",
       "5.0      5006\n",
       "6.0      3318\n",
       "7.0      2192\n",
       "8.0      1513\n",
       "9.0       999\n",
       "10.0      657\n",
       "11.0      365\n",
       "12.0      263\n",
       "13.0      173\n",
       "14.0      151\n",
       "15.0       96\n",
       "16.0       52\n",
       "17.0       37\n",
       "18.0       22\n",
       "19.0       13\n",
       "20.0        8\n",
       "22.0        7\n",
       "21.0        5\n",
       "30.0        3\n",
       "24.0        2\n",
       "28.0        2\n",
       "26.0        1\n",
       "23.0        1\n",
       "25.0        1\n",
       "Name: additives_n, dtype: int64"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df['additives_n'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1 ou 2 additifs          40524\n",
       "Pas d'additif            30489\n",
       "Entre 3 et 7 additifs    28916\n",
       "Plus de 7 additifs        4371\n",
       "Name: additives_n_classes, dtype: int64"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df['additives_n_classes'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RangeIndex(start=0, stop=104300, step=1)"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df['additives_n_classes'].index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Score Nutri-Score')"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "meanprops = {'marker':'o', 'markeredgecolor':'firebrick',\n",
    "            'markerfacecolor':'firebrick'}\n",
    "\n",
    "plt.rcParams[\"figure.figsize\"] = (12, 8)\n",
    "sns.boxplot(x='additives_n_classes', y='nutriscore_score', \n",
    "            data=open_food_facts_df, color='tab:blue',\n",
    "            order=['Pas d\\'additif', '1 ou 2 additifs', 'Entre 3 et 7 additifs', 'Plus de 7 additifs'],\n",
    "            showmeans=True,\n",
    "            meanprops=meanprops)\n",
    "plt.title('Relation du score Nutriscore avec le nombre d\\'additifs parmi les produits')\n",
    "plt.xlabel('Nombre d\\'additifs')\n",
    "plt.ylabel('Score Nutri-Score')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Le score moyen et médian est plus faible parmi les produits sans additif. Il semble donc que ces produits aient une meilleure qualité nutritionnelle."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Confirmons-le avec une ANOVA en commençant par vérifier les hypothèses suivantes :\n",
    "- chaque classe *k* doit suivre une loi normale de moyenne m*k* et de variance v\n",
    "- chaque classe doit avoir la même variance v"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Classe 'Pas d\\'additif'*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(-3.8249999999999957, 0.5, 'Nombre de produits')"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Distribution du score final Nutri-Score parmi la classe 'Pas d'additifs'\n",
    "filtered_data = open_food_facts_df[open_food_facts_df['additives_n_classes'] == 'Pas d\\'additif']\n",
    "nombre_produits = filtered_data['nutriscore_score'].count()\n",
    "sns.displot(x='nutriscore_score', data=filtered_data, discrete=True)\n",
    "plt.title('Répartition des {} produits sans additif selon leur score Nutri-Score'.format(nombre_produits))\n",
    "plt.xlabel('Score Nutri-Score')\n",
    "plt.ylabel('Nombre de produits')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    30489.000000\n",
       "mean         6.094198\n",
       "std          8.751638\n",
       "min        -14.000000\n",
       "25%         -1.000000\n",
       "50%          4.000000\n",
       "75%         13.000000\n",
       "max         37.000000\n",
       "Name: nutriscore_score, dtype: float64"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filtered_data['nutriscore_score'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.40242696396146305"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "st.skew(filtered_data['nutriscore_score'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-0.7744059261756844"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "st.kurtosis(filtered_data['nutriscore_score'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Graphiquement, la distribution est bimodale et ne semble pas suivre une loi normale, cependant le skewness et le kurtosis sont proches de 0. Testons avec un test d'Agostino si cette classe suit la loi normale."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NormaltestResult(statistic=3057.6414829337646, pvalue=0.0)"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "st.normaltest(filtered_data['nutriscore_score'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Au niveau de test 5%, nous rejetons l'hypothèse affirmant que la distribution suit une loi normale."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "La première hypothèse citée ci-avant pour réaliser l'ANOVA n'est pas vérifiée. De ce fait, nous allons utiliser la variable 'nutriscore_grade' pour pouvoir répondre à l'interrogation. Ceci étant posé nous allons calculer le tableau de contingence."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>additives_n_classes</th>\n",
       "      <th>1 ou 2 additifs</th>\n",
       "      <th>Entre 3 et 7 additifs</th>\n",
       "      <th>Pas d'additif</th>\n",
       "      <th>Plus de 7 additifs</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>nutriscore_grade</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>a</td>\n",
       "      <td>4875</td>\n",
       "      <td>1629</td>\n",
       "      <td>7800</td>\n",
       "      <td>170</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>b</td>\n",
       "      <td>5028</td>\n",
       "      <td>3423</td>\n",
       "      <td>5246</td>\n",
       "      <td>417</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>c</td>\n",
       "      <td>9482</td>\n",
       "      <td>6392</td>\n",
       "      <td>6280</td>\n",
       "      <td>862</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>d</td>\n",
       "      <td>10812</td>\n",
       "      <td>10077</td>\n",
       "      <td>7568</td>\n",
       "      <td>2109</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>e</td>\n",
       "      <td>10327</td>\n",
       "      <td>7395</td>\n",
       "      <td>3595</td>\n",
       "      <td>813</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "additives_n_classes  1 ou 2 additifs  Entre 3 et 7 additifs  Pas d'additif  \\\n",
       "nutriscore_grade                                                             \n",
       "a                               4875                   1629           7800   \n",
       "b                               5028                   3423           5246   \n",
       "c                               9482                   6392           6280   \n",
       "d                              10812                  10077           7568   \n",
       "e                              10327                   7395           3595   \n",
       "\n",
       "additives_n_classes  Plus de 7 additifs  \n",
       "nutriscore_grade                         \n",
       "a                                   170  \n",
       "b                                   417  \n",
       "c                                   862  \n",
       "d                                  2109  \n",
       "e                                   813  "
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = 'nutriscore_grade'\n",
    "Y = 'additives_n_classes'\n",
    "\n",
    "tableau_contingence = open_food_facts_df[[X, Y]].pivot_table(index=X, \n",
    "                                                             columns=Y, \n",
    "                                                             aggfunc=len)\n",
    "\n",
    "tableau_contingence"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Testons alors l'indépendance entre ces deux variables avec le test du khi-deux."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(8408.717208004593,\n",
       " 0.0,\n",
       " 12,\n",
       " array([[ 5623.62776606,  4012.75344199,  4231.04301055,   606.5757814 ],\n",
       "        [ 5483.7558581 ,  3912.9474976 ,  4125.80772771,   591.48891659],\n",
       "        [ 8942.47731544,  6380.92671141,  6728.04241611,   964.55355705],\n",
       "        [11875.90205177,  8474.07915628,  8935.0601534 ,  1280.95863854],\n",
       "        [ 8598.23700863,  6135.29319271,  6469.04669223,   927.42310642]]))"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "st.chi2_contingency(tableau_contingence)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "À 5%, nous rejetons l'hypothèse nulle affirmant que les variables 'nutriscore_grade' et 'additives_n_classes' sont indépendantes. Par conséquent, il y a une dépendance significative entre le nombre d'additifs présents dans les produits et leur qualité nutritionnelle."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, \"Nombre d'additifs\")"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtQAAAH1CAYAAADWNWn+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOzdfXxcZZn4/0+aSMEWEOLDypNhjdy6tq4aVNYHNi3FRirqorii2LiufldEyqLuqlDlYQuL+0OFIiqLsqQKrrKiCNTUloKu66pLFG0Ubo0SBR8QB0RaIJCH3x/nTJmZTpJJJ5kzk/m8X6+8kvucyZxrZu45c8117nOflomJCSRJkiTtngVZByBJkiQ1MhNqSZIkqQom1JIkSVIVTKglSZKkKphQS5IkSVUwoZYkSZKqYEKtuhZC6A4h3FXF/38qhPDB2Yxpku3sdpwhhCtCCOtmOZ63hBC+NZv3We9CCC8LIcSs46hXIYSOEMJECKEt61imEkIYDiGsqOH2bg4hvG0W7mfK91zhdkIIbwohfH2K2xb15ZD4QQjhgRDCmmpjrZUQwukhhE9Psq6qfXsWZvIeCiEcEkLYHkJorUVsyl5d71g1P4QQhoGnAGPAdqAfeFeMcfssb+ctwNtijC/NL4sxvmM2t6GZCyGcBZwJvD7GeHW6rA14FDg0xjhcwX1MAM+IMQ5NdpsY438DYTZiVnNK9yHdMca3zOV2YoxXAlcWbLeof5fpy/8M3BxjfN5cxjXbYoznZR1DVmKMvwIW59shhJuBz8UYy37BUOOzQq1aOTbGuBh4LvA84AMZx6Pauhc4Z66qNbWuuoYQWkII7j8zVu/V9ln0NODHWQYwH59rq8eaTfPuDaL6FmP8XQhhE0liDUAIYSFwLvB6YCHwZeC0GONDpf8fQng/8HbgycCdwBkxxi+HEJ4FfAp4XAhhOzAaY3xCCOEK4K4Y49r0/98OvA/YH/gW8I4Y42/SdRPAScB7gCcCV5FU0ne5nGgIYS/gk8Crgd8C/1Gy/gDgYuBIkqr8x2KM66d7fkII+wGfBV5E8v78nzTGsodGQwgHAxcBLyP5gvz5GOO7ytzuIuA4YF/gZ8A/plUwQggvBD4BHAY8BFwZY3x3CGFP4NPAK4DW9P9eGWO8O4SwL/BR4BhgPH38Z8YYxyZ5aP3AEuBEoK9MfDdTUL0pPNoQQvhmerMfpq/R3wN3A58jeY5PAzaHED6T3sdB6X28D1gD7AP8BnhnjPHG9EP0fen9PBn4KfCaGOOdIYQXp8/nYenyU2OM3y6I8X+AbuD5wNIQwj2VPg/p83wR8Kz0ef4S8O4Y4yMhhE8B22OM7y24/bXAN2KMH52qP031eCZ5LfL3X/FrmB5l+AvgYeBvgF8BvTHGW9L1zyJ5PzwX+DXwgRjjV9N1VwAPAoeS9NMfAq8F3g/0kryWJ8QYf1CwyReEENYDTwW+ApwUY3w4hNBNyesOvDmE8EpgHdAB/ITkPfOjSR730en/P5XkvdYyxXN0dRrzXmncJ8UYf5yua0+fs27gdmBTpduptH/HGA8KIWwF/hp4aQjhQpK+1wlcABwM/ImkP1xQJv63kOwvvw+sJtlXnRxjvDFd/3ck1e+DgHuAD8cYL03XdTPJewxYD7yX5KjjScAjwIUk+80L8pXptN90xhhPnOw5Loh1qj5edh9V5j7yMX8CeHd6P2ekRwTyffEhki8ofw28OoTwf+l2X0HSTy8DzosxjqfvrQ8Db0mf54+UbG+Y5HXcUvp4QwgdwB3A44CzSfrREelreAVwCsn7700kn3u/BN4YYxyc7rlSfbLCopoKIRxEsuMqPHT/YZId5XNJPigOBD40yV38nGTHtC/JTupzIYSnxhhvA94B/G+McXGM8Qlltr0c+FeSxP2pJDuw/yy52SuBFwB/md5u5SRxnAk8Pf1ZSZIY5LezALiO5AP4QOAo4B9DCJPdV6EFJB/STwMOIdn5f7zcDdOd/fXp4+hIt1X6ePL+j+T53Z/ki8LVacIMSZJ3UYxxn/TxfDFd3kvyPB8MtJM8v/kvOX3AKMnr9Tzg5cBU41AngA8CZ4YQHjfF7XYRYzwy/fMv09f2C2n7z9LH8zTg/xX+TwghAO8CXhBj3JvkNRpOV78bOIEkkdwHeCvwYAhhf+AGkmShneTD7oY0ccp7c7qtvUme95k8D2MkickTgb8i6RfvTNddBfxtCKEljX+/9L7+s4L+VPbxTBJDoZm+hq8i6V9PAL5K2i/T1/M64OskCf0pwJXpa5D3emBt+thHgP8lSfKeCPwXyXNd6E0kr9nTSfYNawvWFb3uIYTnA5cD/0Dyul0KfDX9ol4khPBEki8y+Vh+Drwkvz7GeEXJcI+vAc9IH9f3KRimAVxC8gXjqSTP+Vsr3U6hKfp3fv1y4L9JvtwvjjH+FPgM8A9p314CbC1336kXAb9I4zgTuCbt6wC/J9nn7QP8HfCx9PnMK/ce+zNgTx7bT19G8kW5i2Tf/KEQwp9PEc8uKujjk+2jyvmz9LEeSLIP+/eSvvhGkgLO3iRFlYtJ9nN/TpJkr06fC0i+jLyS5P1xOPC6mTyuvBjjGRS/hu8ieb8dSdK/nwD8LZDbnftXfbBCrVr5Slp9WUyy8z8TkkPnJDut58QY702XnUeSYOwyLCQ/Bjf1hRDCB4AXAtdWEMObgMtjjN9Pt/MB4L4QQkfBON7zY4x/BP4YQriJJAntL3NfryepeN4L3JtW0/JfAl4APCnGeE7a/kUI4TLgDZRUsco8vhzJBzFpjOcCN01y8xcCBwD/FGMcTZeVPSkqxvi5guZHQghrScZo/pBkLHNnCOGJMcY/AN9Jb/coSYLSmVb7BtKYnkLypegJ6VGEHSGEj5F84F46xWP7agjhDJKk7bLJbjcD4yQV1ZE0rsJ1YyRVn78IIdxTMk77bcA/xxjzJ339MP3/NwM/izF+Nl3++ZCcAHYsSUUJ4IqCCuWMnocY40BBcziEcCnJB/iFJB+2EyQJyTdJPrj/N8b4mxDCi5i6P5V9PFPZzdfwWzHGjen/fxb4x3T5ESTv6/NjjOPA1hDC9SRJ/lnpbb6cf/whhC+TvHc2pO0vkHz5KfTxfIU9fQ9czGNJdenr/nbg0hjjd9P1fSGE09O4vlFyv8cAP4kx/lf6vxeSHJEqK8Z4ef7vtPp4X1rZ305SZV8aY9wBDIYQ+kgSpBlvZzc8StK3fxhjvA+4b4rb/h64MD3S9oUQwnuAVcBnY4w3FNzuGyE5UfJlJF8eoPx77FHg3BjjWAjhP4F/J0l2HwB+HEL4MfAckiS+UtPtMyfbR03mg2nM3wgh3ECyv/6XdN21Mcb/SR/PoySJ7PPS+B8IIXyE5IvzZ9L/u7CgL/4ryRGJ2fAoSVL/TOB7aVFIDcyEWrXymhjjlhDCX5Mky08E/gg8CXg8MFCQELWQDDHYRQhhNUlFriNdtDi9r0ocwGMfFMQYt4cQciSVjOF08e8Kbv8gBSeVlLmvwkPqvyz4+2nAASGEPxYsayVJmqYUQng88DGgB9gvXbx3CKG1zKH4g4FfFiTTU93ve0gSrwNIErd9eOx5+3vgHOD2EMIdwNkxxutJDlMfTFIlfQLJodQz0sf3OOC3Ba/ZAoqfj8msJanAf3a6G1bgnhjjw+VWxBiHQgj/SJLQPTskw4zenQ7vOZikYljqAIpfR9L2gQXtwsc4o+chhHAYSSX2cJI+30b6JSXGOJEmJyeQJNRvJHm+89uZqj9N9nimsjuvYel7Y8+QjKs9ALgzTabzSp+3uwv+fqhMu/R9VvreOqCgXfq6Pw3oDSGcUrBsj5L/ySt636bP+2SvVytJJfN4kv1U/vE9kWQISFuZOGe8nd30WpL30vkhhB8B748x/u8kt/11LB62tvP5DCG8gqS4cRjJ6/94YFvBbcu9x3IF+6L8EavpXs/pTNfHJ9tHlXNf+iUnr7T/FL4OTyTpK78suX2+7061n69KjHFrCOHjJEc6Dkm/aL43xvin2dqGasuEWjUVY/xGOo7tAuA1wB9IdsDPjjH+eqr/DSE8jaSyeRRJ9W4shHArj41N3GWsc4nfkOy48/e3iKQCO+V2J/FbkkQmf6LQIQXr7gTuiDE+Yzfu9z0kleMXxWS8+XOBH1B+nOedJDvitqmS6hDCy0jG2B4F/DgdG3hf/j5jjD8DTkgPux4H/FcIoT39UDobODsdD7gRiOnvEeCJlSTzhWKMm0MIQzw21CFvB8mHed6fVXB3U77eMcargKtCCPuQVF0/TFJ5upPksHHpWMWi/pE6hOIjFIXbvJOZPQ+fJHktT4gxPpAm/IWHkD8PfD2EcD7JYfq/KdjOVP1pssczlZnGPpXfAAeHEBYUJNWHkIzl3l0HF/x9SLqNvNLX/U6Sium5Fdxv/n0L7DxCdvAkt30jyTkSK0i+cO9LUgluIRlvPJr+7+0Fce7OdmYsxvh/JON/H0dS3f/iFPd/YAihpSCpPoTHhsR8iWSIw7UxxkdDCF+heF8z3T51tkzZx6fZR5XaL4SwqGDdIRS/Nwof0x9IKsVPIxl7n799/jOh6HWk+DWGme23dnku0zHi60MITyZ5Df+JZGicGpAJtbJwIckh7+fGGG9ND+19LITwrhjj70MIBwJLYoylwyMWkeyU7oGdJ9QsKVh/N3BQCGGPGOMjZbZ7FUm19SrgNuA84Luxgmnbyvgi8IEQwnfTuAqrY98D/hSSk+LWk5yw8yxgr/SDcCp7k3zB+GM6zvHMKW77PZId/vkhhDNJhjl05Q9nltznKMnz1haSEzv3ya8MIZwIbIox3lNQIRoLISwj+cD5CckJOY8CYzHG36aHhj8Skjm+t5OccHZQjLH0EHs5Z7DrEJ1bgeNCMmftATx2Ylbe3SRjHCedNq9QSMquB5KcRPgwyXOaP2fk08C/hBB+kt7fUpIP0I3AxSGEN5K8vq8lORGvbCVsN56HvUmex+0hhGeSnMx1T8H9/SAkJzl+muT1yL8W0/Wnso8nHT5U1iy8hoW+S5JY/HN6uPwlJMNkXjDD+yl0cjps5EHgdOALU9z2MuDLIYQtJM/V40kOy38zPYxf6Abg4yGE40jGgZ/M5EnQ3iRfOnLpfe6cAi79Mn8NcFYI4a0kR8x6eexI10y2AzPo3yGEPUiq5tfHGO8PIfyJ5L0/mScDa0IInyApYjyLpK/vQTIs6h5gNK1Wv5yZfTGbLVP28cn2UVPc39npsJ8XkYyBLrsfTV/HLwLnpkc/9yc5Apo/wfOLJM/d9SR9/P0ld3Er8IYQwtdIzrt5HeWHCMJjrzEAIYQXkOyTvp/e98PTPCbVOU9KVM3FGO8BNvDYN/H3kXyQfCf9cNhCmfmEY4w/ITnL+n9Jdk5LSRKmvK0kFePfhRD+UOb/b0y3+SWSRPTpJGP0dsfZJIf/7iA5GWvnEIb0cOixJOOv7yBJSj9NUuGazoUkh5Pz4wQn2zkXbqeTZNaFu0jGA5baRHJy1U/TmB+m+DBmD8nYx+0kJ/+8IT3M+2ckJ4z9ieQLyDd4bBjCapIP5J+QVO3+i+TkrGmlCf/3ShZ/jORD9G6Sk+WuLFl/FsnY2D+GEF5fwWYWAueTPI+/I0kqTk/XfZTkg/Lr6WP7DMkHd47kw/c9JEnUP5PMarJLXyowk+fhvSRVzwdIksBySeLnSSqiV+UXVNCfyj6eKWLendgnlX55fRXJmOw/kMywsDrGePuU/zi1q0gezy/Sn0kvfBSTmUbeTnKS5H0k+5K3THLbP5Ako+eTvMbPoHgfUmgDyfvl1yTPUem43XeRDG34HckY+50z/cxwOzDz/v1mkqLEn0hOFp5qFo3vptv/A8kQltfFGHPpl401JH3nPpK++dUKtj3rKujjk+2jyvkdyeP5Dcl+5B3T9MVTSBLaX5Ccg3IVyUmukLxPN5Gcl/B94JqS//0gyefIfSSfCVcxuYuA14UQ7gvJOTf7pPd/H0k/y/FYIq8G1DIxUasjOpIkqVZCmYtdzWchnTYvplNnSrVkhVqSJEmqggm1JEmSVAWHfEiSJElVaOhZPgYGBhaSnEn+Wzw7VpIkSXOjleTE7f/r6uoaKV3Z0Ak1STI97cUyJEmSpFnwMspclbjRE+rfAhx22GHsscceWcciSZKkeeiRRx7hpz/9KaS5Z6lGT6jHAPbYYw8WLlyYdSySJEma38oOMXaWD0mSJKkKJtSSJElSFUyoJUmSpCqYUEuSJElVMKGWJEmSqmBCLUmSJFXBhFqSJEmqggm1JEmSVAUTakmSJKkKJtSSJElSFUyoJUmSpCqYUEuSJElVMKGWJEmSqmBCPU/kcjnWrFlDLpfLOhRJUoMZGhpi1apVDA0NZR2K1JDaarWhEMJXgEOBcWA7cEqM8dYQwmFAH9AO5IDVMcaf1Squ+aKvr49t27axYcMGTjvttKzDkSQ1kHXr1rFjxw7WrVvHFVdckXU4UsOpZYW6N8b4lzHG5wEXAJenyz8FXBJjPAy4BLi0hjHNC7lcjv7+fiYmJujv77dKLUmq2NDQEMPDwwAMDw9bpZZ2Q80S6hjj/QXNfYHxEMKTgecDn0+Xfx54fgjhSbWKaz7o6+tjfHwcgLGxMTZs2JBxRJKkRrFu3bop25KmV7MhHwAhhE8DLwdagB7gYODXMcYxgBjjWAjhN+nyeyq938HBwTmItnFs2rSJ0dFRAEZHR+nv7+fII4/MOCpJUiPIV6cL2wMDA9kEIzWomibUMca3AYQQ3gz8f8AHZ+N+lyxZwsKFC2fjrhrSypUr2bhxI6Ojo7S1tdHT00NXV1fWYUmSGkBHR0dRUt3R0eFniFRiZGRkygJuJrN8xBg/CywD7gIODCG0AqS/DwDuzCKuRtXb28uCBclL2drayurVqzOOSJLUKNauXTtlW9L0apJQhxAWhxAOLmgfC9wL/B64FTghXXUC8IMYY8XDPQTt7e309PTQ0tJCT08P7e3tWYckSWoQnZ2ddHR0AEl1urOzM9uApAZUqwr1IuDqEMK2EMKtwGnAsTHGCeAdwCkhhJ8Cp6RtzVBvby9Lly61Oi1JmrG1a9eyaNEiq9PSbmqZmJjIOobdNjAw0AHc0exjqCVJkjR3CsZQH9rV1TVcut4rJUqSJElVMKGWJEmSqmBCLUmSJFXBhFqSJEmqggm1JEmSVAUTakmSJKkKJtSSJElSFUyoJUmSpCqYUEuSJElVMKGWJEmSqmBCLUmSJFXBhFqSJEmqggm1JEmSVAUTakmSJKkKJtSSJElSFUyoJUmSpCqYUEuSJElVMKGWJKnJDQ0NsWrVKoaGhrIORWpIJtSSJDW5devWsWPHDtatW5d1KFJDMqGWJKmJDQ0NMTw8DMDw8LBVamk3mFBLktTESqvSVqmlmTOhliSpieWr05O1JU3PhFqSpCbW0dExZVvS9EyoJUlqYmvXrp2yLWl6JtSSJDWxzs7OnVXpjo4OOjs7sw1IakAm1JIkNbm1a9eyaNEiq9PSbmrLOgBJkpStzs5ObrjhhqzDkBqWFWpJkiSpCibUkiRJUhVMqCVJkqQqmFBLkiRJVTChliRJkqpgQi1JkiRVwYRakiRJqoIJtSRJklQFE2pJkiSpCibUkiRJUhVMqCVJkqQqmFBLkiRJVTChliRJkqpgQi1JkiRVwYRakiRJqoIJtSRJklQFE2pJkppcLpdjzZo15HK5rEORGpIJtSRJTa6vr49t27axYcOGrEORGpIJtSRJTSyXy9Hf38/ExAT9/f1WqaXdYEItSVIT6+vrY3x8HICxsTGr1NJuMKGWJKmJbdmyhdHRUQBGR0fZvHlzxhFJjceEWpKkJrZixQra2toAaGtr4+ijj844IqnxmFBLktTEent7WbAgSQdaW1tZvXp1xhFJjceEWpKkJtbe3k5PTw8tLS309PTQ3t6edUhSw2nLOgBJkpSt3t5ehoeHrU5Lu8mEWpKkJtfe3s769euzDkNqWA75kCRJkqpgQi1JkiRVwYRakiRJqoIJtSRJTS6Xy7FmzRovOy7tppqclBhCaAc+CzwdGAGGgH+IMd4TQpgAtgHj6c3fHGPcVou4JElScvnxbdu2sWHDBk477bSsw5EaTq0q1BPAv8UYQ4zxOcDPgfML1r84xvjc9MdkWpKkGsnlcvT39zMxMUF/f79Vamk31CShjjHeG2O8uWDRd4Cn1WLbkiRpcn19fYyPJweJx8bG2LBhQ8YRSY2n5vNQhxAWACcBXy1YfHMIoQ34GnBWjHFkJvc5ODg4ixFKktQ8Nm3axOjoKACjo6P09/dz5JFHZhyV1FiyuLDLxcB24ONp+5AY450hhH1Ixll/EFg7kztcsmQJCxcunN0oJUlqAitXrmTjxo2Mjo7S1tZGT08PXV1dWYcl1ZWRkZEpC7g1neUjhHAB8Azgb2OM4wAxxjvT338CPg28pJYxSZLUzHp7e1mwIEkHWltbvfy4tBtqllCHEM4FuoDX5Id0hBD2CyHslf7dBrwOuLVWMUmS1Oza29vp6emhpaWFnp4e2tvbsw5Jaji1mjbv2cDpwE+Bb4cQAO4A/g24NJ0673HAt0mGfEiSpBrp7e1leHjY6rS0m2qSUMcYfwy0TLL6ObWIQZIkldfe3s769euzDkNqWF4pUZIkSaqCCbUkSZJUBRNqSZIkqQom1JIkSdrF0NAQq1atYmhoKOtQ6p4JtSRJknaxbt06duzYwbp167IOpe6ZUEuSJKnI0NAQw8PDAAwPD1ulnoYJtSRJkoqUVqWtUk/NhFqSJElF8tXpydoqZkItSZKkIh0dHVO2VcyEWpIkSUXWrl07ZVvFTKglSZJUpLOzc2dVuqOjg87OzmwDqnMm1JIkSdrF2rVrWbRokdXpCrRlHYAkSZLqT2dnJzfccEPWYTQEK9SSJElSFUyoJUmSpCqYUEuSJElVMKGWJEnSLnK5HGvWrCGXy2UdSt0zoZYkSdIu+vr62LZtGxs2bMg6lLpnQi1JkqQiuVyO/v5+JiYm6O/vt0o9DRNqSZIkFenr62N8fByAsbExq9TTMKGWJElSkS1btjA6OgrA6Ogomzdvzjii+mZCLUmSpCIrVqygrS25/l9bWxtHH310xhHVNxNqSZIkFent7WXBgiRNbG1tZfXq1RlHVN9MqCVJklSkvb2dnp4eWlpa6Onpob29PeuQ6lpb1gFIkiSp/vT29jI8PGx1ugIm1JIkSdpFe3s769evzzqMhuCQD0mSJKkKJtSSJElSFUyoJUmSpCqYUEuSJElVMKGWpCYzNDTEqlWrGBoayjoU1YmtW7fS3d3NTTfdlHUoqiO5XI41a9aQy+WyDqXumVBLUpNZt24dO3bsYN26dVmHojpx3nnnAXDuuedmHInqSV9fH9u2bWPDhg1Zh1L3TKglqYkMDQ0xPDwMwPDwsFVqsXXrVkZHRwEYHR21Si0gqU739/czMTFBf3+/VeppmFBLUhMprUpbpVa+Op1nlVqQVKfHx8cBGBsbs0o9DRNqSWoi+er0ZG01n3x1erK2mtOWLVuKjlxs3rw544jqmwm1JDWRjo6OKdtqPm1tbVO21ZxWrFixsy+0tbVx9NFHZxxRfTOhlqQmsnbt2inbaj6nn356UfuMM87IKBLVk97eXhYsSNLE1tZWVq9enXFE9c2EWpKaSGdn586qdEdHB52dndkGpMwtX768qBK5bNmyjCNSPWhvb6enp4eWlhZ6enpob2/POqS6ZkItSU1m7dq1LFq0yOq0dspXqa1Oq1Bvby9Lly61Ol2BlomJiaxj2G0DAwMdwB1Llixh4cKFWYcjSZKkeWhkZITBwUGAQ7u6uoZL11uhliRJkqpgQi1JkiRVwYRakiRJqoIJtSQ1mVwux5o1a7yUsHbaunUr3d3dXnZc2k0m1JLUZPr6+ti2bZuXEtZO+cuPe9lxafeYUEtSE8nlcvT39zMxMUF/f79VarF169aiS0xbpZZmzoRakppIX18f4+PjAIyNjVml1s7qdJ5VamnmTKglqYls2bKlqBq5efPmjCNS1vL9YbK2pOmZUEtSE1mxYkXRZaaPPvrojCNS1vL9YbK2pOmZUEtSE+nt7WXBgmTX39ra6iWFtfOy43leflyaORNqSWoi7e3t9PT00NLSQk9PD+3t7VmHpIwtX7686KjFsmXLMo5Iajwm1JLUZHp7e1m6dKnVae2Ur1JbnZZ2T8vExETWMey2gYGBDuCOJUuWsHDhwqzDkSRJ0jw0MjLC4OAgwKFdXV3DpeutUEuSJElVMKGWJEmSqmBCLUmSJFXBhFqSmswtt9zC8uXLGRgYyDoU1YmtW7fS3d3tZcdVJJfLsWbNGnK5XNah1L2anJQYQmgHPgs8HRgBhoB/iDHeE0I4ArgU2AsYBk6MMf6+kvv1pERJmrlXvvKVbN++ncWLF3P99ddnHY7qwIoVKxgdHaWtrY0tW7ZkHY7qxEc/+lGuu+46XvWqV3HaaadlHU6m6uWkxAng32KMIcb4HODnwPkhhBbgc8DJMcbDgG8C59coJklqOrfccgvbt28HYPv27VapxdatW4suR2+VWpBUp/v7+5mYmKC/v98q9TRqklDHGO+NMd5csOg7wNOAw4GHY4zfSpd/Cnh9LWKSpGZ01llnFbXPPPPMbAJR3TjvvPOK2ueee25Gkaie9PX1MT4+DsDY2BgbNmzIOKL61lbrDYYQFgAnAV8FDgF+mV8XY/xDCGFBCGH/GOO9ld5nWoKXJE0jX50ubFulbm756nRh2z6hTZs2FR256O/v58gjj8w4qvpV84QauBjYDnwc+JvZuEPHUEtSZRYvXlyUVC9evJiurq4MI1LW2traipLqtrY2+4RYuXIlGzdu3Dm2vqenp6n7RcEY6rJqOstHCOEC4BnA38YYx4FfkQz9yK9/IjAxk+q0JKlypUM+zj777GwCUd3IX3Y8z8uPC6C3t5cFC5I0sbW1ldWrV2ccUX2rWUIdQjgX6AJeE2McSRcPAHuFEF6att8BfLFWMUlSszn88MNZvHgxYHVaieXLl9PWlhywbmtrY9myZRlHpHrQ3t5OT08PLS0t9PT00N7ennVIda0mCXUI4dnA6cABwLdDCLeGEL6cVqnfDHwyhPAz4K+B99ciJsf19CEAACAASURBVElqVmeddRYLFiywOq2d8lVqq9Mq1Nvby9KlS61OV6Am81DPFeehliRJ0lyrl3moJUmSpHnJhFqSJEmqggm1JEmSVAUTakmSmtzpp59Od3c3H/rQh7IORXVkaGiIVatWMTQ0lHUodc+EWpKkJvftb38bgG9+85sZR6J6sm7dOnbs2MG6deuyDqXumVBLktTESi/sYpVakFSnh4eHARgeHrZKPQ0TakmSmli+Op1nlVrALlVpq9RTM6GWJElSkXx1erK2irVVcqMQwgnArTHG20IIAbgMGAXeGWO8fS4DlCRJUm11dHQUJdEdHR2ZxdIIKq1QrwPuTf++APge8E3gE3MRlCRJqo0Xv/jFRe0jjzwyo0hUT9auXTtlW8UqTaifFGO8O4SwJ/BS4AzgHOC5cxaZJEmac+edd15R+5xzzskoEtWTzs7OnVXpjo4OOjs7sw2ozlWaUN8TQugEXgH8X4xxBNgTaJmzyCRJUk3kq9RWp1Vo7dq1LFq0yOp0BSoaQw38CzAAjAF/my47CvjhXAQlSZJqp7RKLUFSpb7hhhuyDqMhTFqhDiG8qqB5JfBU4KAY4+Z02XeBN8xhbJIkSVLdm6pC/Tlgn/TvXIxxn8KVMcbfz1lUkiRJUoOYKqH+XQjhXcBPgLYQwjLKjJmOMW6dq+AkSZKkejfVSYlvAV4DXArsAVwOfKbk59NzHJ+kKuRyOdasWUMul8s6FEl17IQTTqC7u5sTTzwx61BUR2655RaWL1/OwMBA1qHUvUkr1DHGbwMrAEIIQzFG50uRGkxfXx/btm1jw4YNnHbaaVmHI6lO/fa3vwXgrrvuyjgS1ZOzzjqL8fFxzjzzTK6//vqsw6lrFU2bZzItNZ5cLkd/fz8TExP09/dbpZZU1gknnFDUtkotSKrT27dvB2D79u1WqacxaYU6hNAfY+xJ//5vYKLc7WKMTlop1aG+vj7Gx8cBGBsbs0otqax8dTrPKrUgqU4Xsko9talOStxQ8LdjpaUGs2XLFkZHRwEYHR1l8+bNJtSSpIrkq9OTtVVsqjHUVxU0b48xfrf0NiGEF85JVJKqtmLFCjZu3Mjo6ChtbW0cffTRWYckSWoQixcvLkqiFy9enGE09a/SS49vnmR5/2wFIml29fb2smBB8hZvbW1l9erVGUckqR499alPLWofdNBBGUWielI65OPss8/OJpAGMeWlx0MIC0jmnm4JIbRQPA/104HROYxNUhXa29vp6enhuuuuo6enh/b29qxDklSHPv/5z9Pd3b2z/bnPfS67YFQ3Dj/88J1V6sWLF9PV1ZV1SHVtugr1KPAI8Pj070cLfn4CfGJOo5NUld7eXpYuXWp1WtKU8lVqq9MqdNZZZ7FgwQKr0xVomZgoO3kHACGEp5FUpb8BFM7mMQHcE2N8aG7Dm9rAwEAHcMeSJUtYuHBhlqFIkiRpnhoZGWFwcBDg0K6uruHS9VMO+Ygx/jL982mzH5okSZLU+Kaah/rfY4z/L/17w2S3izF6LFmSJElNa6oK9R0Ff/98rgORJEmSGtFU81D/a8HfjkaXJEmSyph0lo8QwvJKfmoZrCaXy+VYs2YNuVwu61BUR+wXKmf9+vV0d3dzySWXZB2K6sSxxx5Ld3c3r371q7MORXVk69atdHd3c9NNN2UdSt2batq8z5T89ANfAz6X/u7HS5LXjb6+PrZt28aGDZMOd1cTsl+onGuuuQaAq6++OuNIVC8eeOABAO6///6MI1E9Oe+88wA499xzM46k/k2aUMcYD83/AJcBFwP7xRgPAPYD1qfLlbFcLkd/fz8TExP09/dbjRRgv1B569evL2pbpdaxxx5b1LZKLUiq06OjyfX7RkdHrVJPo9JLj58GvD/G+CBA+vsDwLvnKjBVrq+vj/HxcQDGxsasRgqwX6i8fHU6zyq18tXpPKvUgseq03lWqadWaUK9A3hhybIXAA/ObjjaHVu2bCn6Frl58+aMI1I9sF9IknZX/vNjsraKVZpQfxDoDyFcFUL4cAjhKpIx1GvnLjRVasWKFbS1JRO2tLW1cfTRR2cckeqB/UKStLvynx+TtVWsooQ6xvhZ4EXAbcA+wO3AEelyZay3t5cFC5KXsrW1ldWrvdaO7Bcq77jjjitqH3/88RlFonqx9957F7X33XffjCJRPTn99NOL2meccUZGkTSGSivUxBh/EmP8lxjjSTHGc2KMP5nLwFS59vZ2enp6aGlpoaenh/b29qxDUh2wX6icNWvWFLVPPvnkjCJRvbjuuuuK2tdee21GkaieLF++vOgo57JlyzKOqL5NdenxzwIT092Blx6vD729vQwPD1uFVBH7hco57rjjuOaaa6xOa6e9996bBx54wOq0ipx++umcc845Vqcr0DIxUT5nDiGcWdB8ItALXAf8EjgEOBboizGuKfPvNTEwMNAB3LFkyRIWLlyYVRiSJEmax0ZGRhgcHAQ4tKura7h0/VSXHt95ufEQwiZgVYzxvwuWvZTkZEVJkiSpaVU6hvoI4Dsly74L/NXshiNJkiQ1lkoT6h8A54UQ9gJIf58L3DpXgUmSJEmNoNKE+i3AS4D7Qwh3A/cDLwU800mqY0NDQ6xatYqhoaGsQ1Ed+fCHP0x3dzcf+chHsg5FdWLZsmV0d3ezfPnyrENRHVm/fj3d3d1ccsklWYdS9yqdh3o4xvhioBN4FdAZY3xxjHF4LoOTVJ1169axY8cO1q1bl3UoqiNf+9rXgF2nS1Pzyk9QMD4+nnEkqifXXHMNAFdffXXGkdS/iuehBogx/gr4HnBXCGFBCGFG/y+pdoaGhhgeHgZgeHjYKrWApDpdyCq1SucXtkotSKrThaxST62ihDiEcEAI4cshhBwwCjxa8COpDpVWpa1SCx6rTudZpVbp9LlWqQWPVafzrFJPrdIK86XAI8BRwHbg+cBXgXfMUVySqpSvTk/WliRJs6PShPrFwFtjjLcCEzHGHwJ/D7xnziKTVJWOjo4p25IkaXZUmlCPkQz1APhjCOFJwA7gwDmJSlLV1q5dO2VbzekVr3hFUfvYY4/NKBLVi5aWlqL2ggWeHiU47rjjitrHH398RpE0hkrfNd8Fjkn/3gR8AbgGuGUugpJUvc7Ozp1V6Y6ODjo7O7MNSHXhfe97X1H7Pe/xQGOzu+mmm4raW7duzSgS1ZM1a9YUtU8++eSMImkMlSbUbwa+kf79j8BWYBB441wEJWl2rF27lkWLFlmdVpF8ldrqtPLyVWqr0yqUr1JbnZ5eS+nZvY1kYGCgA7hjyZIlLFy4MOtwJEmSNA+NjIwwODgIcGhXV9dw6fq2qf45hPCh/N8xxnNmPTpJkiSpwU2ZUAMt06yXJEmSmtqUCXWM8exaBSJJkiQ1okkT6hBCRdcejTF6OnAd2Lp1K+eccw5nnnnmLpeRVfO67LLLuPLKK1m9ejVvfetbsw5HdeLEE0/krrvuoqOjgyuuuCLrcFQHuru7d/598803ZxaH6oufIZWbqkL9mZL2gcAEkAPaSYaD3AX8eSUbCiFcALwW6ACWxhgH0+XDwMPpD8D7YoybKopeO5133nkAnHvuuSbU2unKK68EYMOGDe4MtdNdd90FePVMSVPzM6Ryk86PE2M8NP8DXAZcDOwXYzwA2A9Yny6v1FeAI4Fflln3uhjjc9Mfk+kZ2rp1K6OjyXV3RkdHd5lTVM3pssuK356XX355RpGonpx44olF7be85S3ZBKK6UVidLtdWc/IzZGYqnXDyNOD9McYHAdLfHwDeXemGYozfijHeOfMQNZ18dTrv3HPPzSgS1ZN8ZSFvw4YNGUWiepKvTudZpZZUjp8hMzPdLB95O4AXAv9TsOwFwIOzFMeVIYQW4FvA6THGP87kn9N5AZtWvjpd2B4YGMgoGtUz+4XKsV+olH1C5dgvJldpQv1BoD+EcB1wJ3Aw8EpgNq5D+bIY450hhIXAhcDHgROn+Z8izX5hl7a2tqKkuq2tja6urgwjUr2yX6gc+4VK2SdUTjP3i4ILu5RV0ZCPGONngRcBtwH7ALcDR6TLq5IfBhJjHAE+Abyk2vtsNqeffnpR+4wzzsgoEtWTN73pTUXt1atXZxSJ6slBBx1U1O7o6MgmEEl1zc+Qman5pcfTWT1eGWMcDCEsAtpijPenQz7WAX8RY/ybSu7LS48/ZsWKFYyOjtLW1saWLVuyDkd1wqmwVI79QqXsEyrHfvGY6S49XulJiVULIawPIdwFHARsCSH8GHgKcHMI4UfAIHAY8M5axTSf5KvUVqdVKF9hsLKgQvkqtdVpSVPxM6RyNa9QzyYr1JIkSZprdVOhliRJkuajGSXUIYSDQwhHzFUwkiRJUqOpaNq8EMIhwOeB55JcfnxxCOF1QE+M8W1zGJ8kSZJU1yqtUF8K3ADsDTyaLtsMHD0XQWnmcrkca9asIZfLZR2K6sjb3vY2uru7ecc73pF1KKojr33ta+nu7ub1r3991qGoTnR3d+/8kfJWrlxJd3c3PT09WYdS9ypNqF8InB9jHCepUBNjvB/Yd64C08z09fWxbds2Lw2qIkNDQwDcfvvtGUeiepL/4v373/8+40gk1bORkREAHn744YwjqX+VJtR3A52FC0IIfwH8atYj0ozlcjn6+/uZmJigv7/fKrWApDpdyCq1IKlOF7JKrdKqtFVqQVKdLmSVemqVJtQXANeHEP4OaAshnAB8AfjwnEWmivX19TE+Pg7A2NiYVWoBj1Wn86xSC9jlC7dVaknl5KvTeVapp1bppccvB/4ZOB64E1gNfDDGeOUcxqYKbdmyhdHRUQBGR0fZvHlzxhFJkiQ1j2ln+QghtAJnAufGGL8y9yFpplasWMHGjRt3Xnr86KM9V1SSJKlWpq1QxxjHgJN5bHYP1Zne3l4WLEheytbWVi8RKgA6O4tOe+CZz3xmRpGonrS3txe1n/zkJ2cUiaR6VnoF6j333DOjSBpDpWOo+wDPaKpT7e3t9PT00NLSQk9Pzy4fmGpOn/70p4van/rUpzKKRPXkS1/6UlH7i1/8YkaRqF7cfPPNU7bVnDZt2lTU7u/vzyiSxlDRhV1Ips07JYTwzyRjqCfyK2KMR85FYJqZ3t5ehoeHrU6rSGdnJ0NDQ1anVaS9vZ1cLmd1WtKUFi5cyMjIiNXpCrRMTExMe6MQQu9k62KMfbMa0QwMDAx0AHcsWbJkl0MTkiRJ0mwYGRlhcHAQ4NCurq7h0vUVVaizTJolSZKkelbpkA9CCG8FTgAOAH4D/CdweYxx+hK3JEmSNE9VlFCHEP4NeDVwIfBL4BDgvUAgmZ9akiRJakqVVqjfAjw/xnhXfkEI4Qbg+5hQ14XLLruMK6+8ktWrV/PWt74163BUJ0466SRuu+02li5dysUXX5x1OKoThZeWdkYHgX1C5S1fvpzx8XFaW1u58cYbsw6nrlU6bd4D6U/psj/NbjjaXVdemVy00suOq9Btt90GwLZt2zKORJLUaMbHxwEYGxvLOJL6N2lCHUL48/wPyVCPa0IIR4cQnhVCeDlwNfCxWgWqyV122WVF7csvvzyjSFRPTjrppKL2KaecklEkqieFlchybTUf+4TKWb58eVH7qKOOyiiSxjDVkI8hkvmmWwqWLSu5zXLg47MdlGYmX53O27Bhg8M+tLM6nWeVWpJUqXx1Os8q9dQmTahjjJUOB5EkSZKalkmzJEmSVIVKp807BDgTeB6wuHBdjPGwOYhLM/CmN72paNiHlx8XwLOe9ayiYR9Lly7NMBpJUiNZsGBB0bCP1tbWDKOpf5VWqK8mSb4/BJxU8qOMvf3tby9qO35aAJ/85CeL2k6bJ9h1SjSnSJN9QuVs3bq1qO20eVOrdB7qZwJ/FWMcn/aWykS+Sm11WoXyVWqr05KkmcpXqa1OT69lYmL6K4eHED4HfCbGeNPch1S5gYGBDuCOJUuWsHDhwqzDkSRJ0jw0MjLC4OAgwKFdXV3DpesrrVCvAb4dQvg5cHfhihij4wskSZLUtCpNqP8DGANuAx6au3AkSZKkxlJpQr0cOCDGWHr5cUmSJKmpVZpQ/whoB0yo69T69eu55pprOP744zn55JOzDkd1YuXKlYyMjLDnnnvS39+fdTiqE4WXlnZGB4F9QuXZLypXaUK9Ffh6COE/2HUM9eWzHpVm7JprrgHg6quvNqHWTiMjIwA8/PDDGUciSdL8Vek81C8Ffg28HHhzwc+JcxSXZmD9+vVF7UsuuSSjSFRPVq5cWdTu6enJKBLVk8KKU7m2mo99QuXYL2amogp1jHHZXAei3ZevTudZpRY8Vp3Os0otSdLcqHTIByGE/YBjgQNJqtXXxRjvm6vAJEmSpEZQ0ZCPEMJfAT8H3gE8B/gH4OfpckmSJKlpVTqG+kLgnTHGF8cYT4gxvgQ4CVg/zf+pBo477rii9vHHH59RJKonpVcP3XPPPTOKRJKk+a3ShPow4Isly/4L6JzdcLQ71qxZU9R2/LQANm3aVNR22jzBrlNfORWW7BMqx34xM5Um1D8D3lCy7HiSYSCqA/kqtdVpFcpXqa1OS5I0d1omJiamvVEI4cXA9cBPgV8CHcAzgFfGGL89lwFOZWBgoAO4Y8mSJbsc3pYkSZJmw8jICIODgwCHdnV1DZeur6hCnSbNTwc+DgwAFwOdWSbTkiRJUj2oeNq8dIq8z81hLJIkSVLDmTKhDiHcBEw1JmQixnjU7IYkSZIkNY7pKtSTVaQPBNYAj5/dcCRJkqTGMmVCHWP8TGE7hNAOfAB4O/AF4Jy5C00z8d73vpdbbrmFI444gvPPPz/rcFQnVq1axY4dO9h777257rrrsg5HdaK7u3vn306FJbBPqDz7ReUqGkMdQtgH+CfgXSSzfTw/xuiUeXXklltuAeA73/lOxpGonuzYsQOABx54IONIJEmav6ac5SOEsFcI4QPAL4BnAS+NMb7ZZLq+vPe97y1qv//9788oEtWTVatWFbWPPfbYjCJRPSmsOJVrq/nYJ1SO/WJmpqtQ3wG0Av8G3AI8JYTwlMIbxBi3zlFsqlC+Op1nlVrwWHU6zyq1JElzY7qE+mGSWT5OmmT9BPDnsxqRJEmS1ECmOymxo0ZxSJIkSQ2poislqr4dfvjhRe0jjjgio0hUTxYtWlTU3nvvvTOKRJKk+c2Eeh644IILitpOmyeAG264oajttHmCXae+cios2SdUjv1iZkyo54l8ldrqtArlq9RWpyVJmjstExNTXVm8vg0MDHQAdyxZsoSFCxdmHY4kSZLmoZGREQYHBwEO7erqGi5db4VakiRJqoIJtSRJklQFE2pJkiSpCtNd2EUN4uUvfzmPPPIICxcuZNOmTVmHozpReKlYz9BWnv1CpewTKsd+UbmaJNQhhAuA1wIdwNIY42C6/DCgD2gHcsDqGOPPahHTfPPII48AyaB5SZIk1U6thnx8BTgS+GXJ8k8Bl8QYDwMuAS6tUTzzystf/vKi9sqVKzOKRPWksLJQrq3mZL9QKfuEyrFfzExNEuoY47dijHcWLgshPBl4PvD5dNHngeeHEJ5Ui5jmk3x1Os8qtSRJUu1kOYb6YODXMcYxgBjjWAjhN+nye2ZyR+m8gCowMDCQdQiqQ/YLlWO/UCn7hMqxX0xuXpyU6IVddtXV1ZV1CKpD9guVY79QKfuEymnmflFwYZeyspw2707gwBBCK0D6+4B0uWZgjz32KGr75UKSJKl2MkuoY4y/B24FTkgXnQD8IMY4o+Eegq9//etFbafNE+w6xZFTHgnsF9qVfULl2C9mpiYJdQhhfQjhLuAgYEsI4cfpqncAp4QQfgqckra1G/JVaqvTkiRJtdUyMTGRdQy7bWBgoAO4wzHUkiRJmisFY6gP7erqGi5d76XHJUmSpCqYUEuSJElVMKGWJEmSqjAv5qEWLFu2jImJCRYsWMDWrVuzDkd1ovBSsZ6hrTz7hUrZJ1SO/aJyVqjnifzJpePj4xlHIkmS1FxMqOeBZcuWFbWXL1+eUSSqJ4WVhXJtNSf7hUrZJ1SO/WJmTKjngdKpD61SS5Ik1Y4JtSRJklQFE2pJkiSpCibU80BLS0tRe8ECX1ZJkqRaMfOaB2666aaittPmCXad4sgpjwT2C+3KPqFy7BczY0I9T+Sr1FanJUmSaquldIaIRjIwMNAB3LFkyRIWLlyYdTiSJEmah0ZGRhgcHAQ4tKura7h0veVMSZIkqQom1JIkSVIVTKglSZKkKrRlHYBmR+ElQT0TV3n2C5Vjv5BUCfcVlbNCLUmSJFXBhHoeKPwGWa6t5mS/UDn2C0mVcF8xMw75kCSpTm3atImNGzfO+Xbuu+8+APbbb7853xbAMcccw8qVK2uyLakWTKglSWpyuVwOqF1CLc03JtSSJNWplStX1qSSe+qppwJw0UUXzfm2pPnIMdSSJElSFUyo54HSqWyc2kZgv1B59gtJlXBfMTMm1JIkSVIVHEM9T/jNUeXYL1SO/UJSJdxXVM4KtSRJklQFE2pJkiSpCibUkiRJUhUcQz1PFF4S1DFPkiSpWuYWlbNCLUmSJFXBhHoeKPwGWa4tSZI0E+YWM+OQD6nGNm3axMaNG2uyrfvuuw+A/fbbb863dcwxx9TkEsnzlf1CkhqXCbU0j+VyOaA2iZMah/1CkmaXCbVUYytXrqxZxe7UU08F4KKLLqrJ9rT77BeS1LgcQy1JkiRVwYR6HiidysapbSRJUjXMLWbGhFqSJEmqgmOo5wm/OUqSpNlkblE5K9SSJElSFUyoJUmSpCqYUEuSJElVcAz1PFF4SVDHPEmSJNWOFWpJkiSpCibU80BhdbpcW5IkSXPHIR+SJEkNZNOmTWzcuHHOt3PfffcBsN9++835tgCOOeYYVq5cWZNtzTYTakmSJO0il8sBtUuoG5kJtSRJUgNZuXJlTSq5p556KgAXXXTRnG+r0TmGWpIkSaqCCfU8UDpNntPmSZIk1Y4JtSRJklQFx1DPE1alJUmSsmGFWpIkSaqCCbUkSZJUBRNqSZIkqQom1JIkSVIV6uKkxBDCMPBw+gPwvhjjpswCakDd3d07//YERUmSpNqpi4Q69boY42DWQUiSJEkz4ZCPeaCwOl2uLUmSpLlTTxXqK0MILcC3gNNjjH/MOiBJkiRpOvWSUL8sxnhnCGEhcCHwceDESv95cNCRIqUGBgayDkF1YPv27YD9QcXsFypln1A59ovK1UVCHWO8M/09EkL4BPDVmfz/kiVLWLhw4ZzE1qi6urqyDkF1YPHixYD9QcXsFypln1A59ovHjIyMTFnAzXwMdQhhUQhh3/TvFuANwK3ZRiVJkiRVJvOEGngKcHMI4UfAIHAY8M5sQ2ospdPkOW2eJElS7WQ+5CPG+AvgeVnHIUmSJO2OzBNqzQ6r0pIkSdmohyEfkiRJUsMyoZYkSZKqYEItSZIkVcEx1PNE4eXGHU8tSZJUO1aoJUmSpCqYUM8DhdXpcm1JkiTNHYd8SJI0AxdffDFDQ0NZhzGr8o/n1FNPzTiS2dfZ2ckpp5ySdRia50yoJUmagaGhIW4dvI2xx++fdSizpmUsSQcGfnF3xpHMrtYH7806BDUJE2pJkmZo7PH789Azj8k6DE1jr9s3Zh2CmoRjqCVJkqQqmFDPA6XT5DltniRJUu2YUEuSJElVcAz1PGFVWpIkKRtWqCVJkqQqmFBLkiRJVTChliRJkqrgGOp5ovBy446nliRJqh0r1JIkSVIVTKjngcLqdLm2JEmS5o5DPiTg4osvZmhoKOswZl3+MZ166qkZRzK7Ojs7OeWUU+Z8O/aLxlKrfiGVMx/3F/N1XwGzv78woZZIdhq3Dt7G2OP3zzqUWdUylrzFB35xd8aRzJ7WB++t2bbsF42jlv1CKmc+7i/m474C5mZ/YUItpcYevz8PPfOYrMPQNPa6fWNNt2e/aAy17hdSOe4vGsNc7C9MqCVJamJLf7WNo358I/s+dD/377UvNz77KLYdsjTrsKSG4kmJ80DpNHlOmydJqsTSX23j2B9cxxMeup8W4AkP3c+xP7iOpb/alnVoUkMxoZYkqUkd9eMb2WPs0aJle4w9ylE/vjGjiKTG5JCPecKqtCRppvZ96P4ZLVfzcCjQzFihliSpSd2/174zWq7m4FCgmTOhliSpSd347KN4pPVxRcseaX0cNz77qIwiUj1wKNDMOeRDkpqIh3FVKP/a2ydUyKFAM2dCLUlNIn8YN195yh/GBUygmti2Q5b6+qvI/XvtyxPKJM8OBZqcQz4kqUl4GFdSJRwKNHNWqCWpSXgYV1IlHAo0cybUc2jTpk1s3Fiby+Hed999AOy3335zvq1jjjmGlStXzvl2JM0uD+POjnvvvZfWB3Ne7rwBtD6Y4957Hzf9DWfBfOsXQ8DQoYc9tuDBO9nr9jszi2c2zUW/MKGeJ3K5HFCbhFqNwZPPVOrGZx9VNIYaPIwrSbPBhHoOrVy5smaV3FNPPRWAiy66qCbbU33z5DOV42Hc2bH//vtzxx8f5aFnHpN1KJrGXrdvZP/996/JtuwXjWMu+oUJtTQPTXXymclTc3NGB0mafc7yIc1DnnwmSVLtmFBL85CXE5YkqXZMqKV5yDlEJUmqHcdQS/OQJ59JklQ7JtTSPOXJZ5Ik1YZDPiRJkqQqmFBLkiRJVTChngfuuvZaXnLjjRx1/fVsednLuOvaa7MOSZIkqWk4hrrB3XXttfzojDPY66GHAHjoN7/hR2ecAcBBr351lqFJkiQ1haZLqC+++GKGhob+//buP8iusr7j+HuzLGETsqEJENGGIN34pUhHJVV+lB/jaGXMWLD0F4KmhVGUmTJpQcFBoVRrKoq2kYEBqRZqKR1txak2KQzS2kqwponBBsk3LEmQooOQG/LDTfNjs/3jnG3X7e5mydm95+7m/ZrZyd7nnHue7737zM3nPvc559Zdxrj5lW9+vgPkjQAAC15JREFU83/D9IC+3btZ9ZGP8Ogjj9RU1fjp7u7m6quvnvB+Go0G7b1b6dywYsL7UjXtvVtpNDoOvuM4cFxMHs0cF0V/jSk1Ltr2Ff+P9Hd01lzJ+GrvbQDzmtqf46L1TcS4OOwCdU9PD+vWP0nfjPH9Dve6vGVImB5w1O7drNn0fJOrGV/FgJek1tLd3V13CeNuYKKp++Tmhc/mmNe0v5fjYjIZ/3Fx2AVqgL4Zc9h9yuK6yxgX2zdv5Jhhvk56e+fsSf8Ym/kuf86cOWx+ad+kf84OB50bVjBnTnPeEDsuJo9mjotmfGrWbEuXLgVg+fLlNVcyeTkuDm+elDjJ+Y14kiRJ9TosZ6inEr8RT5IkqV4G6inAb8STJEmqj0s+JEmSpAoM1JIkSVIFBmpJkiSpAgO1JEmSVIGBWpIkSaqgJa7yERGvAe4F5gJbgSWZ+VS9VUmSJEkH1yoz1HcCt2fma4DbgbtqrkeSJEkak9oDdUQcD5wO3F823Q+cHhHH1VeVJEmSNDatsORjPvBcZvYBZGZfRPyobH9hvDtrNBq073yeo9d+abwP/f8dOAD9Bya+n2ZrmwbTmvBerG8/jUbHwfcbJ+29DTo3rJjwftr27Wbavt4J76fZDnTMoL+jc8L7ae9tAPMmvJ/B/TkuDt1UHRfN8uCDD7JixcSPv56eHgCWLl064X0BLF68mAsuuKApfU1FjovW0wqBurL169ePed/Ozk46j5o+gdX8n74+2L9/6gXqI46YRnt7+8R31NFOZ2cna9asmfCuurq66D5p/oT3A7Bz5wF27NjTlL6aqatrOrNmzWxCTzPp6upyXEwSU3FcNNOWLVvYtWvXhPczc2bxN2pGX1A8rqn2t2omx0Xraevv76+1gHLJx0Zgbjk73U5xYuLCzBx1hnrNmjUnAZtPO+00pk9vTkiWJEnS4WXPnj0DE7ivXrRo0Zah22tfQ52ZPwHWAe8qm94FfO9gYVqSJElqBa2y5OMDwL0RcROwDVhScz2SJEnSmLREoM7MDcAZddchSZIkvVy1L/mQJEmSJjMDtSRJklSBgVqSJEmqwEAtSZIkVWCgliRJkiowUEuSJEkVGKglSZKkCgzUkiRJUgUGakmSJKkCA7UkSZJUgYFakiRJqsBALUmSJFVgoJYkSZIqOKLuAipqB9i7d2/ddUiSJGmKGpQ124fbPtkD9QkAGzdurLsOSZIkTX0nAE8PbZzsgXo1cC7wY6Cv5lokSZI0NbVThOnVw21s6+/vb245kiRJ0hTiSYmSJElSBQZqSZIkqQIDtSRJklSBgVqSJEmqwEAtSZIkVWCgliRJkiowUEuSJEkVGKglSZKkCgzU0hQUEf0RcXTddUianHwNkV4eA7UkSZJUwRF1F6DqIuI+IIDpQA9wRWZuq7cqtYAPRsTbgLnADZn593UXpPpFxFnAp4FZZdOHMvOhGktSC4iIi4FlQANYUXM5ahERcQbwSaCrbLopM/+xxpJaljPUU8PSzPzlzPwl4Ang+roLUks4kJlnAxcCn4+I4+suSPWKiDnAA8B1mfk64HRgdb1VqW7la8PdwEXla8bemktSC4iIY4A7gUszcxHwDuCusl1DOEM9NSyJiMuAI4GZwMaa61Fr+AJAZmZErAXOBP6h3pJUs7OAH2TmKoDM7AP8NEtnAmszM8vbnwduqbEetYazgVcDKyNioK0f6Ab+o66iWpWBepKLiHOBq4CzM/OFiLgUuLLmstR62iheCHV4a6u7ALUkx4WG0wZ8PzPPq7uQycAlH5PfMcB2YGtETAeuqLketY7LASJiIfB64N/rLUctYBVwarmOmohoj4ifq7km1e8x4A3lawXAe+ssRi1jFbAwIt480BARb4wI34ANw0A9+a0EngY2lL+vrbcctZA9EfEo8A3g/Zn5k7oLUr0yswFcDHw2Ir4PrAEW1VuV6la+NlwJfD0iVgH7ay5JLaC8uMGFwB9FxOMR8SRwM36iMay2/n4/BZYkSZIOlTPUkiRJUgUGakmSJKkCA7UkSZJUgYFakiRJqsBALUmSJFVgoJakQxQRN0fEX9ddRzNFxGUR8VDddUhSK/GbEiVpFOW3j14DnALsBNYBn8jMb9da2BARcQ/wu8AZmfndsq0beCozD3rd2Ig4CdgMdGTmiNchzsz7gPtGOc7PA8uB84EO4IfAZzLznrE+FkmabJyhlqQRRMQ1wJ8Dy4B5wInAHcBFddY1igbwJxN18IgYyyTMl4BngQXAXGAJ8HwNdUhS0/iiJEnDiIjZwMeAyzPzq4M2fb38Ge4+XwHOBTqBx4GrMvOJctti4FZgPrAD+LPMvDUijgXuAc4BDgBPAOdn5oGIeCVwG3AesKu8z+dGKfte4NKIOD8zvzVMfVuA92bmw+Xtm4HuzHw38K/lbi9FBMCvAgG8D/guxez3HRHRUx7jnBFqeCPwh5n50/L294bUcA7wKeBUihn/GzPznvL5vg14O9AL3A0sK5+H3xtaB/DRiLgC+BDwinLblZn5zCjPjyRNCGeoJWl4ZwFHAQ+8jPusBBYCxwNr+dmlEV+g+Ar4WcBpwCNl+7XAfwHHUcyC3wD0R8Q0iuD+OPAq4C3AH0TEBaP030sxm/6Jl1HzgPPKf4/JzKMz87Hy9hnApvIxjeW43wFuj4hLIuLEwRvK2yspgvNxwOspltBQts0GTqZYLrIEuHzQ3X+mjoh4J8VzdXF5rH8D7h/zo5WkceQMtSQNby7w4mjriYfKzC8O/F7O/m6LiNmZuR3YB5waEY9n5jZgW7nrPuAEYEFm9lAEQyLiTcBxmfmxcr9NEXE3cAnw4Chl3AV8MCLeDjw11tpH8aPMvK38fX85ez2a3wKuB24ETomI/wTel5mrgcuAhzNzIPhuBbZGRDvwO8AbMnMnsDMiPgO8h+KNyHB1vB/408x8EiAilgE3RMQCZ6klNZsz1JI0vK3AsWNdrxsR7RHxyYh4OiJ2AFvKTceW//4GsBh4JiK+FRFnle2fBnqAhyJiU0R8uGxfALwyIl4a+KGYkZ03Wh2ZuQf4ePlz0JMRx+DZkTZExA0Rsav8ubPsf1tmfjgzX1vWug74WkS0USx3eXqYQx0LHAkMDsLPUMzMj1THAmD5oOemQfF4X4UkNZkz1JI0vMeA/wbeCfzdGPa/lOJkxbdShOnZFLPQbQDlDO1FEdEB/D7wZWB+OSN7LXBtRLwW+OeIWE0RIDdn5sJDqP0vgeuAXx/S/lNgxqDbrxj0e/8IxxqpncxcRrHEZKTtL0bErRTrnudQPKY3DbPrixQz9QuAH5RtJwLPjVLHsxRXWxnxiiOS1CwGakkaRmZuj4ibKNYD7wceogh9bwXenJnXDbnLLGAPxcz2DAYFzYg4kmIpxDfK4+4A+spt7wA2UMzcDrT3UZxktyMirgc+B+wFfhHoLMP5aLXvL5ecDD2BcR1wSUSsBF4H/CbwT+W2FyhOijwZ2HjQJ2gEEXELxZU+NlCcnHkV0JOZWyPiPoplGb8NfJXiTcf8zFwXEV+mWBu9hCJ8X0NxEudI7gQ+HhHrMvOJ8qTGt2XmVw61dkk6VC75kKQRZOZnKYLdRykC57MUs8tfG2b3v6JYpvAcxSzrd4Zsfw+wpQzTHwDeXbYvBB6muIrHY8AdmfkvmdkH/BrFiXubKWZx/4IihI7F/cCPh7TdCPwCxcz5HwN/M+ix9lKcdPhouYzizDH2M9QMihM5X6I4iXABcGHZxw8plr1cS7FEYx1FsAe4mmIGfRPw7bK2LzKCzHwAuAX42/I5XU9xhRBJarq2/v4RP82TJEmSdBDOUEuSJEkVGKglSZKkCgzUkiRJUgUGakmSJKkCA7UkSZJUgYFakiRJqsBALUmSJFVgoJYkSZIq+B9b+RG1gNrIzwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "meanprops = {'marker':'o', 'markeredgecolor':'firebrick',\n",
    "            'markerfacecolor':'firebrick'}\n",
    "\n",
    "plt.rcParams[\"figure.figsize\"] = (12, 8)\n",
    "sns.boxplot(x='nutriscore_grade', y='additives_n', \n",
    "            data=open_food_facts_df, color='tab:blue',\n",
    "            order=['a', 'b', 'c', 'd', 'e'],\n",
    "            showmeans=True,\n",
    "            meanprops=meanprops)\n",
    "plt.title('Relation de la classe Nutriscore avec le nombre d\\'additifs parmi les produits')\n",
    "plt.xlabel('Classe Nutri-Score')\n",
    "plt.ylabel('Nombre d\\'additifs')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "On observe des nombres d'additifs moyen et médian plus faible dans les classes de produits de meilleure qualité nutritionnelle. Voyons si cette différence est significative avec une ANOVA."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Vérifions d'abord les hypothèses nécessaires à l'ANOVA."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Classe a*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(-3.8249999999999957, 0.5, 'Nombre de produits de classe a')"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Distribution du nombre d'additifs parmi la classe a des produits\n",
    "filtered_data = open_food_facts_df[open_food_facts_df['nutriscore_grade'] == 'a']\n",
    "nombre_produits = filtered_data['additives_n'].count()\n",
    "sns.displot(x='additives_n', data=filtered_data, discrete=True)\n",
    "plt.title('Répartition des {} produits de classe a selon leur nombre d\\'additifs'.format(nombre_produits))\n",
    "plt.xlabel('Nombre d\\'additifs')\n",
    "plt.ylabel('Nombre de produits de classe a')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NormaltestResult(statistic=9653.887174156096, pvalue=0.0)"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "st.normaltest(filtered_data['additives_n'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "La distribution ne suit donc pas une loi normale avec un risque de 5%. L'ANOVA n'est donc pas faisable."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Conclusion de cette hypothèse :** selon les résultats empiriques, il semblerait que plus nous avons un nombre d'additifs faible dans les produits et meilleure est la qualité nutritionnelle des produits. Cependant, cela n'est pas significatif car les hypothèses du test d'ANOVA n'ont pas été vérifiées et donc cela n'a pas pu être testé.\n",
    "En revanche, il y a une dépendance significative entre le nombre d'additifs et la qualité nutritionnelle."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 104300 entries, 0 to 104299\n",
      "Data columns (total 10 columns):\n",
      "brands_tags            103844 non-null object\n",
      "labels_en              71866 non-null object\n",
      "additives_tags         73811 non-null object\n",
      "nutriscore_grade       104300 non-null object\n",
      "pnns_groups_1          104159 non-null object\n",
      "pnns_groups_2          104300 non-null object\n",
      "additives_n            104300 non-null float64\n",
      "nutriscore_score       104300 non-null float64\n",
      "nova_group             104067 non-null object\n",
      "additives_n_classes    104300 non-null object\n",
      "dtypes: float64(2), object(8)\n",
      "memory usage: 8.0+ MB\n"
     ]
    }
   ],
   "source": [
    "open_food_facts_df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Y a-t-il plus d'additifs dans les produits ultra-transformés ? Ou la présence d'additifs est-elle plus importante parmi ces produits ?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, \"Nombre d'additifs\")"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "meanprops = {'marker':'o', 'markeredgecolor':'firebrick',\n",
    "            'markerfacecolor':'firebrick'}\n",
    "\n",
    "plt.rcParams[\"figure.figsize\"] = (12, 8)\n",
    "sns.boxplot(x='nova_group', y='additives_n', \n",
    "            data=open_food_facts_df, color='tab:blue',\n",
    "            order=[1, 2, 3, 4],\n",
    "            showmeans=True,\n",
    "            meanprops=meanprops)\n",
    "plt.title('Relation de la transformation des produits avec le nombre d\\'additifs')\n",
    "plt.xlabel('Classification Nova Group')\n",
    "plt.ylabel('Nombre d\\'additifs')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "On peut voir que plus le produit est transformé et plus le nombre moyen d'additifs dans le produit est élevé. De plus, les produits avec les nombres d'additifs les plus élevés font partie des produits ultra-transformés."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Montrons que cette corrélation est significative avec un test du khi-2, en utilisant le nombre d'additifs regroupé par classe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>additives_n_classes</th>\n",
       "      <th>1 ou 2 additifs</th>\n",
       "      <th>Entre 3 et 7 additifs</th>\n",
       "      <th>Pas d'additif</th>\n",
       "      <th>Plus de 7 additifs</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>nova_group</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>1.0</td>\n",
       "      <td>848.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>7766.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2.0</td>\n",
       "      <td>22.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1002.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3.0</td>\n",
       "      <td>7196.0</td>\n",
       "      <td>538.0</td>\n",
       "      <td>15058.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4.0</td>\n",
       "      <td>32239.0</td>\n",
       "      <td>28360.0</td>\n",
       "      <td>6663.0</td>\n",
       "      <td>4371.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "additives_n_classes  1 ou 2 additifs  Entre 3 et 7 additifs  Pas d'additif  \\\n",
       "nova_group                                                                   \n",
       "1.0                            848.0                    4.0         7766.0   \n",
       "2.0                             22.0                    NaN         1002.0   \n",
       "3.0                           7196.0                  538.0        15058.0   \n",
       "4.0                          32239.0                28360.0         6663.0   \n",
       "\n",
       "additives_n_classes  Plus de 7 additifs  \n",
       "nova_group                               \n",
       "1.0                                 NaN  \n",
       "2.0                                 NaN  \n",
       "3.0                                 NaN  \n",
       "4.0                              4371.0  "
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = 'nova_group'\n",
    "Y = 'additives_n_classes'\n",
    "\n",
    "tableau_contingence = open_food_facts_df[[X, Y]].pivot_table(index=X, \n",
    "                                                             columns=Y, \n",
    "                                                             aggfunc=len)\n",
    "\n",
    "tableau_contingence"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Regroupons les classes entre elles car les effectifs sont insuffisants pour un test du khi-2.\n",
    "Voici les nouvelles classes :\n",
    "- \"Sans additif\" et \"Avec additifs\"\n",
    "- \"Peu ou pas de transformation\" et \"Ultra-transformation\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "open_food_facts_df.loc[(open_food_facts_df['nova_group'] == 1) | (open_food_facts_df['nova_group'] == 2) |\n",
    "                       (open_food_facts_df['nova_group'] == 3), \n",
    "                       'nova_group_classes'] = 'Peu ou pas de transformation'\n",
    "open_food_facts_df.loc[open_food_facts_df['nova_group'] == 4, 'nova_group_classes'] = 'Ultra transformation'\n",
    "open_food_facts_df.loc[open_food_facts_df['additives_n'] == 0, 'additives_n_classes_k2'] = 'Sans additif'\n",
    "open_food_facts_df.loc[open_food_facts_df['additives_n'] > 0, 'additives_n_classes_k2'] = 'Avec additif'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>additives_n_classes</th>\n",
       "      <th>additives_n_classes_k2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>5037</th>\n",
       "      <td>Entre 3 et 7 additifs</td>\n",
       "      <td>Avec additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29541</th>\n",
       "      <td>Pas d'additif</td>\n",
       "      <td>Sans additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80003</th>\n",
       "      <td>1 ou 2 additifs</td>\n",
       "      <td>Avec additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>79957</th>\n",
       "      <td>Pas d'additif</td>\n",
       "      <td>Sans additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>81444</th>\n",
       "      <td>Pas d'additif</td>\n",
       "      <td>Sans additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>830</th>\n",
       "      <td>Pas d'additif</td>\n",
       "      <td>Sans additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56419</th>\n",
       "      <td>Pas d'additif</td>\n",
       "      <td>Sans additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>61699</th>\n",
       "      <td>1 ou 2 additifs</td>\n",
       "      <td>Avec additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33279</th>\n",
       "      <td>Entre 3 et 7 additifs</td>\n",
       "      <td>Avec additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96600</th>\n",
       "      <td>Entre 3 et 7 additifs</td>\n",
       "      <td>Avec additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40981</th>\n",
       "      <td>1 ou 2 additifs</td>\n",
       "      <td>Avec additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>77485</th>\n",
       "      <td>1 ou 2 additifs</td>\n",
       "      <td>Avec additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9590</th>\n",
       "      <td>Entre 3 et 7 additifs</td>\n",
       "      <td>Avec additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97527</th>\n",
       "      <td>Plus de 7 additifs</td>\n",
       "      <td>Avec additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57553</th>\n",
       "      <td>1 ou 2 additifs</td>\n",
       "      <td>Avec additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51464</th>\n",
       "      <td>Pas d'additif</td>\n",
       "      <td>Sans additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>84048</th>\n",
       "      <td>1 ou 2 additifs</td>\n",
       "      <td>Avec additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42857</th>\n",
       "      <td>Pas d'additif</td>\n",
       "      <td>Sans additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55555</th>\n",
       "      <td>Pas d'additif</td>\n",
       "      <td>Sans additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7211</th>\n",
       "      <td>1 ou 2 additifs</td>\n",
       "      <td>Avec additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52826</th>\n",
       "      <td>1 ou 2 additifs</td>\n",
       "      <td>Avec additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37639</th>\n",
       "      <td>Pas d'additif</td>\n",
       "      <td>Sans additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40336</th>\n",
       "      <td>Entre 3 et 7 additifs</td>\n",
       "      <td>Avec additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38153</th>\n",
       "      <td>Entre 3 et 7 additifs</td>\n",
       "      <td>Avec additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11667</th>\n",
       "      <td>1 ou 2 additifs</td>\n",
       "      <td>Avec additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>86249</th>\n",
       "      <td>Pas d'additif</td>\n",
       "      <td>Sans additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>69128</th>\n",
       "      <td>Entre 3 et 7 additifs</td>\n",
       "      <td>Avec additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30737</th>\n",
       "      <td>Pas d'additif</td>\n",
       "      <td>Sans additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54122</th>\n",
       "      <td>Pas d'additif</td>\n",
       "      <td>Sans additif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76882</th>\n",
       "      <td>Entre 3 et 7 additifs</td>\n",
       "      <td>Avec additif</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         additives_n_classes additives_n_classes_k2\n",
       "5037   Entre 3 et 7 additifs           Avec additif\n",
       "29541          Pas d'additif           Sans additif\n",
       "80003        1 ou 2 additifs           Avec additif\n",
       "79957          Pas d'additif           Sans additif\n",
       "81444          Pas d'additif           Sans additif\n",
       "830            Pas d'additif           Sans additif\n",
       "56419          Pas d'additif           Sans additif\n",
       "61699        1 ou 2 additifs           Avec additif\n",
       "33279  Entre 3 et 7 additifs           Avec additif\n",
       "96600  Entre 3 et 7 additifs           Avec additif\n",
       "40981        1 ou 2 additifs           Avec additif\n",
       "77485        1 ou 2 additifs           Avec additif\n",
       "9590   Entre 3 et 7 additifs           Avec additif\n",
       "97527     Plus de 7 additifs           Avec additif\n",
       "57553        1 ou 2 additifs           Avec additif\n",
       "51464          Pas d'additif           Sans additif\n",
       "84048        1 ou 2 additifs           Avec additif\n",
       "42857          Pas d'additif           Sans additif\n",
       "55555          Pas d'additif           Sans additif\n",
       "7211         1 ou 2 additifs           Avec additif\n",
       "52826        1 ou 2 additifs           Avec additif\n",
       "37639          Pas d'additif           Sans additif\n",
       "40336  Entre 3 et 7 additifs           Avec additif\n",
       "38153  Entre 3 et 7 additifs           Avec additif\n",
       "11667        1 ou 2 additifs           Avec additif\n",
       "86249          Pas d'additif           Sans additif\n",
       "69128  Entre 3 et 7 additifs           Avec additif\n",
       "30737          Pas d'additif           Sans additif\n",
       "54122          Pas d'additif           Sans additif\n",
       "76882  Entre 3 et 7 additifs           Avec additif"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df[['additives_n_classes', 'additives_n_classes_k2']].sample(30)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>nova_group</th>\n",
       "      <th>nova_group_classes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>22438</th>\n",
       "      <td>4</td>\n",
       "      <td>Ultra transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44430</th>\n",
       "      <td>1</td>\n",
       "      <td>Peu ou pas de transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>65123</th>\n",
       "      <td>3</td>\n",
       "      <td>Peu ou pas de transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4853</th>\n",
       "      <td>4</td>\n",
       "      <td>Ultra transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45121</th>\n",
       "      <td>3</td>\n",
       "      <td>Peu ou pas de transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49710</th>\n",
       "      <td>4</td>\n",
       "      <td>Ultra transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>85822</th>\n",
       "      <td>4</td>\n",
       "      <td>Ultra transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14709</th>\n",
       "      <td>3</td>\n",
       "      <td>Peu ou pas de transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>402</th>\n",
       "      <td>3</td>\n",
       "      <td>Peu ou pas de transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96556</th>\n",
       "      <td>4</td>\n",
       "      <td>Ultra transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98766</th>\n",
       "      <td>4</td>\n",
       "      <td>Ultra transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67279</th>\n",
       "      <td>1</td>\n",
       "      <td>Peu ou pas de transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>58286</th>\n",
       "      <td>3</td>\n",
       "      <td>Peu ou pas de transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>79412</th>\n",
       "      <td>1</td>\n",
       "      <td>Peu ou pas de transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100014</th>\n",
       "      <td>4</td>\n",
       "      <td>Ultra transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>102618</th>\n",
       "      <td>4</td>\n",
       "      <td>Ultra transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>74760</th>\n",
       "      <td>3</td>\n",
       "      <td>Peu ou pas de transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>95466</th>\n",
       "      <td>3</td>\n",
       "      <td>Peu ou pas de transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5067</th>\n",
       "      <td>4</td>\n",
       "      <td>Ultra transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>94356</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7258</th>\n",
       "      <td>4</td>\n",
       "      <td>Ultra transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18649</th>\n",
       "      <td>3</td>\n",
       "      <td>Peu ou pas de transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32472</th>\n",
       "      <td>4</td>\n",
       "      <td>Ultra transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2241</th>\n",
       "      <td>4</td>\n",
       "      <td>Ultra transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60025</th>\n",
       "      <td>3</td>\n",
       "      <td>Peu ou pas de transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>102891</th>\n",
       "      <td>4</td>\n",
       "      <td>Ultra transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9612</th>\n",
       "      <td>4</td>\n",
       "      <td>Ultra transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82779</th>\n",
       "      <td>1</td>\n",
       "      <td>Peu ou pas de transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45366</th>\n",
       "      <td>4</td>\n",
       "      <td>Ultra transformation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37742</th>\n",
       "      <td>4</td>\n",
       "      <td>Ultra transformation</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       nova_group            nova_group_classes\n",
       "22438           4          Ultra transformation\n",
       "44430           1  Peu ou pas de transformation\n",
       "65123           3  Peu ou pas de transformation\n",
       "4853            4          Ultra transformation\n",
       "45121           3  Peu ou pas de transformation\n",
       "49710           4          Ultra transformation\n",
       "85822           4          Ultra transformation\n",
       "14709           3  Peu ou pas de transformation\n",
       "402             3  Peu ou pas de transformation\n",
       "96556           4          Ultra transformation\n",
       "98766           4          Ultra transformation\n",
       "67279           1  Peu ou pas de transformation\n",
       "58286           3  Peu ou pas de transformation\n",
       "79412           1  Peu ou pas de transformation\n",
       "100014          4          Ultra transformation\n",
       "102618          4          Ultra transformation\n",
       "74760           3  Peu ou pas de transformation\n",
       "95466           3  Peu ou pas de transformation\n",
       "5067            4          Ultra transformation\n",
       "94356         NaN                           NaN\n",
       "7258            4          Ultra transformation\n",
       "18649           3  Peu ou pas de transformation\n",
       "32472           4          Ultra transformation\n",
       "2241            4          Ultra transformation\n",
       "60025           3  Peu ou pas de transformation\n",
       "102891          4          Ultra transformation\n",
       "9612            4          Ultra transformation\n",
       "82779           1  Peu ou pas de transformation\n",
       "45366           4          Ultra transformation\n",
       "37742           4          Ultra transformation"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df[['nova_group', 'nova_group_classes']].sample(30)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>additives_n_classes_k2</th>\n",
       "      <th>Avec additif</th>\n",
       "      <th>Sans additif</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>nova_group_classes</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>Peu ou pas de transformation</td>\n",
       "      <td>8608</td>\n",
       "      <td>23826</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Ultra transformation</td>\n",
       "      <td>64970</td>\n",
       "      <td>6663</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "additives_n_classes_k2        Avec additif  Sans additif\n",
       "nova_group_classes                                      \n",
       "Peu ou pas de transformation          8608         23826\n",
       "Ultra transformation                 64970          6663"
      ]
     },
     "execution_count": 98,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = 'nova_group_classes'\n",
    "Y = 'additives_n_classes_k2'\n",
    "\n",
    "tableau_contingence = open_food_facts_df[[X, Y]].pivot_table(index=X, \n",
    "                                                             columns=Y, \n",
    "                                                             aggfunc=len)\n",
    "\n",
    "tableau_contingence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(44362.075725243056, 0.0, 1, array([[22931.65798956,  9502.34201044],\n",
       "        [50646.34201044, 20986.65798956]]))"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "st.chi2_contingency(tableau_contingence)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Avec un risque de 5%, nous rejetons l'hypothèse déclarant que les variables 'nova_group_classes' et 'additives_n_classes_k2' sont indépendantes. Par conséquent, la transformation du produit et la présence d'additifs sont significativement corrélés."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Conclusion de cette hypothèse :** la corrélation entre la transformation des produits et la présence d'additifs est significatif pour un risque de 5% (p-value = 0.0). Grâce aux résultats empiriques, on observe que plus un produit est transformé et plus le nombre d'additifs qu'il contient est élevé."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 104300 entries, 0 to 104299\n",
      "Data columns (total 12 columns):\n",
      "brands_tags               103844 non-null object\n",
      "labels_en                 71866 non-null object\n",
      "additives_tags            73811 non-null object\n",
      "nutriscore_grade          104300 non-null object\n",
      "pnns_groups_1             104159 non-null object\n",
      "pnns_groups_2             104300 non-null object\n",
      "additives_n               104300 non-null float64\n",
      "nutriscore_score          104300 non-null float64\n",
      "nova_group                104067 non-null object\n",
      "additives_n_classes       104300 non-null object\n",
      "nova_group_classes        104067 non-null object\n",
      "additives_n_classes_k2    104300 non-null object\n",
      "dtypes: float64(2), object(10)\n",
      "memory usage: 9.5+ MB\n"
     ]
    }
   ],
   "source": [
    "open_food_facts_df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Y a-t-il plus d'additifs dans certaines catégories de produits ? Ou leur présence est-elle plus forte dans certaines d'entre elles ?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, \"Nombre d'additifs\")"
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "meanprops = {'marker':'o', 'markeredgecolor':'firebrick',\n",
    "            'markerfacecolor':'firebrick'}\n",
    "\n",
    "plt.rcParams[\"figure.figsize\"] = (12, 8)\n",
    "sns.boxplot(x='additives_n', y='pnns_groups_1', \n",
    "            data=open_food_facts_df, color='tab:blue',\n",
    "            showmeans=True,\n",
    "            meanprops=meanprops)\n",
    "plt.title('Relation des catégories de produits avec le nombre d\\'additifs')\n",
    "plt.ylabel('Catégorie de produits')\n",
    "plt.xlabel('Nombre d\\'additifs')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Parmi les catégories \"Plats préparés\" et \"Snacks sucrés\", le nombre moyen et médian d'additifs est plus élevé que dans les autres catégories. De plus, la dispersion de ce nombre y sont plus fort."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Vérifions s'il y a une corrélation significative avec un test du khi-2 grâce aux classes créées."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>additives_n_classes</th>\n",
       "      <th>1 ou 2 additifs</th>\n",
       "      <th>Entre 3 et 7 additifs</th>\n",
       "      <th>Pas d'additif</th>\n",
       "      <th>Plus de 7 additifs</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pnns_groups_1</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>Beverages</td>\n",
       "      <td>2869</td>\n",
       "      <td>2021</td>\n",
       "      <td>2786</td>\n",
       "      <td>131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Cereals and potatoes</td>\n",
       "      <td>2584</td>\n",
       "      <td>1245</td>\n",
       "      <td>4822</td>\n",
       "      <td>74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Composite foods</td>\n",
       "      <td>4488</td>\n",
       "      <td>4714</td>\n",
       "      <td>2702</td>\n",
       "      <td>1194</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Fat and sauces</td>\n",
       "      <td>2129</td>\n",
       "      <td>1472</td>\n",
       "      <td>1694</td>\n",
       "      <td>101</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Fish Meat Eggs</td>\n",
       "      <td>5177</td>\n",
       "      <td>3694</td>\n",
       "      <td>3458</td>\n",
       "      <td>360</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Fruits and vegetables</td>\n",
       "      <td>3371</td>\n",
       "      <td>339</td>\n",
       "      <td>3135</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Milk and dairy products</td>\n",
       "      <td>3746</td>\n",
       "      <td>3358</td>\n",
       "      <td>4969</td>\n",
       "      <td>387</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Salty snacks</td>\n",
       "      <td>2943</td>\n",
       "      <td>1599</td>\n",
       "      <td>2250</td>\n",
       "      <td>105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Sugary snacks</td>\n",
       "      <td>10874</td>\n",
       "      <td>8619</td>\n",
       "      <td>2882</td>\n",
       "      <td>1719</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>unknown</td>\n",
       "      <td>2288</td>\n",
       "      <td>1777</td>\n",
       "      <td>1791</td>\n",
       "      <td>285</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "additives_n_classes      1 ou 2 additifs  Entre 3 et 7 additifs  \\\n",
       "pnns_groups_1                                                     \n",
       "Beverages                           2869                   2021   \n",
       "Cereals and potatoes                2584                   1245   \n",
       "Composite foods                     4488                   4714   \n",
       "Fat and sauces                      2129                   1472   \n",
       "Fish Meat Eggs                      5177                   3694   \n",
       "Fruits and vegetables               3371                    339   \n",
       "Milk and dairy products             3746                   3358   \n",
       "Salty snacks                        2943                   1599   \n",
       "Sugary snacks                      10874                   8619   \n",
       "unknown                             2288                   1777   \n",
       "\n",
       "additives_n_classes      Pas d'additif  Plus de 7 additifs  \n",
       "pnns_groups_1                                               \n",
       "Beverages                         2786                 131  \n",
       "Cereals and potatoes              4822                  74  \n",
       "Composite foods                   2702                1194  \n",
       "Fat and sauces                    1694                 101  \n",
       "Fish Meat Eggs                    3458                 360  \n",
       "Fruits and vegetables             3135                   7  \n",
       "Milk and dairy products           4969                 387  \n",
       "Salty snacks                      2250                 105  \n",
       "Sugary snacks                     2882                1719  \n",
       "unknown                           1791                 285  "
      ]
     },
     "execution_count": 111,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = 'pnns_groups_1'\n",
    "Y = 'additives_n_classes'\n",
    "\n",
    "tableau_contingence = open_food_facts_df[[X, Y]].pivot_table(index=X, \n",
    "                                                             columns=Y, \n",
    "                                                             aggfunc=len)\n",
    "\n",
    "tableau_contingence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(12065.09679981136,\n",
       " 0.0,\n",
       " 27,\n",
       " array([[3033.26148485, 2161.486439  , 2285.23337398,  327.01870218],\n",
       "        [3389.93293906, 2415.64867174, 2553.94661047,  365.47177872],\n",
       "        [5088.97898405, 3626.38009197, 3833.99343312,  548.64749086],\n",
       "        [2096.51325378, 1493.96449659, 1579.49523325,  226.02701639],\n",
       "        [4930.06980674, 3513.14223447, 3714.27261206,  531.51534673],\n",
       "        [2662.21438378, 1897.0801947 , 2005.68964756,  287.01577396],\n",
       "        [4841.09620868, 3449.74010887, 3647.24066091,  521.92302153],\n",
       "        [2679.69827859, 1909.53912768, 2018.86186503,  288.90072869],\n",
       "        [9361.26581476, 6670.78958131, 7052.69795217, 1009.24665175],\n",
       "        [2385.96884571, 1700.22905366, 1797.56861145,  257.23348918]]))"
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "st.chi2_contingency(tableau_contingence)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "En effet, la corrélation entre le nombre d'additifs et la catégorie de produits est significative au seuil de 5% (p-value = 0.0). Faisons le même test avec la présence ou absence d'additifs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>additives_n_classes_k2</th>\n",
       "      <th>Avec additif</th>\n",
       "      <th>Sans additif</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pnns_groups_1</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>Beverages</td>\n",
       "      <td>5021</td>\n",
       "      <td>2786</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Cereals and potatoes</td>\n",
       "      <td>3903</td>\n",
       "      <td>4822</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Composite foods</td>\n",
       "      <td>10396</td>\n",
       "      <td>2702</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Fat and sauces</td>\n",
       "      <td>3702</td>\n",
       "      <td>1694</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Fish Meat Eggs</td>\n",
       "      <td>9231</td>\n",
       "      <td>3458</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Fruits and vegetables</td>\n",
       "      <td>3717</td>\n",
       "      <td>3135</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Milk and dairy products</td>\n",
       "      <td>7491</td>\n",
       "      <td>4969</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Salty snacks</td>\n",
       "      <td>4647</td>\n",
       "      <td>2250</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Sugary snacks</td>\n",
       "      <td>21212</td>\n",
       "      <td>2882</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>unknown</td>\n",
       "      <td>4350</td>\n",
       "      <td>1791</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "additives_n_classes_k2   Avec additif  Sans additif\n",
       "pnns_groups_1                                      \n",
       "Beverages                        5021          2786\n",
       "Cereals and potatoes             3903          4822\n",
       "Composite foods                 10396          2702\n",
       "Fat and sauces                   3702          1694\n",
       "Fish Meat Eggs                   9231          3458\n",
       "Fruits and vegetables            3717          3135\n",
       "Milk and dairy products          7491          4969\n",
       "Salty snacks                     4647          2250\n",
       "Sugary snacks                   21212          2882\n",
       "unknown                          4350          1791"
      ]
     },
     "execution_count": 113,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = 'pnns_groups_1'\n",
    "Y = 'additives_n_classes_k2'\n",
    "\n",
    "tableau_contingence = open_food_facts_df[[X, Y]].pivot_table(index=X, \n",
    "                                                             columns=Y, \n",
    "                                                             aggfunc=len)\n",
    "\n",
    "tableau_contingence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(8613.017633496876, 0.0, 9, array([[ 5521.76662602,  2285.23337398],\n",
       "        [ 6171.05338953,  2553.94661047],\n",
       "        [ 9264.00656688,  3833.99343312],\n",
       "        [ 3816.50476675,  1579.49523325],\n",
       "        [ 8974.72738794,  3714.27261206],\n",
       "        [ 4846.31035244,  2005.68964756],\n",
       "        [ 8812.75933909,  3647.24066091],\n",
       "        [ 4878.13813497,  2018.86186503],\n",
       "        [17041.30204783,  7052.69795217],\n",
       "        [ 4343.43138855,  1797.56861145]]))"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "st.chi2_contingency(tableau_contingence)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "La corrélation est également significative au seuil de 5% (p-value = 0.0) entre la présence d'additifs et la catégorie de produits."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Conclusion de cette hypothèse :** la corrélation entre la catégorie des produits et la présence d'additifs est significative pour un risque de 5% (p-value = 0.0) mais également entre la catégorie des produits et le nombre d'additifs. Empiriquement, les produits des catégories \"Plats préparés\" et \"Snacks sucrés\" ont un nombre moyen et médian plus élevés que dans les autres catégories."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Y a-t-il plus d'additifs dans certaines sous-catégories de produits ? Ou leur présence est-elle plus forte dans certaines d'entre elles ?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, \"Nombre d'additifs\")"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x864 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "meanprops = {'marker':'o', 'markeredgecolor':'firebrick',\n",
    "            'markerfacecolor':'firebrick'}\n",
    "\n",
    "plt.rcParams[\"figure.figsize\"] = (12, 12)\n",
    "sns.boxplot(x='additives_n', y='pnns_groups_2', \n",
    "            data=open_food_facts_df, color='tab:blue',\n",
    "            showmeans=True,\n",
    "            meanprops=meanprops)\n",
    "plt.title('Relation des sous-catégories de produits avec le nombre d\\'additifs')\n",
    "plt.ylabel('Sous-catégorie de produits')\n",
    "plt.xlabel('Nombre d\\'additifs')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Le nombre d'additifs moyen et médian sont plus élevés parmi les catégories \"Glaces\", \"Boissons sucrées édulcorées\", \"Sandwichs\", \"Pâtisseries\", \"Pizzas, tartes et quiches\" et \"Biscuits et gâteaux\"."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Vérifions si cette dépendance est significative avec un test du khi-2."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>additives_n_classes</th>\n",
       "      <th>1 ou 2 additifs</th>\n",
       "      <th>Entre 3 et 7 additifs</th>\n",
       "      <th>Pas d'additif</th>\n",
       "      <th>Plus de 7 additifs</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pnns_groups_2</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>Alcoholic beverages</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Appetizers</td>\n",
       "      <td>936.0</td>\n",
       "      <td>792.0</td>\n",
       "      <td>893.0</td>\n",
       "      <td>79.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Artificially sweetened beverages</td>\n",
       "      <td>164.0</td>\n",
       "      <td>826.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>109.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Biscuits and cakes</td>\n",
       "      <td>3314.0</td>\n",
       "      <td>5802.0</td>\n",
       "      <td>867.0</td>\n",
       "      <td>1087.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Bread</td>\n",
       "      <td>894.0</td>\n",
       "      <td>620.0</td>\n",
       "      <td>760.0</td>\n",
       "      <td>48.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Breakfast cereals</td>\n",
       "      <td>634.0</td>\n",
       "      <td>194.0</td>\n",
       "      <td>683.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Cereals</td>\n",
       "      <td>601.0</td>\n",
       "      <td>219.0</td>\n",
       "      <td>2663.0</td>\n",
       "      <td>22.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Cheese</td>\n",
       "      <td>1361.0</td>\n",
       "      <td>358.0</td>\n",
       "      <td>2512.0</td>\n",
       "      <td>19.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Chocolate products</td>\n",
       "      <td>2461.0</td>\n",
       "      <td>301.0</td>\n",
       "      <td>813.0</td>\n",
       "      <td>23.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Dairy desserts</td>\n",
       "      <td>691.0</td>\n",
       "      <td>923.0</td>\n",
       "      <td>505.0</td>\n",
       "      <td>96.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Dressings and sauces</td>\n",
       "      <td>2040.0</td>\n",
       "      <td>1374.0</td>\n",
       "      <td>842.0</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Dried fruits</td>\n",
       "      <td>321.0</td>\n",
       "      <td>29.0</td>\n",
       "      <td>495.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Eggs</td>\n",
       "      <td>15.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>188.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Fats</td>\n",
       "      <td>89.0</td>\n",
       "      <td>98.0</td>\n",
       "      <td>852.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Fish and seafood</td>\n",
       "      <td>739.0</td>\n",
       "      <td>398.0</td>\n",
       "      <td>1475.0</td>\n",
       "      <td>35.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Fruit juices</td>\n",
       "      <td>180.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>1261.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Fruit nectars</td>\n",
       "      <td>37.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>51.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Fruits</td>\n",
       "      <td>1472.0</td>\n",
       "      <td>124.0</td>\n",
       "      <td>598.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Ice cream</td>\n",
       "      <td>518.0</td>\n",
       "      <td>1054.0</td>\n",
       "      <td>91.0</td>\n",
       "      <td>188.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Meat</td>\n",
       "      <td>1029.0</td>\n",
       "      <td>835.0</td>\n",
       "      <td>832.0</td>\n",
       "      <td>79.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Milk and yogurt</td>\n",
       "      <td>1176.0</td>\n",
       "      <td>1023.0</td>\n",
       "      <td>1861.0</td>\n",
       "      <td>84.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Nuts</td>\n",
       "      <td>127.0</td>\n",
       "      <td>47.0</td>\n",
       "      <td>400.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Offals</td>\n",
       "      <td>162.0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>140.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>One-dish meals</td>\n",
       "      <td>4066.0</td>\n",
       "      <td>3579.0</td>\n",
       "      <td>2488.0</td>\n",
       "      <td>560.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Pizza pies and quiche</td>\n",
       "      <td>382.0</td>\n",
       "      <td>778.0</td>\n",
       "      <td>169.0</td>\n",
       "      <td>104.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Plant-based milk substitutes</td>\n",
       "      <td>252.0</td>\n",
       "      <td>49.0</td>\n",
       "      <td>271.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Potatoes</td>\n",
       "      <td>261.0</td>\n",
       "      <td>201.0</td>\n",
       "      <td>205.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Processed meat</td>\n",
       "      <td>3232.0</td>\n",
       "      <td>2417.0</td>\n",
       "      <td>823.0</td>\n",
       "      <td>246.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Salty and fatty products</td>\n",
       "      <td>1880.0</td>\n",
       "      <td>760.0</td>\n",
       "      <td>957.0</td>\n",
       "      <td>24.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Sandwiches</td>\n",
       "      <td>93.0</td>\n",
       "      <td>435.0</td>\n",
       "      <td>45.0</td>\n",
       "      <td>538.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Soups</td>\n",
       "      <td>394.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>425.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Sweetened beverages</td>\n",
       "      <td>1827.0</td>\n",
       "      <td>1034.0</td>\n",
       "      <td>405.0</td>\n",
       "      <td>18.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Sweets</td>\n",
       "      <td>4623.0</td>\n",
       "      <td>1839.0</td>\n",
       "      <td>1022.0</td>\n",
       "      <td>559.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Teas and herbal teas and coffees</td>\n",
       "      <td>12.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Unsweetened beverages</td>\n",
       "      <td>357.0</td>\n",
       "      <td>94.0</td>\n",
       "      <td>651.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Vegetables</td>\n",
       "      <td>1378.0</td>\n",
       "      <td>170.0</td>\n",
       "      <td>2128.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Waters and flavored waters</td>\n",
       "      <td>40.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>116.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>pastries</td>\n",
       "      <td>476.0</td>\n",
       "      <td>677.0</td>\n",
       "      <td>180.0</td>\n",
       "      <td>50.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>unknown</td>\n",
       "      <td>2288.0</td>\n",
       "      <td>1777.0</td>\n",
       "      <td>1791.0</td>\n",
       "      <td>285.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "additives_n_classes               1 ou 2 additifs  Entre 3 et 7 additifs  \\\n",
       "pnns_groups_2                                                              \n",
       "Alcoholic beverages                           2.0                    NaN   \n",
       "Appetizers                                  936.0                  792.0   \n",
       "Artificially sweetened beverages            164.0                  826.0   \n",
       "Biscuits and cakes                         3314.0                 5802.0   \n",
       "Bread                                       894.0                  620.0   \n",
       "Breakfast cereals                           634.0                  194.0   \n",
       "Cereals                                     601.0                  219.0   \n",
       "Cheese                                     1361.0                  358.0   \n",
       "Chocolate products                         2461.0                  301.0   \n",
       "Dairy desserts                              691.0                  923.0   \n",
       "Dressings and sauces                       2040.0                 1374.0   \n",
       "Dried fruits                                321.0                   29.0   \n",
       "Eggs                                         15.0                    4.0   \n",
       "Fats                                         89.0                   98.0   \n",
       "Fish and seafood                            739.0                  398.0   \n",
       "Fruit juices                                180.0                    7.0   \n",
       "Fruit nectars                                37.0                    4.0   \n",
       "Fruits                                     1472.0                  124.0   \n",
       "Ice cream                                   518.0                 1054.0   \n",
       "Meat                                       1029.0                  835.0   \n",
       "Milk and yogurt                            1176.0                 1023.0   \n",
       "Nuts                                        127.0                   47.0   \n",
       "Offals                                      162.0                   40.0   \n",
       "One-dish meals                             4066.0                 3579.0   \n",
       "Pizza pies and quiche                       382.0                  778.0   \n",
       "Plant-based milk substitutes                252.0                   49.0   \n",
       "Potatoes                                    261.0                  201.0   \n",
       "Processed meat                             3232.0                 2417.0   \n",
       "Salty and fatty products                   1880.0                  760.0   \n",
       "Sandwiches                                   93.0                  435.0   \n",
       "Soups                                       394.0                   27.0   \n",
       "Sweetened beverages                        1827.0                 1034.0   \n",
       "Sweets                                     4623.0                 1839.0   \n",
       "Teas and herbal teas and coffees             12.0                    3.0   \n",
       "Unsweetened beverages                       357.0                   94.0   \n",
       "Vegetables                                 1378.0                  170.0   \n",
       "Waters and flavored waters                   40.0                    4.0   \n",
       "pastries                                    476.0                  677.0   \n",
       "unknown                                    2288.0                 1777.0   \n",
       "\n",
       "additives_n_classes               Pas d'additif  Plus de 7 additifs  \n",
       "pnns_groups_2                                                        \n",
       "Alcoholic beverages                         NaN                 NaN  \n",
       "Appetizers                                893.0                79.0  \n",
       "Artificially sweetened beverages            3.0               109.0  \n",
       "Biscuits and cakes                        867.0              1087.0  \n",
       "Bread                                     760.0                48.0  \n",
       "Breakfast cereals                         683.0                 1.0  \n",
       "Cereals                                  2663.0                22.0  \n",
       "Cheese                                   2512.0                19.0  \n",
       "Chocolate products                        813.0                23.0  \n",
       "Dairy desserts                            505.0                96.0  \n",
       "Dressings and sauces                      842.0               100.0  \n",
       "Dried fruits                              495.0                 2.0  \n",
       "Eggs                                      188.0                 NaN  \n",
       "Fats                                      852.0                 1.0  \n",
       "Fish and seafood                         1475.0                35.0  \n",
       "Fruit juices                             1261.0                 1.0  \n",
       "Fruit nectars                              51.0                 NaN  \n",
       "Fruits                                    598.0                 1.0  \n",
       "Ice cream                                  91.0               188.0  \n",
       "Meat                                      832.0                79.0  \n",
       "Milk and yogurt                          1861.0                84.0  \n",
       "Nuts                                      400.0                 2.0  \n",
       "Offals                                    140.0                 NaN  \n",
       "One-dish meals                           2488.0               560.0  \n",
       "Pizza pies and quiche                     169.0               104.0  \n",
       "Plant-based milk substitutes              271.0                 NaN  \n",
       "Potatoes                                  205.0                 3.0  \n",
       "Processed meat                            823.0               246.0  \n",
       "Salty and fatty products                  957.0                24.0  \n",
       "Sandwiches                                 45.0               538.0  \n",
       "Soups                                     425.0                 1.0  \n",
       "Sweetened beverages                       405.0                18.0  \n",
       "Sweets                                   1022.0               559.0  \n",
       "Teas and herbal teas and coffees           28.0                 NaN  \n",
       "Unsweetened beverages                     651.0                 3.0  \n",
       "Vegetables                               2128.0                 3.0  \n",
       "Waters and flavored waters                116.0                 NaN  \n",
       "pastries                                  180.0                50.0  \n",
       "unknown                                  1791.0               285.0  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = 'pnns_groups_2'\n",
    "Y = 'additives_n_classes'\n",
    "\n",
    "tableau_contingence = open_food_facts_df[[X, Y]].pivot_table(index=X, \n",
    "                                                             columns=Y, \n",
    "                                                             aggfunc=len)\n",
    "\n",
    "tableau_contingence"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Certains effectifs sont insuffisants (< 5) pour réaliser le test du khi-2 sur ces données. De ce fait, nous allons utiliser la variable \"additives_n_classes_k2\" qui exprime la présence ou l'absence d'additifs. Et nous allons regrouper les boissons alcoolisés avec les boissons non sucrées."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "open_food_facts_df['pnns_groups_2_classes'] = open_food_facts_df['pnns_groups_2']\n",
    "open_food_facts_df.loc[open_food_facts_df['pnns_groups_2'] == 'Alcoholic beverages', \n",
    "                       'pnns_groups_2_classes'] = 'Unsweetened beverages'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>pnns_groups_2</th>\n",
       "      <th>pnns_groups_2_classes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>11749</td>\n",
       "      <td>Alcoholic beverages</td>\n",
       "      <td>Unsweetened beverages</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>100449</td>\n",
       "      <td>Alcoholic beverages</td>\n",
       "      <td>Unsweetened beverages</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              pnns_groups_2  pnns_groups_2_classes\n",
       "11749   Alcoholic beverages  Unsweetened beverages\n",
       "100449  Alcoholic beverages  Unsweetened beverages"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df.loc[open_food_facts_df['pnns_groups_2'] == 'Alcoholic beverages', \n",
    "                   ['pnns_groups_2', 'pnns_groups_2_classes']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>additives_n_classes_k2</th>\n",
       "      <th>Avec additif</th>\n",
       "      <th>Sans additif</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pnns_groups_2_classes</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>Appetizers</td>\n",
       "      <td>1807</td>\n",
       "      <td>893</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Artificially sweetened beverages</td>\n",
       "      <td>1099</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Biscuits and cakes</td>\n",
       "      <td>10203</td>\n",
       "      <td>867</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Bread</td>\n",
       "      <td>1562</td>\n",
       "      <td>760</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Breakfast cereals</td>\n",
       "      <td>829</td>\n",
       "      <td>683</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Cereals</td>\n",
       "      <td>842</td>\n",
       "      <td>2663</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Cheese</td>\n",
       "      <td>1738</td>\n",
       "      <td>2512</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Chocolate products</td>\n",
       "      <td>2785</td>\n",
       "      <td>813</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Dairy desserts</td>\n",
       "      <td>1710</td>\n",
       "      <td>505</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Dressings and sauces</td>\n",
       "      <td>3514</td>\n",
       "      <td>842</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Dried fruits</td>\n",
       "      <td>352</td>\n",
       "      <td>495</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Eggs</td>\n",
       "      <td>19</td>\n",
       "      <td>188</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Fats</td>\n",
       "      <td>188</td>\n",
       "      <td>852</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Fish and seafood</td>\n",
       "      <td>1172</td>\n",
       "      <td>1475</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Fruit juices</td>\n",
       "      <td>188</td>\n",
       "      <td>1261</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Fruit nectars</td>\n",
       "      <td>41</td>\n",
       "      <td>51</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Fruits</td>\n",
       "      <td>1597</td>\n",
       "      <td>598</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Ice cream</td>\n",
       "      <td>1760</td>\n",
       "      <td>91</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Meat</td>\n",
       "      <td>1943</td>\n",
       "      <td>832</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Milk and yogurt</td>\n",
       "      <td>2283</td>\n",
       "      <td>1861</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Nuts</td>\n",
       "      <td>176</td>\n",
       "      <td>400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Offals</td>\n",
       "      <td>202</td>\n",
       "      <td>140</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>One-dish meals</td>\n",
       "      <td>8205</td>\n",
       "      <td>2488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Pizza pies and quiche</td>\n",
       "      <td>1264</td>\n",
       "      <td>169</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Plant-based milk substitutes</td>\n",
       "      <td>301</td>\n",
       "      <td>271</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Potatoes</td>\n",
       "      <td>465</td>\n",
       "      <td>205</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Processed meat</td>\n",
       "      <td>5895</td>\n",
       "      <td>823</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Salty and fatty products</td>\n",
       "      <td>2664</td>\n",
       "      <td>957</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Sandwiches</td>\n",
       "      <td>1066</td>\n",
       "      <td>45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Soups</td>\n",
       "      <td>422</td>\n",
       "      <td>425</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Sweetened beverages</td>\n",
       "      <td>2879</td>\n",
       "      <td>405</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Sweets</td>\n",
       "      <td>7021</td>\n",
       "      <td>1022</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Teas and herbal teas and coffees</td>\n",
       "      <td>15</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Unsweetened beverages</td>\n",
       "      <td>456</td>\n",
       "      <td>651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Vegetables</td>\n",
       "      <td>1551</td>\n",
       "      <td>2128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Waters and flavored waters</td>\n",
       "      <td>44</td>\n",
       "      <td>116</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>pastries</td>\n",
       "      <td>1203</td>\n",
       "      <td>180</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>unknown</td>\n",
       "      <td>4350</td>\n",
       "      <td>1791</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "additives_n_classes_k2            Avec additif  Sans additif\n",
       "pnns_groups_2_classes                                       \n",
       "Appetizers                                1807           893\n",
       "Artificially sweetened beverages          1099             3\n",
       "Biscuits and cakes                       10203           867\n",
       "Bread                                     1562           760\n",
       "Breakfast cereals                          829           683\n",
       "Cereals                                    842          2663\n",
       "Cheese                                    1738          2512\n",
       "Chocolate products                        2785           813\n",
       "Dairy desserts                            1710           505\n",
       "Dressings and sauces                      3514           842\n",
       "Dried fruits                               352           495\n",
       "Eggs                                        19           188\n",
       "Fats                                       188           852\n",
       "Fish and seafood                          1172          1475\n",
       "Fruit juices                               188          1261\n",
       "Fruit nectars                               41            51\n",
       "Fruits                                    1597           598\n",
       "Ice cream                                 1760            91\n",
       "Meat                                      1943           832\n",
       "Milk and yogurt                           2283          1861\n",
       "Nuts                                       176           400\n",
       "Offals                                     202           140\n",
       "One-dish meals                            8205          2488\n",
       "Pizza pies and quiche                     1264           169\n",
       "Plant-based milk substitutes               301           271\n",
       "Potatoes                                   465           205\n",
       "Processed meat                            5895           823\n",
       "Salty and fatty products                  2664           957\n",
       "Sandwiches                                1066            45\n",
       "Soups                                      422           425\n",
       "Sweetened beverages                       2879           405\n",
       "Sweets                                    7021          1022\n",
       "Teas and herbal teas and coffees            15            28\n",
       "Unsweetened beverages                      456           651\n",
       "Vegetables                                1551          2128\n",
       "Waters and flavored waters                  44           116\n",
       "pastries                                  1203           180\n",
       "unknown                                   4350          1791"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = 'pnns_groups_2_classes'\n",
    "Y = 'additives_n_classes_k2'\n",
    "\n",
    "tableau_contingence = open_food_facts_df[[X, Y]].pivot_table(index=X, \n",
    "                                                             columns=Y, \n",
    "                                                             aggfunc=len)\n",
    "\n",
    "tableau_contingence"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ici encore, il reste un effectif < 5, les catégories \"Boissons sucrées\" et \"Boissons édulcorées\" seront regroupées."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "open_food_facts_df.loc[open_food_facts_df['pnns_groups_2'] == 'Artificially sweetened beverages', \n",
    "                       'pnns_groups_2_classes'] = 'Sweetened beverages'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>pnns_groups_2</th>\n",
       "      <th>pnns_groups_2_classes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>132</td>\n",
       "      <td>Artificially sweetened beverages</td>\n",
       "      <td>Sweetened beverages</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>230</td>\n",
       "      <td>Artificially sweetened beverages</td>\n",
       "      <td>Sweetened beverages</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>231</td>\n",
       "      <td>Artificially sweetened beverages</td>\n",
       "      <td>Sweetened beverages</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>355</td>\n",
       "      <td>Artificially sweetened beverages</td>\n",
       "      <td>Sweetened beverages</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>433</td>\n",
       "      <td>Artificially sweetened beverages</td>\n",
       "      <td>Sweetened beverages</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104105</td>\n",
       "      <td>Artificially sweetened beverages</td>\n",
       "      <td>Sweetened beverages</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104118</td>\n",
       "      <td>Artificially sweetened beverages</td>\n",
       "      <td>Sweetened beverages</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104119</td>\n",
       "      <td>Artificially sweetened beverages</td>\n",
       "      <td>Sweetened beverages</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104246</td>\n",
       "      <td>Artificially sweetened beverages</td>\n",
       "      <td>Sweetened beverages</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104278</td>\n",
       "      <td>Artificially sweetened beverages</td>\n",
       "      <td>Sweetened beverages</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1102 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                           pnns_groups_2 pnns_groups_2_classes\n",
       "132     Artificially sweetened beverages   Sweetened beverages\n",
       "230     Artificially sweetened beverages   Sweetened beverages\n",
       "231     Artificially sweetened beverages   Sweetened beverages\n",
       "355     Artificially sweetened beverages   Sweetened beverages\n",
       "433     Artificially sweetened beverages   Sweetened beverages\n",
       "...                                  ...                   ...\n",
       "104105  Artificially sweetened beverages   Sweetened beverages\n",
       "104118  Artificially sweetened beverages   Sweetened beverages\n",
       "104119  Artificially sweetened beverages   Sweetened beverages\n",
       "104246  Artificially sweetened beverages   Sweetened beverages\n",
       "104278  Artificially sweetened beverages   Sweetened beverages\n",
       "\n",
       "[1102 rows x 2 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df.loc[open_food_facts_df['pnns_groups_2'] == 'Artificially sweetened beverages', \n",
    "                   ['pnns_groups_2', 'pnns_groups_2_classes']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>additives_n_classes_k2</th>\n",
       "      <th>Avec additif</th>\n",
       "      <th>Sans additif</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pnns_groups_2_classes</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>Appetizers</td>\n",
       "      <td>1807</td>\n",
       "      <td>893</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Biscuits and cakes</td>\n",
       "      <td>10203</td>\n",
       "      <td>867</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Bread</td>\n",
       "      <td>1562</td>\n",
       "      <td>760</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Breakfast cereals</td>\n",
       "      <td>829</td>\n",
       "      <td>683</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Cereals</td>\n",
       "      <td>842</td>\n",
       "      <td>2663</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Cheese</td>\n",
       "      <td>1738</td>\n",
       "      <td>2512</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Chocolate products</td>\n",
       "      <td>2785</td>\n",
       "      <td>813</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Dairy desserts</td>\n",
       "      <td>1710</td>\n",
       "      <td>505</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Dressings and sauces</td>\n",
       "      <td>3514</td>\n",
       "      <td>842</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Dried fruits</td>\n",
       "      <td>352</td>\n",
       "      <td>495</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Eggs</td>\n",
       "      <td>19</td>\n",
       "      <td>188</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Fats</td>\n",
       "      <td>188</td>\n",
       "      <td>852</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Fish and seafood</td>\n",
       "      <td>1172</td>\n",
       "      <td>1475</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Fruit juices</td>\n",
       "      <td>188</td>\n",
       "      <td>1261</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Fruit nectars</td>\n",
       "      <td>41</td>\n",
       "      <td>51</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Fruits</td>\n",
       "      <td>1597</td>\n",
       "      <td>598</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Ice cream</td>\n",
       "      <td>1760</td>\n",
       "      <td>91</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Meat</td>\n",
       "      <td>1943</td>\n",
       "      <td>832</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Milk and yogurt</td>\n",
       "      <td>2283</td>\n",
       "      <td>1861</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Nuts</td>\n",
       "      <td>176</td>\n",
       "      <td>400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Offals</td>\n",
       "      <td>202</td>\n",
       "      <td>140</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>One-dish meals</td>\n",
       "      <td>8205</td>\n",
       "      <td>2488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Pizza pies and quiche</td>\n",
       "      <td>1264</td>\n",
       "      <td>169</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Plant-based milk substitutes</td>\n",
       "      <td>301</td>\n",
       "      <td>271</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Potatoes</td>\n",
       "      <td>465</td>\n",
       "      <td>205</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Processed meat</td>\n",
       "      <td>5895</td>\n",
       "      <td>823</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Salty and fatty products</td>\n",
       "      <td>2664</td>\n",
       "      <td>957</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Sandwiches</td>\n",
       "      <td>1066</td>\n",
       "      <td>45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Soups</td>\n",
       "      <td>422</td>\n",
       "      <td>425</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Sweetened beverages</td>\n",
       "      <td>3978</td>\n",
       "      <td>408</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Sweets</td>\n",
       "      <td>7021</td>\n",
       "      <td>1022</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Teas and herbal teas and coffees</td>\n",
       "      <td>15</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Unsweetened beverages</td>\n",
       "      <td>456</td>\n",
       "      <td>651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Vegetables</td>\n",
       "      <td>1551</td>\n",
       "      <td>2128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Waters and flavored waters</td>\n",
       "      <td>44</td>\n",
       "      <td>116</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>pastries</td>\n",
       "      <td>1203</td>\n",
       "      <td>180</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>unknown</td>\n",
       "      <td>4350</td>\n",
       "      <td>1791</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "additives_n_classes_k2            Avec additif  Sans additif\n",
       "pnns_groups_2_classes                                       \n",
       "Appetizers                                1807           893\n",
       "Biscuits and cakes                       10203           867\n",
       "Bread                                     1562           760\n",
       "Breakfast cereals                          829           683\n",
       "Cereals                                    842          2663\n",
       "Cheese                                    1738          2512\n",
       "Chocolate products                        2785           813\n",
       "Dairy desserts                            1710           505\n",
       "Dressings and sauces                      3514           842\n",
       "Dried fruits                               352           495\n",
       "Eggs                                        19           188\n",
       "Fats                                       188           852\n",
       "Fish and seafood                          1172          1475\n",
       "Fruit juices                               188          1261\n",
       "Fruit nectars                               41            51\n",
       "Fruits                                    1597           598\n",
       "Ice cream                                 1760            91\n",
       "Meat                                      1943           832\n",
       "Milk and yogurt                           2283          1861\n",
       "Nuts                                       176           400\n",
       "Offals                                     202           140\n",
       "One-dish meals                            8205          2488\n",
       "Pizza pies and quiche                     1264           169\n",
       "Plant-based milk substitutes               301           271\n",
       "Potatoes                                   465           205\n",
       "Processed meat                            5895           823\n",
       "Salty and fatty products                  2664           957\n",
       "Sandwiches                                1066            45\n",
       "Soups                                      422           425\n",
       "Sweetened beverages                       3978           408\n",
       "Sweets                                    7021          1022\n",
       "Teas and herbal teas and coffees            15            28\n",
       "Unsweetened beverages                      456           651\n",
       "Vegetables                                1551          2128\n",
       "Waters and flavored waters                  44           116\n",
       "pastries                                  1203           180\n",
       "unknown                                   4350          1791"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = 'pnns_groups_2_classes'\n",
    "Y = 'additives_n_classes_k2'\n",
    "\n",
    "tableau_contingence = open_food_facts_df[[X, Y]].pivot_table(index=X, \n",
    "                                                             columns=Y, \n",
    "                                                             aggfunc=len)\n",
    "\n",
    "tableau_contingence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(21555.919881480313, 0.0, 36, array([[1910.73537872,  789.26462128],\n",
       "        [7834.01505273, 3235.98494727],\n",
       "        [1643.2324257 ,  678.7675743 ],\n",
       "        [1070.01181208,  441.98818792],\n",
       "        [2480.41759348, 1024.58240652],\n",
       "        [3007.63902205, 1242.36097795],\n",
       "        [2546.23181208, 1051.76818792],\n",
       "        [1567.51069032,  647.48930968],\n",
       "        [3082.65307766, 1273.34692234],\n",
       "        [ 599.4047651 ,  247.5952349 ],\n",
       "        [ 146.48971237,   60.51028763],\n",
       "        [ 735.98696069,  304.01303931],\n",
       "        [1873.22835091,  773.77164909],\n",
       "        [1025.42798658,  423.57201342],\n",
       "        [  65.10653883,   26.89346117],\n",
       "        [1553.35709492,  641.64290508],\n",
       "        [1309.91525407,  541.08474593],\n",
       "        [1963.81136146,  811.18863854],\n",
       "        [2932.62496644, 1211.37503356],\n",
       "        [ 407.62354746,  168.37645254],\n",
       "        [ 242.0264813 ,   99.9735187 ],\n",
       "        [7567.21977948, 3125.78022052],\n",
       "        [1014.10511026,  418.89488974],\n",
       "        [ 404.79282838,  167.20717162],\n",
       "        [ 474.14544583,  195.85455417],\n",
       "        [4754.19269415, 1963.80730585],\n",
       "        [2562.50844679, 1058.49155321],\n",
       "        [ 786.23222435,  324.76777565],\n",
       "        [ 599.4047651 ,  247.5952349 ],\n",
       "        [3103.88347076, 1282.11652924],\n",
       "        [5691.86838926, 2351.13161074],\n",
       "        [  30.43023011,   12.56976989],\n",
       "        [ 783.40150527,  323.59849473],\n",
       "        [2603.55387344, 1075.44612656],\n",
       "        [ 113.22876318,   46.77123682],\n",
       "        [ 978.72112176,  404.27887824],\n",
       "        [4345.86146692, 1795.13853308]]))"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "st.chi2_contingency(tableau_contingence)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Au niveau de risque fixé à 5%, nous rejetons l'hypothèse nulle, par conséquent les variables \"sous-catégories de produit\" et \"présence d'additifs\" sont dépendantes entre elles."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Conclusion de cette hypothèse :** nous avons une dépendance significative entre la sous-catégorie de produits et la présence d'additifs au seuil de 5%. Avec les résultats empiriques on observe un nombre d'additifs moyen et médian plus élevés dans les sous-catégories suivantes :\n",
    "- Glaces\n",
    "- Boissons sucrées édulcorées\n",
    "- Sandwichs\n",
    "- Pâtisseries\n",
    "- Pizzas, tartes et quiches\n",
    "- Biscuits et gâteaux"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Imputation des données"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 104300 entries, 0 to 104299\n",
      "Data columns (total 9 columns):\n",
      " #   Column            Non-Null Count   Dtype  \n",
      "---  ------            --------------   -----  \n",
      " 0   brands_tags       103844 non-null  object \n",
      " 1   labels_en         71866 non-null   object \n",
      " 2   additives_tags    73811 non-null   object \n",
      " 3   nutriscore_grade  104300 non-null  object \n",
      " 4   pnns_groups_1     104159 non-null  object \n",
      " 5   pnns_groups_2     104300 non-null  object \n",
      " 6   additives_n       104300 non-null  float64\n",
      " 7   nutriscore_score  104300 non-null  float64\n",
      " 8   nova_group        104067 non-null  object \n",
      "dtypes: float64(2), object(7)\n",
      "memory usage: 7.2+ MB\n"
     ]
    }
   ],
   "source": [
    "open_food_facts_df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>nombre</th>\n",
       "      <th>total</th>\n",
       "      <th>pourcentage</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>labels_en</td>\n",
       "      <td>32434</td>\n",
       "      <td>104300</td>\n",
       "      <td>31.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>additives_tags</td>\n",
       "      <td>30489</td>\n",
       "      <td>104300</td>\n",
       "      <td>29.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>brands_tags</td>\n",
       "      <td>456</td>\n",
       "      <td>104300</td>\n",
       "      <td>0.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>nova_group</td>\n",
       "      <td>233</td>\n",
       "      <td>104300</td>\n",
       "      <td>0.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>pnns_groups_1</td>\n",
       "      <td>141</td>\n",
       "      <td>104300</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                nombre   total  pourcentage\n",
       "labels_en        32434  104300         31.1\n",
       "additives_tags   30489  104300         29.2\n",
       "brands_tags        456  104300          0.4\n",
       "nova_group         233  104300          0.2\n",
       "pnns_groups_1      141  104300          0.1"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "donnees_manquantes_df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Variable *additives_tags*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Voyons si les données manquantes de cette variable correspondent à un nombre d'additifs à 0. Cela correspondrait donc à une liste vide d'additif."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "filtered_data = open_food_facts_df[open_food_facts_df['additives_tags'].isna()]\n",
    "filtered_data['additives_tags'].fillna('vide', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>additives_n</th>\n",
       "      <th>0.0</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>additives_tags</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>vide</td>\n",
       "      <td>30489</td>\n",
       "      <td>30489</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>All</td>\n",
       "      <td>30489</td>\n",
       "      <td>30489</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "additives_n       0.0    All\n",
       "additives_tags              \n",
       "vide            30489  30489\n",
       "All             30489  30489"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = 'additives_tags'\n",
    "Y = 'additives_n'\n",
    "\n",
    "tableau_contingence = filtered_data[[X, Y]].pivot_table(index=X, \n",
    "                                                        columns=Y, \n",
    "                                                        aggfunc=len,\n",
    "                                                        margins=True)\n",
    "\n",
    "tableau_contingence"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Toutes les données manquantes correspondent en effet à un nombre nul d'additifs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count                73811\n",
       "unique               19093\n",
       "top       en:e322,en:e322i\n",
       "freq                  3834\n",
       "Name: additives_tags, dtype: object"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df['additives_tags'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "en:e322,en:e322i                                                                                            3834\n",
       "en:e330                                                                                                     2802\n",
       "en:e440                                                                                                     2540\n",
       "en:e14xx                                                                                                    1969\n",
       "en:e300                                                                                                     1870\n",
       "                                                                                                            ... \n",
       "en:e150c,en:e160c,en:e163,en:e330                                                                              1\n",
       "en:e1103,en:e163,en:e270,en:e330,en:e420                                                                       1\n",
       "en:e150a,en:e250,en:e262,en:e316,en:e330,en:e331,en:e450,en:e451,en:e621                                       1\n",
       "en:e160a,en:e160ai,en:e202,en:e211,en:e330,en:e385,en:e412,en:e415                                             1\n",
       "en:e282,en:e322,en:e322i,en:e330,en:e401,en:e407,en:e412,en:e414,en:e450,en:e471,en:e475,en:e500,en:e904       1\n",
       "Name: additives_tags, Length: 19093, dtype: int64"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df['additives_tags'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    73811.0\n",
       "mean        -1.0\n",
       "std          0.0\n",
       "min         -1.0\n",
       "25%         -1.0\n",
       "50%         -1.0\n",
       "75%         -1.0\n",
       "max         -1.0\n",
       "Name: additives_tags, dtype: float64"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df['additives_tags'].str.find(\"no\").describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "On impute donc les valeurs manquantes de cette variable par 'no additives'."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "open_food_facts_df['additives_tags'].fillna('no additives', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "no additives                                                           30489\n",
       "en:e322,en:e322i                                                        3834\n",
       "en:e330                                                                 2802\n",
       "en:e440                                                                 2540\n",
       "en:e14xx                                                                1969\n",
       "                                                                       ...  \n",
       "en:e160a,en:e160ai,en:e202,en:e306,en:e322,en:e322i,en:e330,en:e471        1\n",
       "en:e120,en:e127,en:e133,en:e150,en:e202,en:e220,en:e330,en:e415            1\n",
       "en:e1510,en:e202,en:e415,en:e420,en:e450,en:e450i,en:e471,en:e500          1\n",
       "en:e250,en:e282,en:e301                                                    1\n",
       "en:e322,en:e322i,en:e471,en:e472e,en:e481                                  1\n",
       "Name: additives_tags, Length: 19094, dtype: int64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df['additives_tags'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 104300 entries, 0 to 104299\n",
      "Data columns (total 9 columns):\n",
      "brands_tags         103844 non-null object\n",
      "labels_en           71866 non-null object\n",
      "additives_tags      104300 non-null object\n",
      "nutriscore_grade    104300 non-null object\n",
      "pnns_groups_1       104159 non-null object\n",
      "pnns_groups_2       104300 non-null object\n",
      "additives_n         104300 non-null float64\n",
      "nutriscore_score    104300 non-null float64\n",
      "nova_group          104067 non-null object\n",
      "dtypes: float64(2), object(7)\n",
      "memory usage: 7.2+ MB\n"
     ]
    }
   ],
   "source": [
    "open_food_facts_df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Variable *pnns_groups_1*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Sugary snacks              24094\n",
       "Composite foods            13098\n",
       "Fish Meat Eggs             12689\n",
       "Milk and dairy products    12460\n",
       "Cereals and potatoes        8725\n",
       "Beverages                   7807\n",
       "Salty snacks                6897\n",
       "Fruits and vegetables       6852\n",
       "unknown                     6141\n",
       "Fat and sauces              5396\n",
       "Name: pnns_groups_1, dtype: int64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df['pnns_groups_1'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "filtered_data = open_food_facts_df[open_food_facts_df['pnns_groups_1'].isna()]\n",
    "filtered_data['pnns_groups_1'].fillna('vide', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>pnns_groups_2</th>\n",
       "      <th>Alcoholic beverages</th>\n",
       "      <th>Pizza pies and quiche</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pnns_groups_1</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>vide</td>\n",
       "      <td>2</td>\n",
       "      <td>139</td>\n",
       "      <td>141</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>All</td>\n",
       "      <td>2</td>\n",
       "      <td>139</td>\n",
       "      <td>141</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "pnns_groups_2  Alcoholic beverages  Pizza pies and quiche  All\n",
       "pnns_groups_1                                                 \n",
       "vide                             2                    139  141\n",
       "All                              2                    139  141"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = 'pnns_groups_1'\n",
    "Y = 'pnns_groups_2'\n",
    "\n",
    "tableau_contingence = filtered_data[[X, Y]].pivot_table(index=X, \n",
    "                                                        columns=Y, \n",
    "                                                        aggfunc=len,\n",
    "                                                        margins=True)\n",
    "\n",
    "tableau_contingence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "filtered_data = open_food_facts_df[open_food_facts_df['pnns_groups_2'] == 'Pizza pies and quiche']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>pnns_groups_2</th>\n",
       "      <th>Pizza pies and quiche</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pnns_groups_1</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>Composite foods</td>\n",
       "      <td>1294</td>\n",
       "      <td>1294</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>All</td>\n",
       "      <td>1294</td>\n",
       "      <td>1294</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "pnns_groups_2    Pizza pies and quiche   All\n",
       "pnns_groups_1                               \n",
       "Composite foods                   1294  1294\n",
       "All                               1294  1294"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = 'pnns_groups_1'\n",
    "Y = 'pnns_groups_2'\n",
    "\n",
    "tableau_contingence = filtered_data[[X, Y]].pivot_table(index=X, \n",
    "                                                        columns=Y, \n",
    "                                                        aggfunc=len,\n",
    "                                                        margins=True)\n",
    "\n",
    "tableau_contingence"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Les catégories vides correspondant à la sous-catégorie 'Pizza pies and quiche' seront remplacées par 'Composite foods'. Et celles parmi les boissons alcoolisées seront catégorisées dans 'Beverages'."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "open_food_facts_df.loc[open_food_facts_df['pnns_groups_2'] == 'Pizza pies and quiche',\n",
    "                      'pnns_groups_1'] = 'Composite foods'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count                1433\n",
       "unique                  1\n",
       "top       Composite foods\n",
       "freq                 1433\n",
       "Name: pnns_groups_1, dtype: object"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df[open_food_facts_df['pnns_groups_2'] == 'Pizza pies and quiche']['pnns_groups_1'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "open_food_facts_df.loc[open_food_facts_df['pnns_groups_2'] == 'Alcoholic beverages',\n",
    "                      'pnns_groups_1'] = 'Beverages'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count             2\n",
       "unique            1\n",
       "top       Beverages\n",
       "freq              2\n",
       "Name: pnns_groups_1, dtype: object"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df[open_food_facts_df['pnns_groups_2'] == 'Alcoholic beverages']['pnns_groups_1'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 104300 entries, 0 to 104299\n",
      "Data columns (total 12 columns):\n",
      " #   Column                  Non-Null Count   Dtype  \n",
      "---  ------                  --------------   -----  \n",
      " 0   brands_tags             103844 non-null  object \n",
      " 1   labels_en               71866 non-null   object \n",
      " 2   additives_tags          104300 non-null  object \n",
      " 3   nutriscore_grade        104300 non-null  object \n",
      " 4   pnns_groups_1           104300 non-null  object \n",
      " 5   pnns_groups_2           104300 non-null  object \n",
      " 6   additives_n             104300 non-null  float64\n",
      " 7   nutriscore_score        104300 non-null  float64\n",
      " 8   nova_group              104067 non-null  object \n",
      " 9   additives_n_classes     104300 non-null  object \n",
      " 10  nova_group_classes      104067 non-null  object \n",
      " 11  additives_n_classes_k2  104300 non-null  object \n",
      "dtypes: float64(2), object(10)\n",
      "memory usage: 9.5+ MB\n"
     ]
    }
   ],
   "source": [
    "open_food_facts_df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Variable *nova_group*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1e imputation avec KNNImputer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "imputation_knn_df = open_food_facts_df[['nutriscore_grade', 'pnns_groups_1', 'additives_n', 'nova_group']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>nutriscore_grade</th>\n",
       "      <th>pnns_groups_1</th>\n",
       "      <th>additives_n</th>\n",
       "      <th>nova_group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>a</td>\n",
       "      <td>Cereals and potatoes</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>b</td>\n",
       "      <td>Composite foods</td>\n",
       "      <td>2.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>b</td>\n",
       "      <td>Milk and dairy products</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "      <td>Fruits and vegetables</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>d</td>\n",
       "      <td>Sugary snacks</td>\n",
       "      <td>7.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104295</td>\n",
       "      <td>d</td>\n",
       "      <td>unknown</td>\n",
       "      <td>7.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104296</td>\n",
       "      <td>d</td>\n",
       "      <td>Milk and dairy products</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104297</td>\n",
       "      <td>e</td>\n",
       "      <td>Beverages</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104298</td>\n",
       "      <td>e</td>\n",
       "      <td>Beverages</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104299</td>\n",
       "      <td>a</td>\n",
       "      <td>Fruits and vegetables</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>104300 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       nutriscore_grade            pnns_groups_1  additives_n nova_group\n",
       "0                     a     Cereals and potatoes          3.0          4\n",
       "1                     b          Composite foods          2.0          4\n",
       "2                     b  Milk and dairy products          4.0          4\n",
       "3                     b    Fruits and vegetables          1.0          4\n",
       "4                     d            Sugary snacks          7.0          4\n",
       "...                 ...                      ...          ...        ...\n",
       "104295                d                  unknown          7.0          4\n",
       "104296                d  Milk and dairy products          0.0          3\n",
       "104297                e                Beverages          2.0          3\n",
       "104298                e                Beverages          0.0          3\n",
       "104299                a    Fruits and vegetables          1.0          3\n",
       "\n",
       "[104300 rows x 4 columns]"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "imputation_knn"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Nous allons créer des flags pour chaque modalité de la variable catégorie. Et remplacer le classement Nutri-score par des chiffres, a = 1 et e = 5."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "d    30566\n",
       "c    23016\n",
       "e    22130\n",
       "a    14474\n",
       "b    14114\n",
       "Name: nutriscore_grade, dtype: int64"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "imputation_knn_df['nutriscore_grade'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/sebastien/anaconda3/lib/python3.7/site-packages/pandas/core/indexing.py:494: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  self.obj[item] = s\n"
     ]
    }
   ],
   "source": [
    "imputation_knn_df.loc[imputation_knn_df['nutriscore_grade'] == 'a', 'nutriscore_grade'] = 1\n",
    "imputation_knn_df.loc[imputation_knn_df['nutriscore_grade'] == 'b', 'nutriscore_grade'] = 2\n",
    "imputation_knn_df.loc[imputation_knn_df['nutriscore_grade'] == 'c', 'nutriscore_grade'] = 3\n",
    "imputation_knn_df.loc[imputation_knn_df['nutriscore_grade'] == 'd', 'nutriscore_grade'] = 4\n",
    "imputation_knn_df.loc[imputation_knn_df['nutriscore_grade'] == 'e', 'nutriscore_grade'] = 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4    30566\n",
       "3    23016\n",
       "5    22130\n",
       "1    14474\n",
       "2    14114\n",
       "Name: nutriscore_grade, dtype: int64"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "imputation_knn_df['nutriscore_grade'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>pnns_groups_1_Cereals and potatoes</th>\n",
       "      <th>pnns_groups_1_Composite foods</th>\n",
       "      <th>pnns_groups_1_Fat and sauces</th>\n",
       "      <th>pnns_groups_1_Fish Meat Eggs</th>\n",
       "      <th>pnns_groups_1_Fruits and vegetables</th>\n",
       "      <th>pnns_groups_1_Milk and dairy products</th>\n",
       "      <th>pnns_groups_1_Salty snacks</th>\n",
       "      <th>pnns_groups_1_Sugary snacks</th>\n",
       "      <th>pnns_groups_1_unknown</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   pnns_groups_1_Cereals and potatoes  pnns_groups_1_Composite foods  \\\n",
       "0                                   1                              0   \n",
       "1                                   0                              1   \n",
       "2                                   0                              0   \n",
       "3                                   0                              0   \n",
       "4                                   0                              0   \n",
       "\n",
       "   pnns_groups_1_Fat and sauces  pnns_groups_1_Fish Meat Eggs  \\\n",
       "0                             0                             0   \n",
       "1                             0                             0   \n",
       "2                             0                             0   \n",
       "3                             0                             0   \n",
       "4                             0                             0   \n",
       "\n",
       "   pnns_groups_1_Fruits and vegetables  pnns_groups_1_Milk and dairy products  \\\n",
       "0                                    0                                      0   \n",
       "1                                    0                                      0   \n",
       "2                                    0                                      1   \n",
       "3                                    1                                      0   \n",
       "4                                    0                                      0   \n",
       "\n",
       "   pnns_groups_1_Salty snacks  pnns_groups_1_Sugary snacks  \\\n",
       "0                           0                            0   \n",
       "1                           0                            0   \n",
       "2                           0                            0   \n",
       "3                           0                            0   \n",
       "4                           0                            1   \n",
       "\n",
       "   pnns_groups_1_unknown  \n",
       "0                      0  \n",
       "1                      0  \n",
       "2                      0  \n",
       "3                      0  \n",
       "4                      0  "
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cat_variables = imputation_knn_df[['pnns_groups_1']]\n",
    "cat_flags = pd.get_dummies(cat_variables, drop_first=True)\n",
    "cat_flags.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>nutriscore_grade</th>\n",
       "      <th>additives_n</th>\n",
       "      <th>nova_group</th>\n",
       "      <th>pnns_groups_1_Cereals and potatoes</th>\n",
       "      <th>pnns_groups_1_Composite foods</th>\n",
       "      <th>pnns_groups_1_Fat and sauces</th>\n",
       "      <th>pnns_groups_1_Fish Meat Eggs</th>\n",
       "      <th>pnns_groups_1_Fruits and vegetables</th>\n",
       "      <th>pnns_groups_1_Milk and dairy products</th>\n",
       "      <th>pnns_groups_1_Salty snacks</th>\n",
       "      <th>pnns_groups_1_Sugary snacks</th>\n",
       "      <th>pnns_groups_1_unknown</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2.0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>7.0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   nutriscore_grade  additives_n nova_group  \\\n",
       "0                 1          3.0          4   \n",
       "1                 2          2.0          4   \n",
       "2                 2          4.0          4   \n",
       "3                 2          1.0          4   \n",
       "4                 4          7.0          4   \n",
       "\n",
       "   pnns_groups_1_Cereals and potatoes  pnns_groups_1_Composite foods  \\\n",
       "0                                   1                              0   \n",
       "1                                   0                              1   \n",
       "2                                   0                              0   \n",
       "3                                   0                              0   \n",
       "4                                   0                              0   \n",
       "\n",
       "   pnns_groups_1_Fat and sauces  pnns_groups_1_Fish Meat Eggs  \\\n",
       "0                             0                             0   \n",
       "1                             0                             0   \n",
       "2                             0                             0   \n",
       "3                             0                             0   \n",
       "4                             0                             0   \n",
       "\n",
       "   pnns_groups_1_Fruits and vegetables  pnns_groups_1_Milk and dairy products  \\\n",
       "0                                    0                                      0   \n",
       "1                                    0                                      0   \n",
       "2                                    0                                      1   \n",
       "3                                    1                                      0   \n",
       "4                                    0                                      0   \n",
       "\n",
       "   pnns_groups_1_Salty snacks  pnns_groups_1_Sugary snacks  \\\n",
       "0                           0                            0   \n",
       "1                           0                            0   \n",
       "2                           0                            0   \n",
       "3                           0                            0   \n",
       "4                           0                            1   \n",
       "\n",
       "   pnns_groups_1_unknown  \n",
       "0                      0  \n",
       "1                      0  \n",
       "2                      0  \n",
       "3                      0  \n",
       "4                      0  "
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "imputation_knn_df = imputation_knn_df.drop(['pnns_groups_1'], axis=1)\n",
    "imputation_knn_df = pd.concat([imputation_knn_df, cat_flags], axis=1)\n",
    "imputation_knn_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>nutriscore_grade</th>\n",
       "      <th>additives_n</th>\n",
       "      <th>nova_group</th>\n",
       "      <th>pnns_groups_1_Cereals and potatoes</th>\n",
       "      <th>pnns_groups_1_Composite foods</th>\n",
       "      <th>pnns_groups_1_Fat and sauces</th>\n",
       "      <th>pnns_groups_1_Fish Meat Eggs</th>\n",
       "      <th>pnns_groups_1_Fruits and vegetables</th>\n",
       "      <th>pnns_groups_1_Milk and dairy products</th>\n",
       "      <th>pnns_groups_1_Salty snacks</th>\n",
       "      <th>pnns_groups_1_Sugary snacks</th>\n",
       "      <th>pnns_groups_1_unknown</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.25</td>\n",
       "      <td>0.066667</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.25</td>\n",
       "      <td>0.133333</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.25</td>\n",
       "      <td>0.033333</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.233333</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   nutriscore_grade  additives_n  nova_group  \\\n",
       "0              0.00     0.100000         1.0   \n",
       "1              0.25     0.066667         1.0   \n",
       "2              0.25     0.133333         1.0   \n",
       "3              0.25     0.033333         1.0   \n",
       "4              0.75     0.233333         1.0   \n",
       "\n",
       "   pnns_groups_1_Cereals and potatoes  pnns_groups_1_Composite foods  \\\n",
       "0                                 1.0                            0.0   \n",
       "1                                 0.0                            1.0   \n",
       "2                                 0.0                            0.0   \n",
       "3                                 0.0                            0.0   \n",
       "4                                 0.0                            0.0   \n",
       "\n",
       "   pnns_groups_1_Fat and sauces  pnns_groups_1_Fish Meat Eggs  \\\n",
       "0                           0.0                           0.0   \n",
       "1                           0.0                           0.0   \n",
       "2                           0.0                           0.0   \n",
       "3                           0.0                           0.0   \n",
       "4                           0.0                           0.0   \n",
       "\n",
       "   pnns_groups_1_Fruits and vegetables  pnns_groups_1_Milk and dairy products  \\\n",
       "0                                  0.0                                    0.0   \n",
       "1                                  0.0                                    0.0   \n",
       "2                                  0.0                                    1.0   \n",
       "3                                  1.0                                    0.0   \n",
       "4                                  0.0                                    0.0   \n",
       "\n",
       "   pnns_groups_1_Salty snacks  pnns_groups_1_Sugary snacks  \\\n",
       "0                         0.0                          0.0   \n",
       "1                         0.0                          0.0   \n",
       "2                         0.0                          0.0   \n",
       "3                         0.0                          0.0   \n",
       "4                         0.0                          1.0   \n",
       "\n",
       "   pnns_groups_1_unknown  \n",
       "0                    0.0  \n",
       "1                    0.0  \n",
       "2                    0.0  \n",
       "3                    0.0  \n",
       "4                    0.0  "
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# normalisation des données pour l'algorithme car il se base sur les distances\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "\n",
    "scaler = MinMaxScaler()\n",
    "imputation_knn_df = pd.DataFrame(scaler.fit_transform(imputation_knn_df), columns = imputation_knn_df.columns)\n",
    "imputation_knn_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    104067.000000\n",
       "mean          0.837624\n",
       "std           0.291813\n",
       "min           0.000000\n",
       "25%           0.666667\n",
       "50%           1.000000\n",
       "75%           1.000000\n",
       "max           1.000000\n",
       "Name: nova_group, dtype: float64"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "imputation_knn_df['nova_group'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>nutriscore_grade</th>\n",
       "      <th>additives_n</th>\n",
       "      <th>nova_group</th>\n",
       "      <th>pnns_groups_1_Cereals and potatoes</th>\n",
       "      <th>pnns_groups_1_Composite foods</th>\n",
       "      <th>pnns_groups_1_Fat and sauces</th>\n",
       "      <th>pnns_groups_1_Fish Meat Eggs</th>\n",
       "      <th>pnns_groups_1_Fruits and vegetables</th>\n",
       "      <th>pnns_groups_1_Milk and dairy products</th>\n",
       "      <th>pnns_groups_1_Salty snacks</th>\n",
       "      <th>pnns_groups_1_Sugary snacks</th>\n",
       "      <th>pnns_groups_1_unknown</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>92</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.033333</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>197</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>261</td>\n",
       "      <td>0.25</td>\n",
       "      <td>0.033333</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>578</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>649</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.033333</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>868</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.033333</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1040</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.066667</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1261</td>\n",
       "      <td>0.25</td>\n",
       "      <td>0.033333</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1343</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.066667</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1358</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.033333</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1497</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1725</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.033333</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1775</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.066667</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1776</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.033333</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1813</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.033333</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      nutriscore_grade  additives_n  nova_group  \\\n",
       "92                1.00     0.033333         NaN   \n",
       "197               1.00     0.166667         NaN   \n",
       "261               0.25     0.033333         NaN   \n",
       "578               0.75     0.100000         NaN   \n",
       "649               1.00     0.033333         NaN   \n",
       "868               0.00     0.033333         NaN   \n",
       "1040              0.75     0.066667         NaN   \n",
       "1261              0.25     0.033333         NaN   \n",
       "1343              0.50     0.066667         NaN   \n",
       "1358              0.75     0.033333         NaN   \n",
       "1497              1.00     0.100000         NaN   \n",
       "1725              0.50     0.033333         NaN   \n",
       "1775              1.00     0.066667         NaN   \n",
       "1776              0.50     0.033333         NaN   \n",
       "1813              1.00     0.033333         NaN   \n",
       "\n",
       "      pnns_groups_1_Cereals and potatoes  pnns_groups_1_Composite foods  \\\n",
       "92                                   0.0                            0.0   \n",
       "197                                  0.0                            0.0   \n",
       "261                                  0.0                            0.0   \n",
       "578                                  0.0                            0.0   \n",
       "649                                  0.0                            0.0   \n",
       "868                                  0.0                            0.0   \n",
       "1040                                 0.0                            0.0   \n",
       "1261                                 1.0                            0.0   \n",
       "1343                                 0.0                            0.0   \n",
       "1358                                 0.0                            0.0   \n",
       "1497                                 0.0                            0.0   \n",
       "1725                                 0.0                            0.0   \n",
       "1775                                 0.0                            0.0   \n",
       "1776                                 0.0                            0.0   \n",
       "1813                                 0.0                            0.0   \n",
       "\n",
       "      pnns_groups_1_Fat and sauces  pnns_groups_1_Fish Meat Eggs  \\\n",
       "92                             0.0                           0.0   \n",
       "197                            0.0                           0.0   \n",
       "261                            0.0                           0.0   \n",
       "578                            1.0                           0.0   \n",
       "649                            1.0                           0.0   \n",
       "868                            0.0                           0.0   \n",
       "1040                           0.0                           0.0   \n",
       "1261                           0.0                           0.0   \n",
       "1343                           0.0                           1.0   \n",
       "1358                           0.0                           0.0   \n",
       "1497                           0.0                           0.0   \n",
       "1725                           0.0                           0.0   \n",
       "1775                           0.0                           0.0   \n",
       "1776                           0.0                           0.0   \n",
       "1813                           0.0                           0.0   \n",
       "\n",
       "      pnns_groups_1_Fruits and vegetables  \\\n",
       "92                                    0.0   \n",
       "197                                   0.0   \n",
       "261                                   0.0   \n",
       "578                                   0.0   \n",
       "649                                   0.0   \n",
       "868                                   1.0   \n",
       "1040                                  0.0   \n",
       "1261                                  0.0   \n",
       "1343                                  0.0   \n",
       "1358                                  0.0   \n",
       "1497                                  0.0   \n",
       "1725                                  0.0   \n",
       "1775                                  0.0   \n",
       "1776                                  0.0   \n",
       "1813                                  0.0   \n",
       "\n",
       "      pnns_groups_1_Milk and dairy products  pnns_groups_1_Salty snacks  \\\n",
       "92                                      0.0                         0.0   \n",
       "197                                     0.0                         0.0   \n",
       "261                                     0.0                         0.0   \n",
       "578                                     0.0                         0.0   \n",
       "649                                     0.0                         0.0   \n",
       "868                                     0.0                         0.0   \n",
       "1040                                    0.0                         0.0   \n",
       "1261                                    0.0                         0.0   \n",
       "1343                                    0.0                         0.0   \n",
       "1358                                    0.0                         0.0   \n",
       "1497                                    0.0                         0.0   \n",
       "1725                                    0.0                         0.0   \n",
       "1775                                    0.0                         0.0   \n",
       "1776                                    0.0                         1.0   \n",
       "1813                                    0.0                         0.0   \n",
       "\n",
       "      pnns_groups_1_Sugary snacks  pnns_groups_1_unknown  \n",
       "92                            1.0                    0.0  \n",
       "197                           0.0                    0.0  \n",
       "261                           0.0                    0.0  \n",
       "578                           0.0                    0.0  \n",
       "649                           0.0                    0.0  \n",
       "868                           0.0                    0.0  \n",
       "1040                          1.0                    0.0  \n",
       "1261                          0.0                    0.0  \n",
       "1343                          0.0                    0.0  \n",
       "1358                          1.0                    0.0  \n",
       "1497                          1.0                    0.0  \n",
       "1725                          1.0                    0.0  \n",
       "1775                          0.0                    0.0  \n",
       "1776                          0.0                    0.0  \n",
       "1813                          0.0                    0.0  "
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "imputation_knn_df[imputation_knn_df['nova_group'].isna()].head(15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Int64Index([    92,    197,    261,    578,    649,    868,   1040,   1261,\n",
       "              1343,   1358,\n",
       "            ...\n",
       "            101007, 101093, 101782, 103087, 103256, 103851, 103938, 104120,\n",
       "            104189, 104211],\n",
       "           dtype='int64', length=233)"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "index_donnees_imputees = imputation_knn_df[imputation_knn_df['nova_group'].isna()].index\n",
    "index_donnees_imputees"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "# imputation des données de Nova group avec l'algo k-NN avec 5 voisins\n",
    "from sklearn.impute import KNNImputer\n",
    "\n",
    "imputer = KNNImputer(n_neighbors=5)\n",
    "imputation_knn_df = pd.DataFrame(imputer.fit_transform(imputation_knn_df), columns = imputation_knn_df.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    104300.000000\n",
       "mean          0.837249\n",
       "std           0.291951\n",
       "min           0.000000\n",
       "25%           0.666667\n",
       "50%           1.000000\n",
       "75%           1.000000\n",
       "max           1.000000\n",
       "Name: nova_group, dtype: float64"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "imputation_knn_df['nova_group'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>nutriscore_grade</th>\n",
       "      <th>additives_n</th>\n",
       "      <th>nova_group</th>\n",
       "      <th>pnns_groups_1_Cereals and potatoes</th>\n",
       "      <th>pnns_groups_1_Composite foods</th>\n",
       "      <th>pnns_groups_1_Fat and sauces</th>\n",
       "      <th>pnns_groups_1_Fish Meat Eggs</th>\n",
       "      <th>pnns_groups_1_Fruits and vegetables</th>\n",
       "      <th>pnns_groups_1_Milk and dairy products</th>\n",
       "      <th>pnns_groups_1_Salty snacks</th>\n",
       "      <th>pnns_groups_1_Sugary snacks</th>\n",
       "      <th>pnns_groups_1_unknown</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>92</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.033333</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>197</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>261</td>\n",
       "      <td>0.25</td>\n",
       "      <td>0.033333</td>\n",
       "      <td>0.733333</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>578</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>649</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.033333</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>868</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.033333</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1040</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.066667</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1261</td>\n",
       "      <td>0.25</td>\n",
       "      <td>0.033333</td>\n",
       "      <td>0.533333</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1343</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.066667</td>\n",
       "      <td>0.933333</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1358</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.033333</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1497</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1725</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.033333</td>\n",
       "      <td>0.266667</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1775</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.066667</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1776</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.033333</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1813</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.033333</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      nutriscore_grade  additives_n  nova_group  \\\n",
       "92                1.00     0.033333    1.000000   \n",
       "197               1.00     0.166667    1.000000   \n",
       "261               0.25     0.033333    0.733333   \n",
       "578               0.75     0.100000    1.000000   \n",
       "649               1.00     0.033333    1.000000   \n",
       "868               0.00     0.033333    0.000000   \n",
       "1040              0.75     0.066667    1.000000   \n",
       "1261              0.25     0.033333    0.533333   \n",
       "1343              0.50     0.066667    0.933333   \n",
       "1358              0.75     0.033333    0.666667   \n",
       "1497              1.00     0.100000    1.000000   \n",
       "1725              0.50     0.033333    0.266667   \n",
       "1775              1.00     0.066667    0.666667   \n",
       "1776              0.50     0.033333    0.000000   \n",
       "1813              1.00     0.033333    0.666667   \n",
       "\n",
       "      pnns_groups_1_Cereals and potatoes  pnns_groups_1_Composite foods  \\\n",
       "92                                   0.0                            0.0   \n",
       "197                                  0.0                            0.0   \n",
       "261                                  0.0                            0.0   \n",
       "578                                  0.0                            0.0   \n",
       "649                                  0.0                            0.0   \n",
       "868                                  0.0                            0.0   \n",
       "1040                                 0.0                            0.0   \n",
       "1261                                 1.0                            0.0   \n",
       "1343                                 0.0                            0.0   \n",
       "1358                                 0.0                            0.0   \n",
       "1497                                 0.0                            0.0   \n",
       "1725                                 0.0                            0.0   \n",
       "1775                                 0.0                            0.0   \n",
       "1776                                 0.0                            0.0   \n",
       "1813                                 0.0                            0.0   \n",
       "\n",
       "      pnns_groups_1_Fat and sauces  pnns_groups_1_Fish Meat Eggs  \\\n",
       "92                             0.0                           0.0   \n",
       "197                            0.0                           0.0   \n",
       "261                            0.0                           0.0   \n",
       "578                            1.0                           0.0   \n",
       "649                            1.0                           0.0   \n",
       "868                            0.0                           0.0   \n",
       "1040                           0.0                           0.0   \n",
       "1261                           0.0                           0.0   \n",
       "1343                           0.0                           1.0   \n",
       "1358                           0.0                           0.0   \n",
       "1497                           0.0                           0.0   \n",
       "1725                           0.0                           0.0   \n",
       "1775                           0.0                           0.0   \n",
       "1776                           0.0                           0.0   \n",
       "1813                           0.0                           0.0   \n",
       "\n",
       "      pnns_groups_1_Fruits and vegetables  \\\n",
       "92                                    0.0   \n",
       "197                                   0.0   \n",
       "261                                   0.0   \n",
       "578                                   0.0   \n",
       "649                                   0.0   \n",
       "868                                   1.0   \n",
       "1040                                  0.0   \n",
       "1261                                  0.0   \n",
       "1343                                  0.0   \n",
       "1358                                  0.0   \n",
       "1497                                  0.0   \n",
       "1725                                  0.0   \n",
       "1775                                  0.0   \n",
       "1776                                  0.0   \n",
       "1813                                  0.0   \n",
       "\n",
       "      pnns_groups_1_Milk and dairy products  pnns_groups_1_Salty snacks  \\\n",
       "92                                      0.0                         0.0   \n",
       "197                                     0.0                         0.0   \n",
       "261                                     0.0                         0.0   \n",
       "578                                     0.0                         0.0   \n",
       "649                                     0.0                         0.0   \n",
       "868                                     0.0                         0.0   \n",
       "1040                                    0.0                         0.0   \n",
       "1261                                    0.0                         0.0   \n",
       "1343                                    0.0                         0.0   \n",
       "1358                                    0.0                         0.0   \n",
       "1497                                    0.0                         0.0   \n",
       "1725                                    0.0                         0.0   \n",
       "1775                                    0.0                         0.0   \n",
       "1776                                    0.0                         1.0   \n",
       "1813                                    0.0                         0.0   \n",
       "\n",
       "      pnns_groups_1_Sugary snacks  pnns_groups_1_unknown  \n",
       "92                            1.0                    0.0  \n",
       "197                           0.0                    0.0  \n",
       "261                           0.0                    0.0  \n",
       "578                           0.0                    0.0  \n",
       "649                           0.0                    0.0  \n",
       "868                           0.0                    0.0  \n",
       "1040                          1.0                    0.0  \n",
       "1261                          0.0                    0.0  \n",
       "1343                          0.0                    0.0  \n",
       "1358                          1.0                    0.0  \n",
       "1497                          1.0                    0.0  \n",
       "1725                          1.0                    0.0  \n",
       "1775                          0.0                    0.0  \n",
       "1776                          0.0                    0.0  \n",
       "1813                          0.0                    0.0  "
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "imputation_knn_df.iloc[index_donnees_imputees].head(15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>nutriscore_grade</th>\n",
       "      <th>additives_n</th>\n",
       "      <th>nova_group</th>\n",
       "      <th>pnns_groups_1_Cereals and potatoes</th>\n",
       "      <th>pnns_groups_1_Composite foods</th>\n",
       "      <th>pnns_groups_1_Fat and sauces</th>\n",
       "      <th>pnns_groups_1_Fish Meat Eggs</th>\n",
       "      <th>pnns_groups_1_Fruits and vegetables</th>\n",
       "      <th>pnns_groups_1_Milk and dairy products</th>\n",
       "      <th>pnns_groups_1_Salty snacks</th>\n",
       "      <th>pnns_groups_1_Sugary snacks</th>\n",
       "      <th>pnns_groups_1_unknown</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>2.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>4.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104295</td>\n",
       "      <td>4.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104296</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104297</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104298</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104299</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>104300 rows × 12 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        nutriscore_grade  additives_n  nova_group  \\\n",
       "0                    1.0          3.0         4.0   \n",
       "1                    2.0          2.0         4.0   \n",
       "2                    2.0          4.0         4.0   \n",
       "3                    2.0          1.0         4.0   \n",
       "4                    4.0          7.0         4.0   \n",
       "...                  ...          ...         ...   \n",
       "104295               4.0          7.0         4.0   \n",
       "104296               4.0          0.0         3.0   \n",
       "104297               5.0          2.0         3.0   \n",
       "104298               5.0          0.0         3.0   \n",
       "104299               1.0          1.0         3.0   \n",
       "\n",
       "        pnns_groups_1_Cereals and potatoes  pnns_groups_1_Composite foods  \\\n",
       "0                                      1.0                            0.0   \n",
       "1                                      0.0                            1.0   \n",
       "2                                      0.0                            0.0   \n",
       "3                                      0.0                            0.0   \n",
       "4                                      0.0                            0.0   \n",
       "...                                    ...                            ...   \n",
       "104295                                 0.0                            0.0   \n",
       "104296                                 0.0                            0.0   \n",
       "104297                                 0.0                            0.0   \n",
       "104298                                 0.0                            0.0   \n",
       "104299                                 0.0                            0.0   \n",
       "\n",
       "        pnns_groups_1_Fat and sauces  pnns_groups_1_Fish Meat Eggs  \\\n",
       "0                                0.0                           0.0   \n",
       "1                                0.0                           0.0   \n",
       "2                                0.0                           0.0   \n",
       "3                                0.0                           0.0   \n",
       "4                                0.0                           0.0   \n",
       "...                              ...                           ...   \n",
       "104295                           0.0                           0.0   \n",
       "104296                           0.0                           0.0   \n",
       "104297                           0.0                           0.0   \n",
       "104298                           0.0                           0.0   \n",
       "104299                           0.0                           0.0   \n",
       "\n",
       "        pnns_groups_1_Fruits and vegetables  \\\n",
       "0                                       0.0   \n",
       "1                                       0.0   \n",
       "2                                       0.0   \n",
       "3                                       1.0   \n",
       "4                                       0.0   \n",
       "...                                     ...   \n",
       "104295                                  0.0   \n",
       "104296                                  0.0   \n",
       "104297                                  0.0   \n",
       "104298                                  0.0   \n",
       "104299                                  1.0   \n",
       "\n",
       "        pnns_groups_1_Milk and dairy products  pnns_groups_1_Salty snacks  \\\n",
       "0                                         0.0                         0.0   \n",
       "1                                         0.0                         0.0   \n",
       "2                                         1.0                         0.0   \n",
       "3                                         0.0                         0.0   \n",
       "4                                         0.0                         0.0   \n",
       "...                                       ...                         ...   \n",
       "104295                                    0.0                         0.0   \n",
       "104296                                    1.0                         0.0   \n",
       "104297                                    0.0                         0.0   \n",
       "104298                                    0.0                         0.0   \n",
       "104299                                    0.0                         0.0   \n",
       "\n",
       "        pnns_groups_1_Sugary snacks  pnns_groups_1_unknown  \n",
       "0                               0.0                    0.0  \n",
       "1                               0.0                    0.0  \n",
       "2                               0.0                    0.0  \n",
       "3                               0.0                    0.0  \n",
       "4                               1.0                    0.0  \n",
       "...                             ...                    ...  \n",
       "104295                          0.0                    1.0  \n",
       "104296                          0.0                    0.0  \n",
       "104297                          0.0                    0.0  \n",
       "104298                          0.0                    0.0  \n",
       "104299                          0.0                    0.0  \n",
       "\n",
       "[104300 rows x 12 columns]"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "denormalized_df = pd.DataFrame(scaler.inverse_transform(imputation_knn_df), columns = imputation_knn_df.columns)\n",
    "denormalized_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [],
   "source": [
    "open_food_facts_df['nova_group'] = denormalized_df['nova_group']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     4.0\n",
       "1     4.0\n",
       "2     4.0\n",
       "3     4.0\n",
       "4     4.0\n",
       "5     4.0\n",
       "6     4.0\n",
       "7     4.0\n",
       "8     4.0\n",
       "9     4.0\n",
       "10    4.0\n",
       "11    4.0\n",
       "12    4.0\n",
       "13    4.0\n",
       "14    4.0\n",
       "Name: nova_group, dtype: float64"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df['nova_group'].head(15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     1.0\n",
       "1     1.0\n",
       "2     1.0\n",
       "3     1.0\n",
       "4     1.0\n",
       "5     1.0\n",
       "6     1.0\n",
       "7     1.0\n",
       "8     1.0\n",
       "9     1.0\n",
       "10    1.0\n",
       "11    1.0\n",
       "12    1.0\n",
       "13    1.0\n",
       "14    1.0\n",
       "Name: nova_group, dtype: float64"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "imputation_knn_df['nova_group'].head(15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "92      4.0\n",
       "197     4.0\n",
       "261     3.2\n",
       "578     4.0\n",
       "649     4.0\n",
       "868     1.0\n",
       "1040    4.0\n",
       "1261    2.6\n",
       "1343    3.8\n",
       "1358    3.0\n",
       "1497    4.0\n",
       "1725    1.8\n",
       "1775    3.0\n",
       "1776    1.0\n",
       "1813    3.0\n",
       "Name: nova_group, dtype: float64"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_food_facts_df.iloc[index_donnees_imputees]['nova_group'].head(15)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Arrondir les valeurs à l'entier le plus proche ?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 104300 entries, 0 to 104299\n",
      "Data columns (total 9 columns):\n",
      "brands_tags         103844 non-null object\n",
      "labels_en           71866 non-null object\n",
      "additives_tags      104300 non-null object\n",
      "nutriscore_grade    104300 non-null object\n",
      "pnns_groups_1       104300 non-null object\n",
      "pnns_groups_2       104300 non-null object\n",
      "additives_n         104300 non-null float64\n",
      "nutriscore_score    104300 non-null float64\n",
      "nova_group          104300 non-null float64\n",
      "dtypes: float64(3), object(6)\n",
      "memory usage: 7.2+ MB\n"
     ]
    }
   ],
   "source": [
    "open_food_facts_df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "L'imputation avec KNNImputer nécessite de travailler uniquement avec des variables quantitatives. Pour cette raison, toutes les variables utilisées ont été transformées de manière à ce qu'elles soient exploitées par l'algorithme. Malgré ces transformations, on obtient des valeurs décimales pour *nova_group* alors que ses valeurs possibles sont 1, 2, 3 et 4. Cette manière d'imputer *nova_group* n'est donc pas retenue."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Imputation avec tout type de variable avec *[knn_impute.py](https://gist.github.com/YohanObadia/b310793cd22a4427faaadd9c381a5850)*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ce code Python permet de réaliser une imputation de variables quantitatives et qualitatives. Il pourrait donc pallier au problème rencontré lors de l'imputation précédente."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Administrateur\\anaconda3\\lib\\site-packages\\pandas\\core\\series.py:4517: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  return super().fillna(\n"
     ]
    },
    {
     "ename": "MemoryError",
     "evalue": "Unable to allocate 81.1 GiB for an array with shape (104300, 104300) and data type float64",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mMemoryError\u001b[0m                               Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-29-0f5c7c4f5906>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m knn.knn_impute(target=open_food_facts_df['nova_group'],\n\u001b[0m\u001b[0;32m      2\u001b[0m                \u001b[0mk_neighbors\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m5\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      3\u001b[0m                \u001b[0mattributes\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mopen_food_facts_df\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;34m'nutriscore_grade'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m'pnns_groups_1'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m'additives_n'\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      4\u001b[0m                aggregation_method=\"mode\")\n",
      "\u001b[1;32m~\\Desktop\\OpenClassrooms\\P3_Concevoir-une-application-au-service-de-la-sante-publique\\librairies\\knn_impute.py\u001b[0m in \u001b[0;36mknn_impute\u001b[1;34m(target, attributes, k_neighbors, aggregation_method, numeric_distance, categorical_distance, missing_neighbors_threshold)\u001b[0m\n\u001b[0;32m    188\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    189\u001b[0m     \u001b[1;31m# Get the distance matrix and check whether no error was triggered when computing it\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 190\u001b[1;33m     \u001b[0mdistances\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mdistance_matrix\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mattributes\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mnumeric_distance\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mcategorical_distance\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    191\u001b[0m     \u001b[1;32mif\u001b[0m \u001b[0mdistances\u001b[0m \u001b[1;32mis\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    192\u001b[0m         \u001b[1;32mreturn\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Desktop\\OpenClassrooms\\P3_Concevoir-une-application-au-service-de-la-sante-publique\\librairies\\knn_impute.py\u001b[0m in \u001b[0;36mdistance_matrix\u001b[1;34m(data, numeric_distance, categorical_distance)\u001b[0m\n\u001b[0;32m    118\u001b[0m             \u001b[0mresult_matrix\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mcdist\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmetric\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mcategorical_distance\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    119\u001b[0m     \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 120\u001b[1;33m         \u001b[0mresult_numeric\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mcdist\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdata_numeric\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mdata_numeric\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmetric\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mnumeric_distance\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    121\u001b[0m         \u001b[1;32mif\u001b[0m \u001b[0mcategorical_distance\u001b[0m \u001b[1;33m==\u001b[0m \u001b[1;34m\"weighted-hamming\"\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    122\u001b[0m             \u001b[0mresult_categorical\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mweighted_hamming\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdata_categorical\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\scipy\\spatial\\distance.py\u001b[0m in \u001b[0;36mcdist\u001b[1;34m(XA, XB, metric, *args, **kwargs)\u001b[0m\n\u001b[0;32m   2718\u001b[0m     \u001b[0mout\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mpop\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m\"out\"\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2719\u001b[0m     \u001b[1;32mif\u001b[0m \u001b[0mout\u001b[0m \u001b[1;32mis\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 2720\u001b[1;33m         \u001b[0mdm\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mempty\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mmA\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmB\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mnp\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mdouble\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   2721\u001b[0m     \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2722\u001b[0m         \u001b[1;32mif\u001b[0m \u001b[0mout\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mshape\u001b[0m \u001b[1;33m!=\u001b[0m \u001b[1;33m(\u001b[0m\u001b[0mmA\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmB\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mMemoryError\u001b[0m: Unable to allocate 81.1 GiB for an array with shape (104300, 104300) and data type float64"
     ]
    }
   ],
   "source": [
    "knn.knn_impute(target=open_food_facts_df['nova_group'],\n",
    "               k_neighbors=5,\n",
    "               attributes=open_food_facts_df[['nutriscore_grade', 'pnns_groups_1', 'additives_n']],\n",
    "               aggregation_method=\"mode\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Etant donné l'incapacité à allouer de la mémoire pour cette fonction, cette méthode est abandonnée."
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "PSante_01_notebooknettoyage.ipynb",
   "provenance": [],
   "toc_visible": true
  },
  "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}