src/seance5-analyse.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "source": [
    "# Analyse de données sous `Python`\n",
    "\n",
    "Nous utilisons dans ce TP le module [`scikit-learn`](http://scikit-learn.org/), dédié au *Machine Learning*. Pour mieux la découvrir, nous allons réaliser une étude de cas, avec les données `iris`.\n",
    "\n",
    "Dans cette étude, nous allons chercher à savoir s'il existe différentes sous-espèces d'iris. Pour cela, nous allons procéder par étapes ::\n",
    "\n",
    "1. Visualisation des données, via une **ACP**\n",
    "1. Classification non-supervisée interne à chaque espèce, via **DBSCAN**\n",
    "1. Analyse des résultats\n",
    "\n",
    "Dans un premier temps, il va nous falloir importer les données (`iris`) que nous allons utiliser (via la librairie `pydataset`). Nous allons aussi utiliser d'autres librairies (telles que `seaborn`, `numpy`, `matplotlib` et `pandas`).\n",
    "\n",
    "Nous importerons les éléments de `scikit-learn` (module `sklearn`) au fur et à mesure."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy\n",
    "import pandas\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn\n",
    "seaborn.set_style(\"white\")\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pour rappel, la table se présente comme ceci :"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>Sepal Length</th>\n",
       "      <th>Sepal Width</th>\n",
       "      <th>Petal Length</th>\n",
       "      <th>Petal Width</th>\n",
       "      <th>Species</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5.1</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.6</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Sepal Length  Sepal Width  Petal Length  Petal Width Species\n",
       "0           5.1          3.5           1.4          0.2  setosa\n",
       "1           4.9          3.0           1.4          0.2  setosa\n",
       "2           4.7          3.2           1.3          0.2  setosa\n",
       "3           4.6          3.1           1.5          0.2  setosa\n",
       "4           5.0          3.6           1.4          0.2  setosa"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iris = pandas.read_csv(\"Iris.txt\", sep=\"\\t\")\n",
    "iris.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ACP\n",
    "\n",
    "Dans le sous-module `decomposition`, nous allons importer la fonction [`PCA()`](http://scikit-learn.org/stable/modules/generated/sklearn.decomposition.PCA.html). Le fonctionnement de celle-ci est similaire à toutes les autres dans `scikit-learn`. \n",
    "\n",
    "1. On créé d'abord un objet (nommé ici `pca`) qui va contenir le résultat de l'ACP. Dans la fonction `PCA()`, on pourra indiquer les paramètres tels que le nombre de composants à retenir (`n_components`) ou la méthode de calcul (`svd_solver`).\n",
    "2. Ensuite, on ajuste (*fit*  en anglais) sur des données avec la fonction `fit()` de l'objet précédemment créé. Dans cette fonction, nous devons donc passer les données à utiliser.\n",
    "\n",
    "Si l'on souhaite une ACP normée, il nous faut standardiser les données en amont, avec la fonction `scale()` du sous-module `preprocessing` (importée aussi ici). Bien évidemment, il ne faut passer que des variables quantitatives."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PCA(copy=True, iterated_power='auto', n_components=None, random_state=None,\n",
       "  svd_solver='auto', tol=0.0, whiten=False)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "from sklearn.preprocessing import scale\n",
    "\n",
    "pca = PCA()\n",
    "pca.fit(scale(iris[iris.columns[:4]]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "L'objet `pca` comprend maintenant un certain nombre d'objets et de fonctions. Le premier objet est le tableau des variances expliquées (`explained_variance_`) par chaque dimension, et le ratio (proportion) de variance expliquée par dimension (`explained_variance_ratio_`)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2.93808505 0.9201649  0.14774182 0.02085386]\n",
      "[0.72962445 0.22850762 0.03668922 0.00517871]\n"
     ]
    }
   ],
   "source": [
    "print(pca.explained_variance_)\n",
    "print(pca.explained_variance_ratio_)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Bien évidemment, il est possible (et préférable) de faire un tableau récapitulatif, avec les valeurs propres, les proportions de variance expliquée simples et cumulées. Voici un petit code permettant de faire ceci."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>Dimension</th>\n",
       "      <th>Valeur propre</th>\n",
       "      <th>% variance expliquée</th>\n",
       "      <th>% cum. var. expliquée</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Dim1</td>\n",
       "      <td>2.938085</td>\n",
       "      <td>73.0</td>\n",
       "      <td>73.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Dim2</td>\n",
       "      <td>0.920165</td>\n",
       "      <td>23.0</td>\n",
       "      <td>96.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Dim3</td>\n",
       "      <td>0.147742</td>\n",
       "      <td>4.0</td>\n",
       "      <td>99.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Dim4</td>\n",
       "      <td>0.020854</td>\n",
       "      <td>1.0</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Dimension  Valeur propre  % variance expliquée  % cum. var. expliquée\n",
       "0      Dim1       2.938085                  73.0                   73.0\n",
       "1      Dim2       0.920165                  23.0                   96.0\n",
       "2      Dim3       0.147742                   4.0                   99.0\n",
       "3      Dim4       0.020854                   1.0                  100.0"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "eig = pandas.DataFrame(\n",
    "    {\n",
    "        \"Dimension\" : [\"Dim\" + str(x + 1) for x in range(4)], \n",
    "        \"Valeur propre\" : pca.explained_variance_,\n",
    "        \"% variance expliquée\" : numpy.round(pca.explained_variance_ratio_ * 100),\n",
    "        \"% cum. var. expliquée\" : numpy.round(numpy.cumsum(pca.explained_variance_ratio_) * 100)\n",
    "    },\n",
    "    columns = [\"Dimension\", \"Valeur propre\", \"% variance expliquée\", \"% cum. var. expliquée\"]\n",
    ")\n",
    "eig"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "L'idée est de représenter graphiquement ces proportions de variances expliquées (qu'on passe en pourcentage par convenance)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(3.25,26,'25%')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9f380f6ef0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "g_eig = seaborn.barplot(x = \"Dimension\", \n",
    "                y = \"% variance expliquée\",\n",
    "                palette = [\"lightseagreen\"],\n",
    "                data = eig)\n",
    "g_eig.set(ylabel = \"Variance expliquée (%)\")\n",
    "g_eig.figure.suptitle(\"Variance expliquée par dimension\")\n",
    "plt.axhline(y = 25, linewidth = .5, color = \"dimgray\", linestyle = \"--\") # 25 = 100 / 4 (nb dimensions)\n",
    "plt.text(3.25, 26, \"25%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "On remarque ici qu'avec seulement deux dimensions suffisent à expliquer 96 %  de la variance des données. Nous allons maintenant calculer les coordonnées des iris sur les dimensions, avec la fonction `transform()` de l'objet `pca`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "iris_pca = pca.transform(iris[iris.columns[:4]])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Afin de manipuler plus facilement l'objet obtenu par la suite, nous allons créer un `DataFrame` `pandas` en ne prenant que les deux premières dimensions, ainsi que les espèces."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>Dim1</th>\n",
       "      <th>Dim2</th>\n",
       "      <th>Species</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2.640270</td>\n",
       "      <td>5.204041</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2.670730</td>\n",
       "      <td>4.666910</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2.454606</td>\n",
       "      <td>4.773636</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2.545517</td>\n",
       "      <td>4.648463</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2.561228</td>\n",
       "      <td>5.258629</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Dim1      Dim2 Species\n",
       "0  2.640270  5.204041  setosa\n",
       "1  2.670730  4.666910  setosa\n",
       "2  2.454606  4.773636  setosa\n",
       "3  2.545517  4.648463  setosa\n",
       "4  2.561228  5.258629  setosa"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Transformation en DataFrame pandas\n",
    "iris_pca_df = pandas.DataFrame({\n",
    "    \"Dim1\" : iris_pca[:,0], \n",
    "    \"Dim2\" : iris_pca[:,1], \n",
    "    \"Species\" : iris.Species\n",
    "})\n",
    "\n",
    "# Résultat (premières lignes)\n",
    "iris_pca_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Il est maintenant possible de représenter les données sur le premier plan factoriel, en ajoutant bien évidemment l'information sur les espèces."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,0.98,'Premier plan factoriel')"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9f17c59978>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "g_pca = seaborn.lmplot(\"Dim1\", \"Dim2\", hue = \"Species\", data = iris_pca_df, fit_reg = False)\n",
    "g_pca.set(xlabel = \"Dimension 1 (73%)\", ylabel = \"Dimension 2 (23 %)\")\n",
    "g_pca.fig.suptitle(\"Premier plan factoriel\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Il est aussi possible de différencier l'affichage de ce premier plan par espèce, grâce à l'option `col` de `lmplot()`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x7f9f17c01748>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9f17c016d8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "g_pca2 = seaborn.lmplot(\"Dim1\", \"Dim2\", hue = \"Species\", col = \"Species\", col_wrap = 2, data = iris_pca_df, fit_reg = False)\n",
    "g_pca2.set(xlabel = \"Dimension 1 (73%)\", ylabel = \"Dimension 2 (23 %)\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## DBSCAN\n",
    "\n",
    "[`DBSCAN`](https://fr.wikipedia.org/wiki/DBSCAN) est un algorithme de classification non supervisée, basé sur la densité. Il est intéressant car il ne nécessite pas de connaître le nombre de classes, mais une estimation de la densité (globale) des données. En effet, les points proches (distance inférieure à $\\varepsilon$) sont consiédérés dans la même classe. Si toutefois cette classe comporte au moins un cetain nombre de points au final. Si ce n'est pas le cas, les points sont considérés comme *outliers* et mis à part.\n",
    "\n",
    "Nous utilisons ici la fonction [`DBSCAN()`](http://scikit-learn.org/stable/modules/generated/sklearn.cluster.DBSCAN.html) du sous-module `cluster`. Celle-ci peut prendre en paramètre, entre autres, la valeur de $\\varepsilon$ (`eps`, `.5` par défaut) et le nombre minimal d'objets dans une classe (`min_samples`, `5` par défaut).\n",
    "\n",
    "Puis nous ajustons les données standardisées (pour ne pas donner plus d'influences à une variable qu'à une autre)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DBSCAN(algorithm='auto', eps=0.6, leaf_size=30, metric='euclidean',\n",
       "    metric_params=None, min_samples=3, n_jobs=1, p=None)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.cluster import DBSCAN\n",
    "\n",
    "db = DBSCAN(eps = .6, min_samples = 3)\n",
    "db.fit(scale(iris[iris.columns[:4]]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Les classes dont contenues dans l'objet `labels_` de `db`. On a pour chaque individu sa classe (de $0$ à $K-1$ pour $K$ classes obtenues). Les *outliers* sont étiquettés $-1$ ."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,\n",
       "        0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,\n",
       "        0,  0,  0,  0,  0,  0,  0, -1,  0,  0,  0,  0,  0,  0,  0,  0,  1,\n",
       "        1,  1,  1,  1,  1,  1,  2,  1,  1, -1,  1, -1,  1,  1,  1,  1,  1,\n",
       "        3,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,\n",
       "        1,  1,  3,  1,  1,  1,  1,  1,  2,  1,  1,  1,  1,  2,  1,  1,  1,\n",
       "        1,  1,  1,  1, -1,  1, -1, -1,  1,  1,  1,  1,  1,  1,  1, -1, -1,\n",
       "        3,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1, -1,  1,  1,  1,  1,\n",
       "        1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1])"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "db.labels_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "On peut faire un tableau récapitulatif des effectifs par classes."
   ]
  },
  {
   "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>Classe</th>\n",
       "      <th>Effectif</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-1</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Classe  Effectif\n",
       "0      -1         9\n",
       "1       0        49\n",
       "2       1        86\n",
       "3       2         3\n",
       "4       3         3"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "eff = numpy.unique(db.labels_, return_counts = True)\n",
    "pandas.DataFrame({\n",
    "    \"Classe\" : eff[0],\n",
    "    \"Effectif\" : eff[1]\n",
    "})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pour la représentation des données, on ajoute ces labels de classe aux projections des iris sur le premier plan factoriel."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "iris_pca_db = iris_pca_df.assign(Labels = db.labels_)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "On peut ainsi représenter les données en fonction des espèces et des classes obtenues."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x7f9f14c60dd8>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9f14c6f400>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "seaborn.lmplot(\"Dim1\", \"Dim2\", hue = \"Labels\", col = \"Species\", data = iris_pca_db, fit_reg = False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "On peut aussi *splitter* le graphique en ligne pour chaque classe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x7f9f14be2128>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9f14c0acf8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "seaborn.lmplot(\"Dim1\", \"Dim2\", hue = \"Labels\", col = \"Species\", row = \"Labels\", data = iris_pca_db, fit_reg = False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Exercice\n",
    "\n",
    "### ACP\n",
    "\n",
    "Nous allons travailler ici sur les données de température mondiale HadCRUT4, fournies par [Climate Research Unit](https://crudata.uea.ac.uk/). Vous trouverez plus d’informations sur ces données sur ce [lien](https://crudata.uea.ac.uk/cru/data/temperature/). Nous avons ici l’historique des anomalies moyennes mensuelles et annuelles depuis 1850, au niveau mondial, par rapport à la période 1961-1990.\n",
    "\n",
    "Le code ci-dessous télécharge directement les dernières données disponibles et les met dans un DataFrame dont vous avez un aperçu en dessous (en supprimant l’année 2019, incomplète)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": true
   },
   "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>Year</th>\n",
       "      <th>Jan</th>\n",
       "      <th>Feb</th>\n",
       "      <th>Mar</th>\n",
       "      <th>Apr</th>\n",
       "      <th>May</th>\n",
       "      <th>Jun</th>\n",
       "      <th>Jul</th>\n",
       "      <th>Aug</th>\n",
       "      <th>Sep</th>\n",
       "      <th>Oct</th>\n",
       "      <th>Nov</th>\n",
       "      <th>Dec</th>\n",
       "      <th>Annual</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>165</th>\n",
       "      <td>2015.0</td>\n",
       "      <td>0.705</td>\n",
       "      <td>0.689</td>\n",
       "      <td>0.708</td>\n",
       "      <td>0.673</td>\n",
       "      <td>0.711</td>\n",
       "      <td>0.743</td>\n",
       "      <td>0.697</td>\n",
       "      <td>0.740</td>\n",
       "      <td>0.796</td>\n",
       "      <td>0.845</td>\n",
       "      <td>0.840</td>\n",
       "      <td>1.024</td>\n",
       "      <td>0.763</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>166</th>\n",
       "      <td>2016.0</td>\n",
       "      <td>0.934</td>\n",
       "      <td>1.111</td>\n",
       "      <td>1.106</td>\n",
       "      <td>0.937</td>\n",
       "      <td>0.707</td>\n",
       "      <td>0.744</td>\n",
       "      <td>0.744</td>\n",
       "      <td>0.790</td>\n",
       "      <td>0.729</td>\n",
       "      <td>0.598</td>\n",
       "      <td>0.553</td>\n",
       "      <td>0.620</td>\n",
       "      <td>0.797</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>167</th>\n",
       "      <td>2017.0</td>\n",
       "      <td>0.739</td>\n",
       "      <td>0.845</td>\n",
       "      <td>0.873</td>\n",
       "      <td>0.737</td>\n",
       "      <td>0.659</td>\n",
       "      <td>0.641</td>\n",
       "      <td>0.651</td>\n",
       "      <td>0.714</td>\n",
       "      <td>0.557</td>\n",
       "      <td>0.571</td>\n",
       "      <td>0.554</td>\n",
       "      <td>0.600</td>\n",
       "      <td>0.677</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>168</th>\n",
       "      <td>2018.0</td>\n",
       "      <td>0.554</td>\n",
       "      <td>0.528</td>\n",
       "      <td>0.615</td>\n",
       "      <td>0.627</td>\n",
       "      <td>0.587</td>\n",
       "      <td>0.573</td>\n",
       "      <td>0.594</td>\n",
       "      <td>0.586</td>\n",
       "      <td>0.598</td>\n",
       "      <td>0.684</td>\n",
       "      <td>0.591</td>\n",
       "      <td>0.604</td>\n",
       "      <td>0.595</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>169</th>\n",
       "      <td>2019.0</td>\n",
       "      <td>0.738</td>\n",
       "      <td>0.669</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.705</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Year    Jan    Feb    Mar    Apr    May    Jun    Jul    Aug    Sep  \\\n",
       "165  2015.0  0.705  0.689  0.708  0.673  0.711  0.743  0.697  0.740  0.796   \n",
       "166  2016.0  0.934  1.111  1.106  0.937  0.707  0.744  0.744  0.790  0.729   \n",
       "167  2017.0  0.739  0.845  0.873  0.737  0.659  0.641  0.651  0.714  0.557   \n",
       "168  2018.0  0.554  0.528  0.615  0.627  0.587  0.573  0.594  0.586  0.598   \n",
       "169  2019.0  0.738  0.669  0.000  0.000  0.000  0.000  0.000  0.000  0.000   \n",
       "\n",
       "       Oct    Nov    Dec  Annual  \n",
       "165  0.845  0.840  1.024   0.763  \n",
       "166  0.598  0.553  0.620   0.797  \n",
       "167  0.571  0.554  0.600   0.677  \n",
       "168  0.684  0.591  0.604   0.595  \n",
       "169  0.000  0.000  0.000   0.705  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "had = pandas.read_csv(\"https://crudata.uea.ac.uk/cru/data/temperature/HadCRUT4-gl.dat\", header=None)\n",
    "donnees = pandas.DataFrame(\n",
    "    [list(map(lambda v: float(v), filter(lambda v: v!= \"\", h.split(\" \")))) for h in had[0][::2]],\n",
    "    columns = [\"Year\", \"Jan\", \"Feb\", \"Mar\", \"Apr\", \"May\", \"Jun\", \"Jul\", \"Aug\", \"Sep\", \"Oct\", \"Nov\", \"Dec\", \"Annual\"]\n",
    ")\n",
    "donnees.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Vous devez donc réaliser les étapes suivantes :\n",
    "\n",
    "- Décrire rapidement les données\n",
    "- Réaliser une ACP centrée ou normée (choix à justifier), sur les données mensuelles\n",
    "- Ajouter la moyenne annuelle en variable supplémentaire\n",
    "- Produire les graphiques nécessaires à l’interprétation\n",
    "- Identifier des années particulières\n",
    "    - Refaire l’ACP en les indiquant comme individus supplémentaires\n",
    "    - Comparer les résultats pour voir s’il est opportun ou non de les garder dans l’analyse\n",
    "- Que peut-on dire globalement ?\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### DBSCAN\n",
    "\n",
    "La banque mondiale fournit un grand nombre de données, dont des indicateurs de gouvernance au niveau mondial (voir [ici](https://data.worldbank.org/data-catalog/worldwide-governance-indicators)). Le code ci-dessous importe les données du fichier [WGI_Data.csv](https://fxjollois.github.io/cours-2017-2018/analyse-donnees/WGI_Data.csv) (que vous devez donc télécharger) pour les importer. Les informations concernant la définition des indicateurs et leur source se trouvent dans le fichier [WGI_Definition and Source.csv](https://fxjollois.github.io/cours-2017-2018/analyse-donnees/WGI_Definition%20and%20Source.csv)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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>Country Name</th>\n",
       "      <th>Country Code</th>\n",
       "      <th>Series Name</th>\n",
       "      <th>Series Code</th>\n",
       "      <th>Value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>AFG</td>\n",
       "      <td>Control of Corruption: Estimate</td>\n",
       "      <td>CC.EST</td>\n",
       "      <td>-1.56373512744904</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>AFG</td>\n",
       "      <td>Government Effectiveness: Estimate</td>\n",
       "      <td>GE.EST</td>\n",
       "      <td>-1.2221063375473</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>AFG</td>\n",
       "      <td>Political Stability and Absence of Violence/Te...</td>\n",
       "      <td>PV.EST</td>\n",
       "      <td>-2.74916744232178</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>AFG</td>\n",
       "      <td>Regulatory Quality: Estimate</td>\n",
       "      <td>RQ.EST</td>\n",
       "      <td>-1.32791936397552</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>AFG</td>\n",
       "      <td>Rule of Law: Estimate</td>\n",
       "      <td>RL.EST</td>\n",
       "      <td>-1.61642956733704</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Country Name Country Code  \\\n",
       "0  Afghanistan          AFG   \n",
       "1  Afghanistan          AFG   \n",
       "2  Afghanistan          AFG   \n",
       "3  Afghanistan          AFG   \n",
       "4  Afghanistan          AFG   \n",
       "\n",
       "                                         Series Name Series Code  \\\n",
       "0                    Control of Corruption: Estimate      CC.EST   \n",
       "1                 Government Effectiveness: Estimate      GE.EST   \n",
       "2  Political Stability and Absence of Violence/Te...      PV.EST   \n",
       "3                       Regulatory Quality: Estimate      RQ.EST   \n",
       "4                              Rule of Law: Estimate      RL.EST   \n",
       "\n",
       "               Value  \n",
       "0  -1.56373512744904  \n",
       "1   -1.2221063375473  \n",
       "2  -2.74916744232178  \n",
       "3  -1.32791936397552  \n",
       "4  -1.61642956733704  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wgi_one = pandas.read_csv(\"WGI_Data.csv\")\n",
    "wgi_one.head()"
   ]
  },
  {
   "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>Series Code</th>\n",
       "      <th>CC.EST</th>\n",
       "      <th>GE.EST</th>\n",
       "      <th>PV.EST</th>\n",
       "      <th>RL.EST</th>\n",
       "      <th>RQ.EST</th>\n",
       "      <th>VA.EST</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Country Name</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",
       "      <th>Afghanistan</th>\n",
       "      <td>-1.563735</td>\n",
       "      <td>-1.222106</td>\n",
       "      <td>-2.749167</td>\n",
       "      <td>-1.616430</td>\n",
       "      <td>-1.327919</td>\n",
       "      <td>-1.086220</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Albania</th>\n",
       "      <td>-0.397567</td>\n",
       "      <td>-0.000573</td>\n",
       "      <td>0.257720</td>\n",
       "      <td>-0.347468</td>\n",
       "      <td>0.188962</td>\n",
       "      <td>0.159191</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Algeria</th>\n",
       "      <td>-0.691359</td>\n",
       "      <td>-0.544695</td>\n",
       "      <td>-1.139282</td>\n",
       "      <td>-0.849319</td>\n",
       "      <td>-1.171843</td>\n",
       "      <td>-0.877241</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Samoa</th>\n",
       "      <td>1.230341</td>\n",
       "      <td>0.531472</td>\n",
       "      <td>1.192534</td>\n",
       "      <td>1.328246</td>\n",
       "      <td>0.136647</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Andorra</th>\n",
       "      <td>1.230341</td>\n",
       "      <td>1.860201</td>\n",
       "      <td>1.396047</td>\n",
       "      <td>1.564011</td>\n",
       "      <td>0.871409</td>\n",
       "      <td>1.197540</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Series Code       CC.EST    GE.EST    PV.EST    RL.EST    RQ.EST    VA.EST\n",
       "Country Name                                                              \n",
       "Afghanistan    -1.563735 -1.222106 -2.749167 -1.616430 -1.327919 -1.086220\n",
       "Albania        -0.397567 -0.000573  0.257720 -0.347468  0.188962  0.159191\n",
       "Algeria        -0.691359 -0.544695 -1.139282 -0.849319 -1.171843 -0.877241\n",
       "American Samoa  1.230341  0.531472  1.192534  1.328246  0.136647       NaN\n",
       "Andorra         1.230341  1.860201  1.396047  1.564011  0.871409  1.197540"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wgi_bis = wgi_one[[\"Country Name\", \"Series Code\", \"Value\"]][wgi_one.Value != \"..\"]\n",
    "wgi_ter = wgi_bis.astype({\"Value\": float})\n",
    "wgi = wgi_ter.pivot_table(values = \"Value\", index = \"Country Name\", columns = \"Series Code\")\n",
    "wgi.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Vous devez donc réaliser les étapes suivantes :\n",
    "\n",
    "- Décrire rapidement les données\n",
    "- Réaliser une ACP centrée ou normée (choix à justifier), sur les données mensuelles\n",
    "- Ajouter la moyenne annuelle en variable supplémentaire\n",
    "- Produire les graphiques nécessaires à l’interprétation\n",
    "- Identifier des années particulières\n",
    "    - Refaire l’ACP en les indiquant comme individus supplémentaires\n",
    "    - Comparer les résultats pour voir s’il est opportun ou non de les garder dans l’analyse\n",
    "- Que peut-on dire globalement ?\n",
    "\n",
    "### Description des données"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": true
   },
   "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>Year</th>\n",
       "      <th>Jan</th>\n",
       "      <th>Feb</th>\n",
       "      <th>Mar</th>\n",
       "      <th>Apr</th>\n",
       "      <th>May</th>\n",
       "      <th>Jun</th>\n",
       "      <th>Jul</th>\n",
       "      <th>Aug</th>\n",
       "      <th>Sep</th>\n",
       "      <th>Oct</th>\n",
       "      <th>Nov</th>\n",
       "      <th>Dec</th>\n",
       "      <th>Annual</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>170.00</td>\n",
       "      <td>170.00</td>\n",
       "      <td>170.00</td>\n",
       "      <td>170.00</td>\n",
       "      <td>170.00</td>\n",
       "      <td>170.00</td>\n",
       "      <td>170.00</td>\n",
       "      <td>170.00</td>\n",
       "      <td>170.00</td>\n",
       "      <td>170.00</td>\n",
       "      <td>170.00</td>\n",
       "      <td>170.00</td>\n",
       "      <td>170.00</td>\n",
       "      <td>170.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1934.50</td>\n",
       "      <td>-0.09</td>\n",
       "      <td>-0.09</td>\n",
       "      <td>-0.14</td>\n",
       "      <td>-0.10</td>\n",
       "      <td>-0.10</td>\n",
       "      <td>-0.07</td>\n",
       "      <td>-0.05</td>\n",
       "      <td>-0.05</td>\n",
       "      <td>-0.07</td>\n",
       "      <td>-0.08</td>\n",
       "      <td>-0.12</td>\n",
       "      <td>-0.12</td>\n",
       "      <td>-0.09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>49.22</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.33</td>\n",
       "      <td>0.31</td>\n",
       "      <td>0.30</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.29</td>\n",
       "      <td>0.29</td>\n",
       "      <td>0.31</td>\n",
       "      <td>0.33</td>\n",
       "      <td>0.34</td>\n",
       "      <td>0.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1850.00</td>\n",
       "      <td>-0.97</td>\n",
       "      <td>-0.78</td>\n",
       "      <td>-0.83</td>\n",
       "      <td>-0.72</td>\n",
       "      <td>-0.77</td>\n",
       "      <td>-0.58</td>\n",
       "      <td>-0.57</td>\n",
       "      <td>-0.70</td>\n",
       "      <td>-0.56</td>\n",
       "      <td>-0.71</td>\n",
       "      <td>-0.77</td>\n",
       "      <td>-0.89</td>\n",
       "      <td>-0.54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1892.25</td>\n",
       "      <td>-0.32</td>\n",
       "      <td>-0.35</td>\n",
       "      <td>-0.40</td>\n",
       "      <td>-0.34</td>\n",
       "      <td>-0.32</td>\n",
       "      <td>-0.28</td>\n",
       "      <td>-0.25</td>\n",
       "      <td>-0.25</td>\n",
       "      <td>-0.28</td>\n",
       "      <td>-0.32</td>\n",
       "      <td>-0.37</td>\n",
       "      <td>-0.36</td>\n",
       "      <td>-0.30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>1934.50</td>\n",
       "      <td>-0.11</td>\n",
       "      <td>-0.15</td>\n",
       "      <td>-0.21</td>\n",
       "      <td>-0.17</td>\n",
       "      <td>-0.17</td>\n",
       "      <td>-0.14</td>\n",
       "      <td>-0.10</td>\n",
       "      <td>-0.09</td>\n",
       "      <td>-0.11</td>\n",
       "      <td>-0.11</td>\n",
       "      <td>-0.13</td>\n",
       "      <td>-0.17</td>\n",
       "      <td>-0.16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1976.75</td>\n",
       "      <td>0.12</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.06</td>\n",
       "      <td>0.09</td>\n",
       "      <td>0.06</td>\n",
       "      <td>0.07</td>\n",
       "      <td>0.05</td>\n",
       "      <td>0.08</td>\n",
       "      <td>0.07</td>\n",
       "      <td>0.07</td>\n",
       "      <td>0.03</td>\n",
       "      <td>0.11</td>\n",
       "      <td>0.05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2019.00</td>\n",
       "      <td>0.93</td>\n",
       "      <td>1.11</td>\n",
       "      <td>1.11</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.71</td>\n",
       "      <td>0.74</td>\n",
       "      <td>0.74</td>\n",
       "      <td>0.79</td>\n",
       "      <td>0.80</td>\n",
       "      <td>0.84</td>\n",
       "      <td>0.84</td>\n",
       "      <td>1.02</td>\n",
       "      <td>0.80</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Year     Jan     Feb     Mar     Apr     May     Jun     Jul  \\\n",
       "count   170.00  170.00  170.00  170.00  170.00  170.00  170.00  170.00   \n",
       "mean   1934.50   -0.09   -0.09   -0.14   -0.10   -0.10   -0.07   -0.05   \n",
       "std      49.22    0.36    0.37    0.36    0.33    0.31    0.30    0.28   \n",
       "min    1850.00   -0.97   -0.78   -0.83   -0.72   -0.77   -0.58   -0.57   \n",
       "25%    1892.25   -0.32   -0.35   -0.40   -0.34   -0.32   -0.28   -0.25   \n",
       "50%    1934.50   -0.11   -0.15   -0.21   -0.17   -0.17   -0.14   -0.10   \n",
       "75%    1976.75    0.12    0.14    0.06    0.09    0.06    0.07    0.05   \n",
       "max    2019.00    0.93    1.11    1.11    0.94    0.71    0.74    0.74   \n",
       "\n",
       "          Aug     Sep     Oct     Nov     Dec  Annual  \n",
       "count  170.00  170.00  170.00  170.00  170.00  170.00  \n",
       "mean    -0.05   -0.07   -0.08   -0.12   -0.12   -0.09  \n",
       "std      0.29    0.29    0.31    0.33    0.34    0.31  \n",
       "min     -0.70   -0.56   -0.71   -0.77   -0.89   -0.54  \n",
       "25%     -0.25   -0.28   -0.32   -0.37   -0.36   -0.30  \n",
       "50%     -0.09   -0.11   -0.11   -0.13   -0.17   -0.16  \n",
       "75%      0.08    0.07    0.07    0.03    0.11    0.05  \n",
       "max      0.79    0.80    0.84    0.84    1.02    0.80  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "numpy.round(donnees.describe(), 2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f97a11c0710>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZIAAAEKCAYAAAA4t9PUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAIABJREFUeJzsvXec3Hd95/+cPjszO9v7rrSqX1tylStuBGyMqSYJEEoSSiCFB3eXI+Qglxy/XHLHmcsRQhIuuZxztBQMhIADBmNMccFFrrIl66su7Wp7m95nfn98y3yn7c5qy+xq38/HQw9N+czMZ1arz+v77rZCoYAgCIIgXCj2Rm9AEARB2NyIkAiCIAgrQoREEARBWBEiJIIgCMKKECERBEEQVoQIiSAIgrAiREgEQRCEFSFCIgiCIKwIERJBEARhRTgbvYHV5oYbbigMDAw0ehuCIAibisOHD8+oqtp1Ia+96IRkYGCAb33rW43ehiAIwqZCUZSzF/pacW0JgiAIK0KERBAEQVgRIiSCIAjCihAhEQRBEFZEQ4PtiqLcBXwecAD3qqp6T9nz24AvA636mk+qqvrAum9UEARBqEnDLBJFURzAF4A3APuAdyuKsq9s2R8BX1dV9WrgXcD/Xt9dCoIgCEvRSNfW9cAJVVVPqaqaBr4G3F22pgAE9dstwNg67k8QBEGog0YKyQAwYrk/qj9m5Y+BX1UUZRR4APh367M1QRCEjUk+X+Cbz44SSmQavRWTjR5sfzfwJVVVB4E3Al9VFGWj71kQBGHN+PnJWT7+jRf57qGN46Bp5KF8Hhiy3B/UH7PyG8DXAVRVfQLwAp3rsjtBEIQNyL8+rx2Tk6Fkg3dSpJFCchDYoyjKDkVR3GjB9PvL1pwDbgdQFOVSNCGZXtddCoIgbBAS6Rw/eHkcgMlwqsG7KdIwIVFVNQt8FHgQeAUtO+uwoih/oijKW/Vlvwd8WFGUF4F/Bt6vqmqhMTsWBEFoLA+9MkksncPjtDMZ2TgWSUPrSPSakAfKHvuU5fYR4Ob13pcgCMJG5NvPn6evxcu+viDnFxKN3o6JBK4FQRA2CQdPz/HaS7rpbfEyFdk4rq2Lro28IAjCxUg2lyeSytLV7MFuszEXS5PK5vA4HY3emlgkgiAIm4FwMgtAS5OLnqAHgOkNYpWIkAiCIGwCjALEliYX3UEvsHEyt0RIBEEQNgFWIelp1oRkKrwxMrdESARBEDYBJUKiu7YmRUgEQRCEerEKSZvPjcthY1JiJIIgCEK9WIXEbrfR3ewVi0QQBEGon7AuJMEmFwDdQQ9TlmD75x46xj3fP9qQvUkdiSAIwiYglMjgcdrxurS6kZ5mLyeno+bzP3plku5mT0P2JhaJIAjCJiAUz9CiWyMAPUEPExbX1kQoSW9LUyO2JhaJIAjCZiCUKBWS7qCXSDJLPJ3FbrMxG0vT3+JtyN5ESARBEDYB5ULSYylKtNu0x3pFSARBEIRahBIZ+ixCMdCqubFG5uK4HFqUoq9Bri2JkQiCIGwCyi2SXV1+AE7PxJgIay3l+1rFIhEEQRBqEE5kzNRfgK5mD363g9MzMbr1Svc+cW0JgiAI1cjlC0RS2RKLxGazsaPLz6mZGLl8gZYmFz53Y450ERJBEIQNTthS1W5lR2eAF0bmcTvsDbNGQGIkgiAIG55QTSHxMzqf4OxsTIREEARBqE0tIdnV5adQgONT0YYVI4IIiSAIwobHFBJfpUVi0EiLpKExEkVR7gI+DziAe1VVvafKmncCfwwUgBdVVX3Pum5SEAShwdSySIY3iJA0zCJRFMUBfAF4A7APeLeiKPvK1uwB/gC4WVXV/cDvrvtGBUEQGkwtIQl6XXQGjNTfrenauh44oarqKVVV08DXgLvL1nwY+IKqqvMAqqpOrfMeBUEQGk4tIQHYqVsljSpGhMYKyQAwYrk/qj9mZS+wV1GUxxVFeVJ3hQmCIGwpwokMbksLeStGnGTLxkjqwAnsAX4BGAQeURTlclVVFxq6K0EQhHWkvD2KlV86MEDA62xYMSI0VkjOA0OW+4P6Y1ZGgadUVc0ApxVFOYYmLAfXZ4uCIAiNZzEhuWFnBzfs7FjnHZXSSNfWQWCPoig7FEVxA+8C7i9b8200awRFUTrRXF2n1nOTgiAIjWY6kqLd7270NmrSMCFRVTULfBR4EHgF+LqqqocVRfkTRVHeqi97EJhVFOUI8BPg91VVnW3MjgVBENaffL7A0YkIl/Q2N3orNWlojERV1QeAB8oe+5TldgH4mP5HEAThouU7L5zn0r4ge3tKBWNkPk40lWVfX7BBO1saqWwXBEFoMPF0lo99/UW++PjpiueOjIUB2NcvQiIIgrBlOTIWZj6Wrvn8iyMhcvkC05FU5WvHwzjstgpLZSMhQiIIgrAMJkJJHj0+Xff6QqHAu//vk/zpd4/UXPPcuXkApqOVYnNkLMyuLn/VGpKNggiJIAjCMvj7x07x4a88U/f6+XiGUCLDQ69Mksnlq6559qwmJDM1LJL9/S0Xttl1QoREEARhGUxFUiQzeVLZXF3rR+fjAESSWZ46NVfxfKFQMC2SmWiKQqFgPjcXSzMeSm7oQDuIkAiCICyLOT3WEUvVKyQJ8/YPj0xUPH9qJsZCPMOuLj+pbJ5IKms+txkC7SBCIgiCsCxm9ThGNJldYqWGYZG8amcHPzw8WWJxQNGt9fr9vUCpe+vIeAiAS8UiEQRBuHgwLJJoqraQnJ6J8fMTM4BmkTR7nfzyNYNMhJM8P1LaKvC5s/O0NLnMNiczloD7k6fmGO7wbeiqdhAhEQRBqJtCoVB0baUrhaRQKHDfwXO84fOP8Ov/72ni6Szn5xMMtvl43aU9tPpcfOKbh0pE6NhkhH19QbqbtbkiM1HNIkllczxxcpbb9natwzdbGSIkgiAIdRJNZUnrmVfVXFs/Vaf5xL+8RHezl2y+wOGxMKPzCQbbmmjxufjf7znAqZkY//G+F0wX10QoSX9rkzmgyhCS584ukMjkuHWPCIkgCMJFw5ylqLCaa+uxEzN4nHa+9ps3AvDCuQVG5+MMtmnTC2/a3cnv3bmXh45McmQ8TC5fYDKSoq/FS7vfjd1WjJE8cnwap93GjTvb1+GbrQwREkEQhDqZXUJInjs3z+UDLfS3NjHQ2sTPjk0TS+cYbPOZa27TLYyRuTjTkRS5fIG+Vi8Ou412v9ssSnz0+DQHtrfR7K3ePn4jIUIiCIJQJ3OWQHisTEhS2RyHz4c5sL0NgKuGWvn5SS3gblgkAEO6qIzMJRgPaanBxnTDzoCH6UiK2WiKl8+HuW1P59p9mVVEhEQQBKFOrK6tSFmM5PBYmHQuz4FtrQBcOdRCXs/0tQpJi89Fs9fJyHyciVASgN6g9nxXs4eZaIrH9IyvzRAfARESQRCEujFcW067rcIieU6vBzmwTbNIrhxsNZ+zuraM+6PzCcZ1IbFaJDPRFI8cm6HV5+KygY3dGsVgo89sFwRB2DDMxVJ4XXaava6K9N/nzy0w0NpEd1AThcsGWrDbwO9xVozJHWpr4vRMjIlwEo/TTqtPe74z4GYmmuLR49PcsrsTh922Pl9shYiQCIIg1MlsLE2H34PHaa9wbT13bp5r9PgIaAJSq/X7ULuPR4/PMLaQoL+1CZtNE4zOgIdkJk8ykzKD8psBERJBEIQ6mYulzSpzq2trKpxkPJTk6m1tJev/y5v3kcuXtkQBLWaSyOQ4MhamV7dgQIuRGNyySQLtIEIiCIJQN4aQpLK5kvRfI9axrb00FnLz7upiYGRunZqJcdVQMZZiFCXu7g7Q39pU9bUbEQm2C4Ig1MlsNE2H303A4yJq6f5rZHN1BOrriTXYXhSJ3paiRWIIyWZya0GDLRJFUe4CPg84gHtVVb2nxrpfBr4JXKeqav0TZQRBEFYRwyLJFwolri0jm6ujzuaK1iyuPouQ7Ozyc8el3bzzusFV2vH60DCLRFEUB/AF4A3APuDdiqLsq7KuGfgPwFPru0NBEIQiiXSORCZHe8BNwOsscW3N6v2x6u3SG/A4adMztfpaitaJ1+Xg3vddxyW9G7ttfDmNdG1dD5xQVfWUqqpp4GvA3VXW/SnwGSC5npsTBEGwMhvTxKLD78bvKRWSuVgat8NOwFO/k2dIj6dYXVublUYKyQAwYrk/qj9moijKAWBIVdXvrefGBEEQyjHiIO1+DwG3k3Q2TzqrdQKejaXpCLjNNN56MKrd+0RI1g5FUezAnwO/1+i9CIIgzJpC4iLg1SwPI05iTQuuF6UnSJvPteGHVtVDI4XkPDBkuT+oP2bQDFwG/FRRlDPAjcD9iqJcu14bFARBMFiIa0LS5tNcW1DsADwbTS1bEH7r1Tv5we/etiwrZqPSyKytg8AeRVF2oAnIu4D3GE+qqhoCzCRsRVF+CnxcsrYEQWgERiV7s9dFc7mQxNLs7Aos6/28Lgdel2N1N9kgGmaRqKqaBT4KPAi8AnxdVdXDiqL8iaIob23UvgRBEKpRFBKnaZGsxLV1MdHQOhJVVR8AHih77FM11v7CeuxJEAShGuFkBrfTjtflMIUkksqSSOeIp3NbWkg2bLBdEARhIxFJZgnqQfZmS7Ddmha8VREhEQRBqINIMmuOvTWD7cmspT2Kp+ZrL3ZESARBEOognMiYlkjAEmwvpgWLRSIIgiAsQiRZFBK/W8u2iqayzEaX12frYkSERBAEoQ4iySzNHs215XTY8brsxFJZ5vQYSXudnX8vRkRIBEEQavBbX32Gbz+v1UlrMZJioqvRSn5W77PVvIw+WxcbIiSCIAhVSGVzPHh4ksdPzACGa6s4ez3gcRBNZZmLajUkF0OF+oWydSVUEARhEabCmstqNpYmm8sTS+dKLRKvk1gqSyy1tQPtIEIiCIJQlcmwNrliNpoyW6EEm4oWSZvPzeGxEE0uh9kSfqsiri1BELYkX3z8NC+MLNR8flK3SGai6ZL2KAYfe91eYqkcZ2bjWzpjC0RIBEHYghQKBT79wCv81cPHa66ZMCySWIpwMgNgVrYDXL2tjX/40A0EvU6GO/1ru+ENjri2BEHYcsTTOTK5Ao+fnCGZyVXtwmu4tpKZvHnbGmwHuGqolSf+4HaaLpIuvheKWCSCIGw5FhKahZHM5Hny1GzVNYZ4AJyeiQOlri0Dv8eJ3b51M7ZAhEQQhC1IKJ4xb//k6FTVNROhopCcmYkBlRaJoCFCIgjClmMhobU1aWly8RN1mkKhULFmMpxkuEPLxjozawiJRAOqIUIiCMKWw7BI3nxFH+fm4pycjpU8XygUmAgn2dcfBODUtAjJYoiQCIKw5QjpMZI3XNYHUJEGHE5mSWby7O9vAWAslMDttONxbu2gei1ESARB2HIYwfa9vdqc9dloquR5I9A+1O6j2eOkUICgxEdqIkIiCMKWI5TI4HLY6Ap4cDvt5nAqAyPQ3hv00qF39Q2KW6smIiSCIGw5FuIZWpq0Rosdfrc5nMrAKEbUhESbfCjxkdrU/MkoitK+2AtVVZ1b/e0IgiCsPeFEhpYm7fhr97srLJIpXUi6gx6z/Ymk/tZmMYl9FigA1SptCsDOlX64oih3AZ8HHMC9qqreU/b8x4APAVlgGvigqqpnV/q5giBsbRYSaVp9mkC017BIWn0uvC6HWCR1UPMno6rqjrX8YEVRHMAXgNcBo8BBRVHuV1X1iGXZ88C1qqrGFUX5HeB/Ar+ylvsSBOHiZyGeoSfoBTQhMepEDCZCKXr15zsDhkUiQlKLun4yiqK0AXsAr/GYqqqPrPCzrwdOqKp6Sv+MrwF3A6aQqKr6E8v6J4FfXeFnCoKwRUmkcyQyOdr9bkKJDEpPM6C7tqKlFsmp6Si7u7WMrk7TIhHXVi2WDLYrivIh4BHgQeC/6n//8Sp89gAwYrk/qj9Wi98Avr8KnysIwkXG0Ykw9x08V/W5fL7A3/7sJLd85se8/i8eoVAoEIpnzNkiHX43sXSOZCYHQDiZ4dRMjCsGtRqSDrFIlqSerK3/AFwHnFVV9TXA1UDtJv5rgKIovwpcC/zZen6uIAibg3966hz/5TuHqz734OEJ7vn+UdxOO9ORFGOhJJFUllafJiTtfs3iMALuL58PAXD5YCsAHX6xSJaiHiFJqqqaBFAUxaOq6lFAWYXPPg8MWe4P6o+VoCjKHcAfAm9VVTVV/rwgCMJCPEM6myeVzVU8d34hAcCn3rwPgBfOadfBrU2GkGgWhyEkh0Z1IRnQLJKuZk/JeqGSemy1UUVRWoFvAw8pijIPrEbm1EFgj6IoO9AE5F3Ae6wLFEW5Gvg/wF2qqlZv0SkIwpbHaHkSSWbxBErbmMzG0rgcNq4c0iyM587NA9CiWySG68rI3HppNMRQe5MpMLu6/Pz5O6/krst61/6LbFKWFBJVVX9Rv/nHiqL8BGgBfrDSD1ZVNasoykfRYi4O4P+pqnpYUZQ/AZ5RVfV+NFdWAPiGoigA51RVfetKP1sQhIsLq5AYwXGD2WiKNp+bvhYvPrfDFJLWpmL6L8BcTHN4HDq/wBUDrebrbTYbv3RgcM2/w2ZmSSFRFGWb5e5p/e9eoHpkaxmoqvoA8EDZY5+y3L5jpZ8hCMLFT1FIMhXPzcXSdAQ82Gw2dnT6OXw+DFASbAeYjaaZj6UZmUvw3hu2r9POLw7qcW19j2JhohfYAajA/jXclyAIQt1YLZJyZmNpUyx2dgU4PKYJiRFsD3pdOOw25mJpXtID7Vfo8RGhPupxbV1uva8oygHgI2u2I0EQajIfS2O322iRwK9JoVAwhSScqG6RDLVpA6p2dfnNx42fod1uo82ntUk5NKoF4veLkCyLZTdtVFX1OeCGNdiLIAhL8JF/fI7//K2XGr2NhpPK5rjn+0eZiaaIpXPk8tqEw6oWSTRtxkF2dgXMx61ibDRufPzELJf0NotQL5N6YiQfs9y1AweAsTXbkSAINTk3F2c+nl564UXOz0/M8rc/O8mOTh+37OkyHw+XxUhS2RzRVNZsc7KzU7NIAh4nLkfxOrrd72ZkLs7J6SgfvGVNu0NdlNQTI2m23M6ixUz+ZW22IwhCLQqFAjPRFIlMZa3EVuOJU7MATEdS5thcqLRIjNoQo+hwp+7aKrc42gNu8z1fvbcLYXnUEyP5r+uxEUEQFieezpHK5kll06SyuS099vWJkxYhSdQWktmoISSaReJzO+lr8VYIiRGM97sdXLt90QkaQhXqcW3tBT4ODFvXq6r62rXbliAI5VhnZkyFUwy1+xq4m8YRSmQ4PKZlV01Hy4Wk1LVlFBkari2A63e0Y7eVTscwhObm3Z24nTLvb7nU49r6BvC3wL2A2NSC0CCsMzOmIsktKyRPn54jXwCf26FbJNrPxed2VHFtaUWGhlAA/MWvXIWtTEgMi+QXlO613PpFSz1CklVV9W/WfCeCICyKcSgCTIYvrO1cKJFhLpZmR6d/6cUblCdOzuJx2rl1TydHJyKmRTLU5iOSKrNIdNeW0XgRqBARgCsGWxlsa+KOS0VILoR6hOTfFEX5CPCvgPnbK6N2BWF9mbXMzJjUR8Eul//1oMrDr0zy8z+4fbW2te78/OQM1w63Mdjm49HjM4QSGRx2Gz0tXhbKMtrmYmmcdhvBpsWPuiuHWnnsE+Ktv1DqEZL36X//vuWxVRm1KwhC/RgxErtNGwV7IRw6H2I6mqJQKFS9Mt/ojMzFOToR4RN3XYLNpiUgjC8kCXqdBL1ORubiJeuNGpLN+F03E/VkbUlStSBsAGZjadxOO10BD1MX4NrK5wscm4iQyRVIZfN4XZsv6+vfDmklbG++oo+nT2tOkRPTUVqaXDR7XVWD7db4iLA21Dtq9yYqs7a+skZ7EgShCrNRrWdUT9BzQa6tc3NxswYlmspuSiG5/4UxDmxrZajdZ85ZPzEVZU93gKDXSbhKsL0jIEKy1tQzaverwP8CbkGblHgd2rRCQRDWEeNQ7G3xXpCQHJ0Im7djqcpWIo0gm8sTT9e3l2OTEY5ORLj7Km0itzFwKp7OEWxy0ex1Vgy3moulSwLtwtpQj0VyLbBPVdXCWm9GEITazMXStPs9dDd7efTYzLJff3QiYt6u1pOqEfzNT0/yr8+f58cf/4Ul197/whh2G7zx8j4AuixzRwzXFpQOt7L22RLWjnoqb15Gmz8iCEIDMdqh9wS9RFLZZVsVqkVIoiu0SPL5gtkocSUcnYxwrixAXovHT85w7fZ20xJp87lx2rUgeotukUBRJFPZHJFU1qwREdaOeoSkEziiKMqDiqLcr//5zlpvTBCEUub0wHFvi3aQLte9dXQiwkBrEwDRRSySH7w8wR99+yUKhdpC8bkfHeNNf/nosj6/GtPhFNl8gXQ2X/Hcy+dDvOHzj5qNGCdDpUWYdrvNnIZYapFo6+dj2t8dAXFtrTX1CMkfA28DPg38Odqs9d1ruCdhhSQzubr9zsLmIJHOEU/naPe76Wn2AssrSkykc5yZjXHN9jZgcYvkH548yz88eY7nRxbMxzK5PP/rQZXpSIpMLs8/PXWO0zOxC/w2RaYiSXN/5bx0PsQr42GOT0bJ5wtMRVL0BEtFwbBOWn0ugmUWyfkFzdLpa/GueJ/C4iwpJKqq/gwIA28GvgS8Fq1lirBB+W/fO8Kv//3Tjd6GsIrM6lXtHX433UHtYDQO4Xo4PhWhUIBrhzUhidQQkmwuz/P6TPOvPnHWfPz5cwv89U9O8Nkfqjx2YobZWJpUNl+Xe+u+g+f45rOjFY8XCpo4AMQzlfuJ6+IyHkowG0uTzRfoLRMFQ0iqWSTHJ6MA7O4OIKwtNYPterPGd+t/ZoD7AJuqqq9Zp70JF8jR8QjnFxKN3oawihTbobvNw3QiVL+QGIH2A9s0IakVXzk6ESGWzjHQ2sT3Do3zn994KV3NHjPV9hvPjnJqumiJxNNZ8wCvxRcfP0Mqm+ft1wyWPB5NZU2xiKUqLZKEblWPLyRNN153c5mQlLi2tOPMSAE+PhXF67Kb7jxh7VjMIjmKZn28WVXVW1RV/StWuWmjoih3KYqiKopyQlGUT1Z53qMoyn36808pijK8mp9/sTIeSm6Y9E5h+UxHUnz6gVe4+Z4fm11ujYaNHQE3AY+TZo+T8WUIycnpKC6HjUt6m7HbasdInj2rWSOf/qXLSefyfO3pcwCcnY3hsNtw2m08fWaOgEc7tONVXFLlzMXSnJ6JlXTpBUxrBKq7toyal7FQwrS+yl1b3fr9YJOLoCVrC7T6kt3dAex2qWpfaxYTkl8CxoGfKIryfxVFuR1YtX8RRVEcwBeANwD7gHcrirKvbNlvAPOqqu4GPgd8ZrU+/2Ilny8wGU7W9R9c2Hhkc3ne+tePce+jpxgLJfj+SxMAzEVLBzQNtvsq2oEsxqnpGMMdfpwOOwGPs2aM5Jmz8/QEPdy2p5Nrt7fxo1cmATgzG2ewrYlfu3E7AHdf1Q8sLSSFQsG0pl4aDZU8Z63OrxbTM957IpRkIqSt7QnWdm0FzBiJJlgnpqLs7hK31npQU0hUVf22qqrvAi4BfgL8LtCtKMrfKIpy5yp89vXACVVVT6mqmga+BtxdtuZu4Mv67W8CtyuKIpcXizAT1bJgamXCCBub0zMxxkNJPv2Ll3P1UCuPn9TqRayuLYDBtiZG5pcjJFFzOqDWSqSGRXJmjmuH27HZbFw51Io6GSGXL3B2Nsb2Dj+/d6fCvb9+Lbfq422XsnzDiSxZPY7y4uhCyXPWGE81QTKslLGQ5tqy2YrCYXDTrk5u29vFjk4/DrsNv9tBOJElmspyfiHBnp7mivcVVp96gu0xVVX/SVXVtwCDwPPAJ1bhsweAEcv9Uf2xqmtUVc0CIaBjFT77osXq7pDMrc2Btbr7yLhWfX7lUCs37+7k0GiISDLDbCyNy2EzM5OG2nyMzicWTdG1vv+5uTg79atzzSLJVKwbW0gwFkpyrZ7ZdWlfkGQmz+mZGGdn4gx3+GhyO7hjXw9+j1bwt9TY3xlL6/tDZUIyHbFaJJXvYwbbFzTXVoffUzJnHbRA+lc+eD0+t/ZzMfptnZySQPt6sqxRYKqqzquq+neqqm7eHtQXOeOhYpA9Ju6tTcHvfeNF3vj5RykUChwZD+N22NnVFeBVuzrI5Qs8dWqOn5+cYajdZ3axHWpvIp7OlUxNrMXIfIJMrsBOfQaJ3+Oo6tp6/px20Bspwvv6goDWtj2SyrK9ozjDxDi4l7JIjP11+N0cKnNtWetgFnNtTUdTjM4nKuIj1Wj2Ookks5wQIVlXGjlT8jwwZLk/qD9WdY2iKE6gBZhdl91tUkosEgm4b3h+dmya77wwxpnZOCemorwyHmF3dwC3086BbW14nHY+84OjHBoN8Tuv3mW+brBNK8wbnV86O+/UtHaomhaJ10W0SpaUkWLcr2c57e4O4HLYeOClcQCGO4rFgD63bpEscbEyG9Xe89VKF+OhZIk7ayqSWjRon9BTggsFraakN7h0PUiwycW5uTjHJiO4HDa2b9EpkutNI4XkILBHUZQdiqK4gXcB95etuZ/iPJS3Az+Wnl+LYxUSsUg2NslMjv/vOy+bB+RjJ2Y4MhZmX79mCXhdDq4dbuO43t32lw4U02eH2rXDvp44iZGuu8uIkXicRJOVri0jBdevWxtup53d3c1mu3arRWKsWep3zMg2e40+wvbQSNEqmQqn2K6LUy3XlkPPuFqIZ8z6mcV4+zWDHBkP85UnzrKzM4DTIfPX14OG/ZT1mMdHgQeBV4Cvq6p6WFGUP1EU5a36sr8HOhRFOQF8DKhIERZKkRjJ5uF7h8Y5MxvnM2+/guEOH99+YYyZaIpLdZcSwM27OwH4+OsV81CFokUyMre0RXJyOkq7302rTwvU18raiqez2GzgdRWPhUv7mskXwGYrihdAk2mRLP47Zkx1vG1PF3YbvDxmEZJIkm3tPmy26u+TSOcYait+Zj2urXddN8TdV/WTyOTErbWO1DWPZK1QVfUB4IGyxz5luZ0E3rHe+9rMjC8kzIMiXsV9IWwczs3Fsdngpl0d3Ly7k398SqvZ2GcRkl+7cTvb2/3cua+n5LUBj5M2n4vR+TjRVJbvHRrD73Gyry9ourBajt5uAAAgAElEQVQMTk3HzPgIQMDrrFpHEkvl8LudJdME9/UF+Rbn6W9pwuMszi8xgu1LWSRzsTRBr5MWn4t2v7ukdmQqkuLWPV00uRxV3yeRybG7K8CZWc3qqse1ZbPZ+PQvXs5cLM2d+3uWXC+sDg0VEmH1GQ8l2dXl58XREDGxSDY0WiaSG5fDzi01hKTZ6+JNV/RVff1gm4+R+QR/98gp/vLh44BWT/H0H95ecuifmony2ku6zfsBj5NYOkcuXyixcuLprBn7MDD2ssMiRABep8McdbsYM9GU2TSxzedmXnd1JdI5Isks3UEPPrezpmurO+gxL4zKa0hq4fc4+epv3FDXWmF1EAfiRYRRjLhLvyKVosSNzVQ4Zbb8eNWuDmw26G/x0uJbvOWIwVB7EyNzcb713Civ2tnBZ99xJaFEhp+q0+aaUCLDTDRdYqUYAe7yC41YOoffU3ptabjZtneUBq3tdhtNLseSCR1zeut70ITEyOIygu7dzV58bkdN11aTy2k2Xeyuw7UlNAYRkosIoxhxl+4bljYpG5vJSNL0+7f63Ny4o4MbdtZfJjXY5uP0TIzR+QS/oscGOvxu7n9hzFxzVu+RNdxR6tqCyt+PeKrSImnzu/nY6/byjmuHKMfndi4dbLcMlmrzu5iPG0Kiubi6mz343JWurUKhYFpIfXoWWb0WibD+iGvrIsIItBvZOUulZgqNZTKc4rL+FvP+lz54HXZb/Y0bjEC03+3gzv09OB123nRFH/cdHCGayhLwOM1W8/2txUPYsEiiyayWUK8TS2fNbCwr//72PVU/v5YlYWU2luaAXpfS7nfznF6rYrRH0Vxbjorf1VQ2T76gBfX7gl5cDhvtPhlQtVERi+QiwihGHGzz4XbaJf13A5PN5ZmJpkpSWj1OR0Xl9mIM6jUSb7y8zywQvPuqflLZPD88rPXosrqQDMyeVOUWSTqHz1NqkSxGuSUxFUly0/94mP943wucmNJmiMzHi66tVj1GUigUSrr5ajGS0r0YwuJzO3jfTcP897ddLs0XNzAiJBcRhkXS1+LF73ZI+u8GZiaaplCoL6W1FpcPtLC3J8D7bho2HzuwrY2B1iYe1IVkMpzCZoPOQPFqvtlqkViIpapbJLUotySeP7fAWCjJ/S+O8aa/fJQj42Fy+QId+me3+9xk8wUiqSxTkRQuh402n4smt6MinhfXW680uRzs6w/yzusqXWvCxkGE5CLCOBiCTS7Nfy3pvw2hnoFTteZrLIfOgIcf/sdXc9lA0T9ls9nY1x/kzIyWMjsVTtIZ8JQU5hkWSXktSTydq4iRLIbf4ywJ2B+f1Gae/Mvv3EQqmzez0MysLd0yWYhlmIok6W72YrPZ9Iue0t9VQ6CalrEfoXGIkFxExDM5XA4bLocdn1gkDeHwWIgbPv0wjx2fWXSdEWxeiUVSi8G2Jkbn46YLqfwzDKujqkXiuXCLRJ2MMtDaxFVDrSg9zdz/gtbxqJi1pWWjzcXTTEdSZiffpirpv0XXloRxNwMiJBcRiXQOr0u7gvN5qufmC2vLo8dnKBTgZ8emFl1nWCRrkYk00NpELJ0jlMgwGU6ZM94NmqtYJFqW1PIsEi1rq9QiUXq1tu137u8x4yeGa8uwSOZjaT312aO/T2XQ3rgIWs5+hMYhQnIRkcwUDwKJkawN2dziM16eOqX1FH1K709Vi6lwEruteLW+mlgbOk5FUhX1F4bVYRWSdC5PNl+4YIskk8tzajrGnh4t9fzOfb3mOiP918i6moulmYwkzX353Q7imVxJS3wzRiJCsikQIbmISGRyNBkWicRIVp1zs3H2ferBikl/Brl8gWfOzOO023j5fKjmFELQguDlsYvVYlBPCz4zG2M2lqqIw7gcdrwue8n+jHY6y7NIHObv2NnZGOlcHkUfJHXZQJB+vZDQEJA2/e/JSJKFeMa0lJrcTgoFSGaKIm3N2hI2PiIkFxElri2xSFadVybCpHN5XjpfXUheGQ8TSWX5pQMD5AvwzJnaVolWjLg2BXaGkLxwbkHPDKv8nICndEqi4aJaXtaWk0QmRz5f4Nik1qp+ry4kNpuNt109wM5OvymWzV4nDruNYxNaUN6wSAyxsP6+Gm5Z48JI2NiIkFxEJDI50xXg91RvhCdcOGMLWp3O+YXqrdsNd9Zvv3oXTrvNbL9ejclwak0C7aDPL/c4efbcPFA9oN/sLe0AbBzcy3VtgfZ7p05EsNlKB0n93p0KD/yHW837druW7qvqomNYSkUhKf6+GjETcW1tDkRILiKSZa4tqWxfXYw6nbGF6um9T52aZVu7j51dAa4YbCmJk9x38BzfPVRsXTIdSdY1X+NCsNlsDLQ2cfi8Nra3ukXiJJQoziQx2qUsqyDR0rPr+FSE7e0+0yIGcNhtJfdBK0o0xuB2mcH2yuFWxghfydraHIiQXERYYyR+t4NYOlvXTO+LmclwknR28QB5vZgWSZWphIVCgYNn5rh+RzsA1+/o4NDoAkn9QPw/j5zizx86BmiB6Zlo2sxaWgsG25pI64kB1Zod7uj0m3UfYLFIluPachWnJB6bjJpurcVo97kr9iWurc2PCMlFRCKdw+supv+WBzC3Gqlsjts/+zP++elzq/J+hkVyfqFSSBKZHPPxjNl5WekNkMkVzLWz0TSnpmNMhJJMmzUka9eEcECPk2iZYZVCcsVgC+OhJDP6KFzTIllWQaK2NpTIcHqmmLG1GG1+rZbEYbeZ+6ru2srhcdpL2twLGxcRkosIre120SKBylbhW4nxhSTRVLZkauRKMCySiXCyIg3YCFwbNRqdejX3TCRFOps33Ug/PznD4THN5TRomf632hjv3dXsqXoYG9XwRuLAhcVItLXqRIRcvsDOzjqERM/c6gy4zX0Z7xNLZbn9sz/lK0+cWXZNi9BYREguIhKWOpIm/T/nVo6TGNbAarTTz+byTOrtRnL5ApOWSX+wiJBE02brdIDHT8xy38EROgMeblxGy/jlMtCq1ZLUsnr263PhX9ZTmYtZW8tL/wVMYRwuG35VDaMo0ZqSbATUj09FOTkd45kz88QtF0XCxkeE5CKiPEYCW9siWU0hmYqkyBfgumGtJXp5nCSS1CyOSiFJme4jv9vBj49O8hN1irdfM7isTr/LxbBIavXyava62Nnp55BhkRh1JBdgkRzRhWRnHUJi1JRY40OGi+ygni59bi5OIpOVjK1NhAjJRUI+XyCZyZe0SAFWpSjxL350jH99fnTF77PeGIf9YoWB9WK06L9Gn61RngJctEi0GEC7343dpgnJbFSzSO7c38t8PEMuX+Bda9zN1oiRLJZifPlgCy+fL7VIlmMFFC2SEK0+l2ltLEar3m/LmrHmc2m/q8+e1dKVR+biumtLMrY2CyIkFwkpPTOpyV1qkaxGUeI/PnWOB16aWPH7rDdGTGM1rLLzesrvtcPt+nuXxl0MsTKGRjnsNtr9Hk1IYppF8pYrtdnrr9rZUZcbaCV0+N1cu73NzCKrxuUDxYC74UpaTnDbSBWOpXMVM91r0e6vtEiM31lDjGdjaWaiKbFINhENkXxFUdqB+4Bh4AzwTlVV58vWXAX8DRAEcsB/V1X1vvXd6eYhkSlNlywGMFdmkeTzBeZi6WUJUjKT429+epIP3ryj7vnja8FYyLBIlv4Z5PMFvvfSOHdd1lvV5TSui9KuLj/tfjejS7i2QAsoT0fSpkVyzfZ2fuu2nbz+sl7WGpvNxjd/56ZF11gD7lrn3+Ud3FaLYUdHfUJixkgslpLbacflsJHJFfS09RzHJ6O8atfaxZCE1aVRFskngYdVVd0DPKzfLycO/LqqqvuBu4C/UBSldR33uKkoFxLjUEhkVnY1Ph9Pk8sX6jqMDZ47O8/nHz7O//j+Kyv67JViuLbqiZE8fnKGf/fPz/Mzdbrq8+OhJM0eJ81eFwOtTRUpwOWuLdAyprQYSRq3w07Q6+QP3ngpB7a1XehXWlX29wex2bSAeyyVXbYryeoGq9ci2d0d4Nrtbdywo1QkjPd6zSXdgGZhS9bW5qFRQnI38GX99peBt5UvUFX1mKqqx/XbY8AU0LVuO9xkGNlZXjNrS3c76AKQzOT46hNnCCczVV9fixn9ano5AWvjM752cIRnz1a2Ccnm8mteKJnPFxjT037r2fshPXvJsGLKGVtI0KfPPe9v9ZpuMwNDSAIeq0Wiu7aiKToCbmzLmMe+HhiieHwqSuwC0m0ddpspADu66hOSoNfFN3/nppJWKlC0bu6yWGtNLomRbBYaJSQ9qqqO67cngJ7FFiuKcj3gBk6u9cY2K4mySmC/2XZCO+C+/PMz/JfvHOY3vnRwWSnB5QVr9RBO6FfnHid/9O3DJaKRzOS44dMP841n1jZ4PxPT6jfcDntdwfYXRxYAmKhRczIWStDXogWwB1p9jM7H+cuHj/OPT50FNCHxu0tjDJ0Bt5m11RFY/Xbxq8Hu7gAnpqLE08sbamVgiM9wna6tmu+jW9A37uwgqLsHm9wSwt0srNm/lKIoP1IU5eUqf+62rlNVtQDUvDxVFKUP+CrwAVVVt26Z9hIUexPpFomraJGks3m++PgZhtqbePbsPB/5x2frtgguSEh0i+QDNw/zynjYtGpAy8iZjaX56RKDn1aKEQzf2eUnllq6VYxRmDcRri4k4wtJ+nWLZFt7E8lMnj9/6Bif+f5RQIuRWN1aoFkkyUyes3PxqtXlG4FdXQFOzUSJJrMX5EoyBKBe11bN93E76Al66Ax42Nbh0x8Ti2SzsGb/Uqqq3lHrOUVRJhVF6VNVdVwXiqqniqIoQeB7wB+qqvrkGm21hLOzMT76T8/z8dcrvHrv5vGkGUJipP/a7Tazlfx3D40xEU7ypQ9cx4sjIT73o2OMhZIMtC5dWW2084iltcFD9bhnIsksNhts169SI8mM2aDv9EwMgOfPLSz/Sy4Dw/W0t6eZoxMRkpl8zSygqUjSrH6vZpEk0jlmY2n6dYvkl68ZpD3g4Zkzc3zlibNkcnmiqWxJoB2KtSRnZmJcNbQxw3u7uwMkM3lOTse4ZXfnsl/vcznpCXouyJqxcsVgK8Zv1rZ2Hy+fD0tB4iaiUbbj/cD79NvvA75TvkBRFDfwr8BXVFX95npt7KXzIV46H+IDX3ya//fY6fX62CUZDyW472DtnlHlri3QrugOj4X5wk9OoPQ08+q9XVyrF9Sd0Q/0pTCsiVy+YKYYL0U4mSHgdtLSpF2hW+denJ2N698nWdONtBoYgfa9ev+nxdxbRi1FV7OnqkVyakbrVrtT76PV7HXx1iv7zSaF87E0kWSWQLmQ6OKZLxRFZaNhxCqiqeyyOv8atPvddTVrXIpP/+Ll/PdfvByAoXbDIhEh2Sw0SkjuAV6nKMpx4A79PoqiXKsoyr36mncCtwHvVxTlBf3PVWu9sfmYdnBeN9zOn3z3yJoedsvha0+P8Il/eYmpGq6XZJXRpK0+Fz8/OcvIXIKPv17BZrOZLojTdQtJsRVIvYV9kaR2dW5coVuF5Mxs8XNfGJmveO1qcX4hQcDjNOMai6UvHxoNYbPBa5VuJqv8e5+c1va8q7vUfdOpxz1moukarq1iXGQtRuquBkaTSVhe51+DP3vHFfzPt1+xmltimwjJpqMhTkhVVWeB26s8/gzwIf32PwD/sM5bYy6m+fc/9ZZ9vOkvH+PHR6d4zw3b1uWz//dPT3D4fJi/fs/VFS4kw1VzYipadY5FefovwF+/52rmYmkObGszXV69QS9el/2ChCSeysHSffmIJDMEm1wEdYvEmil2ZjbGvr4gJ6aiPH9ugbsu66trH8vl/EKCgdamqvPJy3lpNMSurgC7uv3E0rkKUTg5FcVmqwwot+txj9lYikgqy6B+ABp0WayQjg1qkbT73bT73czF0hdkkRjz4VcTQ0iaJEayaZC0iDLm42mCXif7+oIMtjXx46OTADx9eo6RueqT8VaLh45M8r2Xxnnk+EzFc4YP/+R0tOprq7m2LukNctOuzpLhQna7jeEO/zJcW8u3SMKJcovEIiQzcfb2BNg/EFzTOMl4SEvXDSzRKqZQKHDofIgrBlvMBoflVuipmRiDbU0VQ5qMTKzZqObaai6LE7T73RjXAxs1awtgt26VXIhFshbs729hqL2JS3pX7jIT1gcRkjLmYmn9ALBxx6U9PHZihiNjYd5775N8+CvPkM+vXv3DCyMLHPjTh8xD3fj7zx86VpFlZNQ3nJiqISRGsL2OlMkdnf76LZJIml79gK231UgkpV3RG1f1hmsrlc0xFkqwvcPP1UNtHDq/QCa3Nol487EMHX5PsY1HDRF86vQc05EUN+7sML9neZzk5FS0xAVk0GlaJIZrq/QgdjrsxbbpGzRrC4ouu43iSmr3u3n0P73WrLwXNj4iJGXMxdJmG4fXXtJNMpPng186SC5f4OhEhB8cXr2eU3/705PMxdI8fWaO+Via+XiGS3qbeXFkgR8fLSayFQoFxhcMi6S6ACTSOew2cNfRUXZHp59zc/GKmRrlFAoFZmMptuvpmMuxSIJep2kNhHUhGZmLUyhon3/VtlaSmTzqRGSxt7pgQomMObt8sb1/8fHTtPm04HlvS6VFks8XOD0TqzprI9jkxGm3MRlOkszkK2IkUIyTdDZvXIvEEMmVZl4JWxcRkjLmYmmz1fUNO9vxuR1MhJN84q5L2N0d4HMPHSO3ClbJyFycHx7RREmdiHBaD0L/7h172dHp5w++9ZLZhiOUyJDI5LDZFrdIfG5nXem5Ozr9ZPOFin5R5YQSGTK5ghkbiNfZJsWIMTjsNgIep+naOjOjuQa3d/jYpVdCj86vvrvQSMdt9bnMw7GaRXJuNs4Pj0zynhu24XU5TNfWpMUiGQ8nSWRyFYF20PpZtfvdpiUZqHIQG9la7Rs02A7FzK2NYpEImw8RkjLm42nzP73H6eD1+3vZ2xPgAzfv4Hfv2MPxqSgPHZlc8ed85Ykz2Gw2BlqbUCci5mG0pyfA3/7qNSTSOT7wxacJJTJmcd3+/iAT4WRJzMEgkclV+PBrUW/mlhEfMQrE6ilKLBQKZtYWQNDrNCvdjYyt4Q5/yeCn1SasTyNsaXIRcNe2SL78xBkcNhu/duMwoNXgtPlcjIeShJPa+NiTunBXc22BFkQ3UprLXVugCUmz14nHuXEP6QPb23iN0rVheoAJmw8REguFQsGMkRj82duv4P6P3oLbaeeu/b24HDYOjVYPEifSWhV5Nf7nD47y2199FoB0Ns/XDo7whst6uXFnB+pkhNMzMRx2G0NtPpTeZv7Pr13Dscko//TUOXMWxq17tALJU1XcW8l0ru6WEoaQnFpCSKYj2iFvWCT1uLYSmRzZfMF08zR7XUWLZDZGS5M2t8KIHcwuISSHx0LL7su1YBESv6e055iVB14a5/ZLu02XFmgTBSfDST7yD8/xhs8/wsOvaBcNO2v0kuoMuDk7p/0cq7m2fuW6If79a/csa//rTdDr4osfuN6s3xCE5SJCYiGRyZHK5ksG9DgddvNK3+mwM9TmK6mFsPIrf/cEn36gesfbI+Nhsw3HbCxFJJnlpl2dXNLbzHQkxbNn5xlsa8Lt1P5JbtrdyfYOHy+MzJvNB2/VK4+rubes0xGXot3vJuh1Lpm5ZVgk25dhkRiB9WCTdnXe7HWaj52djTOsv5fbaaelyWXO6qjGCyMLvOkvH+OZs8urNzHmo7f4XDgddjxOe0WiwFRYq2a/vqwLbW+Ll6dPz/HYiRmSmTxffuIszV5nSSqvlXa/m2Qmb37Xcm7e3cmHb9u5rP0LwmZDhMTCnF6MaMRIqrG9w2f6+ss5PRPj5ycrU3dBO4SNK/qoZb63oqc4PnV6rqJf0RWDrRwaDTG+kMBpt3HNcBtOu61qCvByhMQoTKzXtdXX4tWaH9aRtVWcy2FYJE4iKe2xkbl4yVVvR8DNbKy2RXJ8UgvEjy+zKDRksUhAi12UW1Mv6t1+rxoqzQzqDXoJJ7Vkgf/2tssAraK9VuzJ2kOrmpAIwlZAhMSCKSSLBEaHO/2cmY1VuFtyeS02cHwqWtUFFE3liOrNA40spoDXaebK5/KFCiG5clCbYPfS+RA9QS8ep4PhTn91iyRdf4zE+j0WYyaawmG30eZz4/c46gq2hxJFkdT+dhFJat97MpwyU2xBq/aejda2SEbmi8kGyyEULxUSv8dZYU29OLKAw25jX1+ZkOhurl9/1TDvvWEbH751B++8drDmZ1nrQ6q5tgRhKyBCYsEQksVmTw93+Imnc0yXHYBGgLdQKPZushJLZcnlCyQyOVNogl4nXc0ec451NYsE4ImTs/TpB9zurkBVIUlmcssaTdrb4mUqnFq0LmYmosWLtAaQlYdxNQyLJKgfqsEmJ+FEhmgqSyKTK5mM1+H3LBojMQpAw8sVEn1962JCMrqA0tNc8TM7sK2N4Q4f7795GJvNxh++aR/vvWF7zc+ytkGplrUlCFsBERIL8/GlLRIjXmBk6hhYr5qrBeONXk+RZNY8bAMeFzabDUVvelcuJJcNBLHbIJsv0Kd36r20L8jp2VjFwbgc1xZAX9BLOpdnLl77IJ+KJM3Z2tXcQ9UwYyRlFslkWBPeHqtFsoRryxCSZVsk+vqg6dpylOw9ny/w4sgCV1bpyHvb3i5++vuvqbvJYru4tgRBhMSK0WdrsRhJrdRZ62Fn+N+tGAeZJiSl7h/DvVUuJD630+ys2q9bJPv7gxQKcHQiXLI2nl6uRaIJ02JNKcdDSbPpod/jqKuyPVwlRpLNFzinZzYZ7eRBS52dj6drFkaO6DUmC4uIXTUW4hn8boc5e12zSIpuuTOzMcLJbEV85EIwXFtuS1KGIGw1REgszMfSOOy2Ra8sB1qbcNptnJ2tLiTdzZ4KiySby5uZPZFkpiTYDnD31QO845pBc96FlSsGtcPOcG3tHwgCcHisVEiSy7RIqlVxl6MJibbOOIwLhQKPn5ipmZJbmbWlCcrJKe3nZbVIOgNuCgWYj1daHMlMzrRiLsQiabVcDPjL3HIv6v8+1SyS5WK0PilvIS8IWwkREgtz8TRtPi0mUAunw85gW1NF5pZx2N2yp5ORuYQZbwGIZ4pXw9GU5tqy2YpN8g5sa+PP3nFl1c814iSGa6s36KXN5+Lw+VIhSaSX6drSBWK8Rlv6eDpLKJExBcc4jJ88Ncd7732KJ09VzmIHTSits7wNF5cR1+luLo2RACU/KwNr1f3yhSRturWg0pp6cSREk8thNitcCYZFIm4tYSsjQmJhLpqm3b905k21jCfjsLtNLxq0WiXWq+FIMksklSXgdi4qWAa3X9rNjTvbzapjm83G/v4WDo8X3WeFghbEX45rqzPgwWG3VZ2/AUVLxRgvawSsT0xpKblG+5ZyjM6/RrqsEXQ/MR2lyeUoCUgXu+dWZm4Zbq02n8vMBKsXrc9W8XPKXVtjCwmG2ptw1tGXbCl8bgcep12ERNjSiJBYMCySpRju8HN2Nl7i3rFaJKDNuDCwCklUj5HU6wrpa2nia7/5qpLYwv7+IMcmombn3HQuT77Asnz0DruNnmZPzRoNQ0h6g5olZASsz+hJBlOR6q+LJDOmeEDxSv3EVJSeoKekHsMcDFXNItED7ZcNtFxQ1lZrU2k2VSxdnNs+HU2V/DxXgs1mozPgkYwtYUsjQmJhvqw9Si2GO3xEU9mSPlHhRAaP005nwMP2Dh9HxouuJ+vVcFiPkazkCnZff5B0Ls/xSc1dlExrgrLcGdc9LV4mwtUtC0NgrDGSeDpnVsNPhavXf4TLvpsRIwklMnQ3lw7kMlxb1S2SBG6nnd3dgQsKtreUuLacFApaQgJo9TGrOfr28oEWLukNrtr7CcJmQy6jLFgbNi7Gdr331NnZmHlla7QtB81iePm8VUgsFkkqSySVWdEV7P5+LQB/eCzEvv5gcTriMru39rV4a7ZxN2Zy9FqEJJsvcEx3bU1HqgtJ+VwO621rDQloBYMOu61qLcm52TiDbU20+dzE0jkyubyZhbUUWrC9VEhA+3fwe5zMRNKrKiR/+2vXrNp7CcJmRCwSnXy+wHw8U5eQDLZp7h5rnKBUSFo4Nxc3U2GjZTESzSK58CroHZ1+mlwOs/Cx2pjdeugNNjEeSpoun6MTYV7/uUc4NxtnbCFBm89lusv8ukiNzGnfubZrK1vi2rIGvcstErteNV+t39bIfJxt7T7zZ1qveyup90uzfm5Ab9wYTWWJ6YWRqykkgrDVESHRSWZz5AuFug6Yfj2DymjvDqVCsq9fc3Mc0VN0rRlDRoxkJa4th93Gzbs7+bdD4yTSObPYcbkWSW+Lh3g6R0Rv3fKn3z2COhnhR69MMhFKmrUmUDr0yG6DqRoWSThROu/c73Zg5BT0BCt/tp0Bd9VW8iNzcYbaikJSb+ZWeZ8tbQ/FcbtG/7DVipEIgtAg15aiKO3AfcAwcAZ4p6qqVVu8KooSBI4A31ZV9aNrtSef28m9v34tN+zsWHKt3+Ok1efi/EIxBXghninWevQXaz1u3NlhxkiCegPD8jjChfDhW3fwK383yb88N8qlfVrR4rItEl0oJkNJnl1I8PiJWUCbTz8eSppFkFDa/mNff5CTU1q/sfJmhuUiabNpw63CyWyFawu0zK3y9N9oKks4mWWgremChcTq2rJOSUzntH+Lzg08Q10QNhuNskg+CTysquoe4GH9fi3+FHhkPTZ1+6U9dccu+luaalok3c1eOgMeDo9pricjRtLX0qS5tlKZFTf4u35HO1cMtnDvo6f4zA9UAAbaKgsaF8MQvtH5BP/jgVcY7vDxliv7OXhmjvFQomROh9UiuW64vaRnmEE+XyCazpq1IwbGd+0pc22B0W+r1Lox7ncGPKaLaqFOIVmIV1okRu+0mWjKnLEiri1BWD0aJSR3A1/Wb38ZeFu1RYqiXAP0AJxeTlUAABJNSURBVD9cp33VTX9rE2OWGEk4kSnxy+/vDxZdW6ksNpsWbJ6Pp0lm8itOF7XZbHz41p2cmY3z/Ll5/urdV9ec4lcLoxPvZx9SOTYZ5Q/eeCk37+pgVp8f31ciJJq143HazWr7cvfWd18ap1DQ2q5bMSyUWhZJebDd6L/VEXAvO0ZSzbVV7I8WM5ttimtLEFaPRglJj6qq4/rtCTSxKEFRFDvwWeDj67mxehlsa+K8Xn2dyxeIpLIlh9f+/iDHp6IkMzli6Rx+t5Nmr9Osz1iNArY3XNbLb962k6988AbecmX/sl9vHOwvnw/zGqWLO/f1cN2OdvP5ajGS4Q6/aVlYU4DDyQx/+t0jXDHYUrEXQ2C7g5UWSWfAQySVJWmp/jeEpdNf7Ixcr2vLSBW21pH43E66mz2cnokzo4vfRp6hLgibjTWLkSiK8iOgt8pTf2i9o6pqQVGUao2bPgI8oKrqqKIoa7HFFdHf6iWSyhJOZsjltO2XCkkLuXyBE1NRYqksPreDZo/LDCyvRgGb02HnP7/x0gt+vcfpoDPgJpLM8l/fehk2m42dnX4zAF5ikegB6+0dPlOArJlbn3voGDPRFH//vmtxlFXsB71OvC47zVW+s1E5PzqfYHe3ZskYri2rRRKq0o+rGtUsEtC6EZydjeF12Wn3u+tOJRYEYWnWTEhUVb2j1nOKokwqitKnquq4oih9wFSVZa8CblUU5SNAAHArihJVVXWxeMq6YWRunZ9PmEFu6+G1TZ8EODqfIJrKEvA4S6rZN8oQpPffNExP0Ms23f1js9m4brid7788URIjMYRvR6efLt0iMWpJCoUC33lhjLdc0W/2BrMy2OZjd3f1KYOGO+7EVLQoJJYBYy6HHZ/bUbdFEk5ofczKLb4dHX4ePjpJR8AtgXZBWGUaVZB4P/A+4B797++UL1BV9b3GbUVR3g9cu1FEBLQuwKD1bTICt1Yh6WkpXrXH0zn8HmfJ4VYekG4UH33tnorHXrevhydPzZrfEbTv9o5rBnnD5X0EvU48TrsZIxkPJZmLpbluuK3qZ3zyDZeQylZvFW8IiXV88Ew0RcDjNGtYWppcdQfb5+Nai5byPmbDnX5momnOzMQl0C4Iq0yjTrN7gK8rivIbwFngnQCKolwL/Laqqh9q0L7qxiokhpvEmnLa6deaIk6EkkRTWfye0oaFG7nt+C9ePcDdVw2UuKjsdht/9o4rzfvdQQ9TevW7URi5f6D6fA+vy1GzD5jf46SvxctJy9TH2Wi6ZIRtS5OrLovk+XPzfOu50ar72NGpWVzHpiK8pXf58SRBEGrTkNNMVdVZ4PYqjz8DVIiIqqpfAr605htbBp0BD26HndGFhDn7wmqR2O02ups9TISTxFJZeoPesmaGG8O1VQ2bzYZjicbE3c1e0yI5PBbGboNLL7Df1O7uACcsFslsLEWHJRgerENIzszEeP8XD9IR8PBX77664nmjrU2hIKm/grDaSMTxArHbbfS1ehlbSNYM8PYEvUzqQuIvi5Fs9m6x3c0ei5CE2NUVWHZlvcGurgAnp6JmqxbNIike9q1NriXTf7/30jihRIavfPD6kuFZBsMdxemTkvorCKuLCMkK0IoSExUzwg16g14mwykt/bcsRrLZ51d0N1tdW2Gzmv9C2NUdIJbOmY0iZ2PpkoB4Pa6t8VCCliYXw2Xjig2a3A6zbkaC7YKwuoiQrIABvZYkpLeQL48D9LZ4mQzpFom7GCNxOWx4nJv7R98d9BJOZnllPMxEOMllNeIj9bDbkrmVzxeYi6XNFvOgB9uXSP8dX0iWpCtXY1iPk3SKRSIIq8rmPs0azLZ2HxPhJN87NF7h1gItIB1JZS1ZW9qaZq+rairsZuKuy3pxO+186MvPAMVGlRfCrm7Nijg5FSWUyJDLFyqC7YlMjnSNzC+AsVCyJMusGjt0a6VLYiSCsKqIkKyA971qmF8+MMhEOFlSc2HQGyytwzDcWZvdrQVaXOM/vV4xW+kbM1IuhK6Ah6DXyYnpqNlS3hoj6dF/tmfLxhtbGQ8l6Gtd3CLZ2alZPtVatQiCcOFs/hOtgbT4XHz2nVfysTv3Vn3eKiTWGMlmD7QbfODmHTx0ZJL5eLqqRVYvNptNy9yaipqV/52WrK1X6R2Zf35ylj09zRWvj6ezevflxS2Sd10/pFXmV2keKQjChXNxnGgNppZLpaes6WGTS5vNcTFYJKDNRfnyB68v6ZN1oSi9Qb57aIxJPeDebnFtDbX72Nbu49HjM7zvpuGK1xpdmPuXsEiavS7u3F+ta48gCCtBXFtriDUN1e92mrM5Ap6NW0OyXLwuh1lHsxJevbeLSDLLDw9PApQE2wFu3t3Jk6dmyea0OEkuX+Bj971gtrwHLYtOEIT1R4RkDdFEQ7M+jO65vS1eelvER1/OLXs6cTvs/PDIBDYbtPlKxfaW3Z1EU1leHNWq6B85Ps23nj/P1w+OMG5aJCIkgtAIREjWGGO8rCEoX/rA9fz+nZc0cksbkoDHyQ0728nkCrT53DjLuvO+alcHNhs8fmIGgG88MwLAs+fmGQslsNmoWogoCMLaI0KyxhjZXD59MFR/axMtvovHtbWa3HGpNpamo8qskHa/m/39QX70yiRTkSQPHZmk2evk1HSMw2NhrWXNJq/NEYTNivzPW2OMq+SLJVNrLXntJd0AJTUkVt59/TYOjYZ4y189RiZX4Pdfr82p+dmxaXFrCUIDESFZYwwh8YuQLMlQu4+rt7Wyp7syxRfgvTds57+97TKmIymuGGzhHdcM4bTbSGfz9C9R1S4Iwtohp9sa81Z97Kz/AhsabjW+9ps34rTXvr751Ru3c/lAC20+N01uB/v7g7w4GlqyhkQQhLVDhGSNubQvyKV9F94+ZKvhcS4tuFcOFacwHtjexoujoSVrSARBWDvEtSVsaq7Zrk1lFItEEBqHCImwqXntJd186JYd3Lq3s9FbEYQti7i2hE2Nz+3kj968r9HbEIQtjVgkgiAIwooQIREEQRBWhAiJIAiCsCIaEiNRFKUduA8YBs4A71RVdb7Kum3AvcAQUADeqKrqmXXbqCAIgrAkjbJIPgk8rKrqHuBh/X41vgL8maqqlwLXA1PrtD9BEAShThqVtXU38Av67S8DPwU+YV2gKMo+wKmq6kMAqqpG13F/giAIQp00Skh6VFUd129PAD1V1uwFFhRF+RawA/gR8ElVVVc+jk8QBEFYNdZMSBRF+RFQba7pH1rvqKpaUBSlUGWdE7gVuBo4hxZTeT/w96u7U0EQBGElrJmQqKp6R63nFEWZVBSlT1XVcUVR+qge+xgFXlBV9ZT+mm8DN7KEkBw+fHhGUZSzK9i6IAjCVmT7hb6wUa6t+4H3Affof3+nypqDQKuiKF2qqk4DrwWeWeqNVVXtWs2NCoIgCIvTqKyte4DXKYpyHLhDv4+iKNcqinIvgB4L+TjwsKIoLwE24P82aL+CIAhCDWyFQrXwhCAIgiDUh1S2C4Ig/P/tnXuMXVUVxn+1UypSigQQCUWtWr7wkD7AgqhUjIwtgVSCYOukrdqgfUVA7R8YLKE1RqsxaQyKDZbyKhRi0SaCfcQoWFKJtCBW+JRSDAViI41OQxUcWv/Y+zin49yhlzP3zu10/ZKbObPP2btf192z1zl7r7N2UIlwJEEQBEElwpEEQRAElRj0+5FIWgFcAuyyfWYuGwfcDLwV6ALm2X5U0sdIEWQ7cvU1thfnOpOBZcBQ4Bbb324RzQuBjly1DTgNOMH2bknPAXuA14Eu2+c0WfPYrHkEKadah+3OfO46YHbW9mXb63L5QNu5V82SLiIFhRwBvAYstP2rXOfXwEnAv3LT7bYbks6nTs3vAZ4CnKtvtj0n1zkbWAkcCTwAXG27IQumdWruABaWqp8FTLD9eJPtfAopRdOJpDx/y20vq5UnUNIQUr+9GNgLfM72ltzWLOD63PQ3bd/WIpo7SBlFhpDGibm2n8htPUcdY8fh8ESyEpjco2wpcKPtccCi/HvBw7bH5U/hRIYCNwFTgNOB6TmFy4Brtv3dQi9wHfAb27tL9S7M5xvmRPrQfAspG8EHgPvJA0S23TTgjFznh5KGtoide9UM/B24NJfPAu7oUa+j1G8amROuHs0A20u65pTKfwRcBYzJn55t9icre2m/V8227yr15xnADtuPl+o1y85dwFdtn056f21+7ou18gROoduWXyTZt0hQewNwLilf4A2Sjm0RzTuASfk7WAIs79HeQY8dg96R2H4I2N2jeD8wMh8fA7z4Bs1MBJ6x/azt14B7SPnCGkIFzdOBuxulqy9qaD4VeCgfbwAuz8dTgXtsv2p7B/AMycatYOdeNdvearuw+TbgSEnDG6WtFnXauVfyS8AjbW/OTyG3A5/qb60FFTRPJ/WBpmP7peKJwvYe0pPdyaT+WDxR3Ea33aYCt9veb3sz6R24k4BPAhts784ZzjfQIKddr2bbj5Syrm8GRr3Zf3vQT23V4BpgnaTvkZzp+aVzH5L0BGmg/prtbaQv4/nSNTtJdxjNpC/NSHobqYMuKBXvB9bnFDQ/tt3zjqPRbCN14p8BV5C2A4Bkz82l63bmMhh4O9fSXOZyYIvtV0tlt0p6HfgpafqimXH1fWkeLWkr0Alcb/thkq13lq4p279ZHIydP8P/30g03c55inA88Dtq5wnsbYw4uY/yhnKQmsvMBh4s/V7X2DHon0hqMBe41vYpwLV0p13ZArzb9ljgB6RO3irU0lxwKbCpx7TWR2xPID12z5d0QXOk/o8vAPMkPQYcTVpbaHX61CzpDOA7wJdKxR15euCj+TOjSVoLaml+CXiX7fHAV4BVkkbWaKPZvJGdzwX22v5jqbjpdpY0guS0rinW9wqyE2u5F/Hq1SzpQpIjKWdgr2vsOFwdySxgTT6+jzSlgu1O53T1th8Ahkk6HniBA++YRuWyZtKr5hLT6DGtZfuF/HMXaR66Z52GYvtp2+22z87atudTtew54HbuQzOSRpHsONP29lKdws57gFW0iJ3z1OHL+fixXH4qyablaYyWsnOmr/7cFDtLGkYakO+yXfzt/S1PWRVThMU6TUv06To1I+ks0nrV1KKvQP1jx+HqSF4EJuXjjwN/AZD0zhx9gaSJJPu8TMr7NUbSaElHkDr52lbQnLUek8/9vFR2lKSji2OgHSjf3TUcSe/IP99Cilq5OZ9aC0yTNFzSaNIC5aO0gJ1raZb0duAXpAXiTaXr2/LNRvFHfAktYmdJJ+QABiS9l2TnZ/M0R6ek83J/n0nv+e6arrlUdiWl9ZFm2znb5SfAU7a/XzpV5AmEA/MErgVmShoi6Tzgn9nO64B2ScfmRfb2XDbgmpV2oF0DzLD951I7dY8dg36NRNLdpE20jpe0kxRBcRWwTFIb8G9SlAXAp4G5krpIIYbT8qNgl6QFpA4wFFiR105aQTPAZcB626+Uyk4E7pcE6XteZfuXTdY8QtL8fMka4FYA29sk3Qv8iRRpMt95n5kWsHOvmklrT+8HFklalMvagVdIa1fDsuaNNDAnXJ2aLwAWS/oPsA+YU5r6nEd3+O+DHDg/PpCaC93PO2f+zgyniXYGPkyaOntSUhE19nVSCPi9kmYDfyU5PEgh1BeTAkf2Ap8HcArDX0K6SQJY3GP6eSA1LwKOI0VNQneYb91jR+TaCoIgCCpxuE5tBUEQBP1EOJIgCIKgEuFIgiAIgkqEIwmCIAgqEY4kCIIgqEQ4kiCoSH534LeSppTKrpDUsHDrIGglIvw3CPoBSWeSMg6MJ8XebwUml9+AfxNtttnu6ieJQdAwwpEEQT8haSnpBcWjgD22lyjtRTGftI/JI8AC2/skLQcmkF4IXO3uLQt2AneSssZ+y/Z9A/BfCYK6iKmtIOg/bgQ+S0p0tzQ/pVwGnJ/312gjpX2BlGrlHGAscJEO3Hdll+3x4USCQ4VwJEHQT+QUNauBO3KK+U8AHwR+n1NWTALely+fLmkLKeP0aaSNvApWN091EFRn0OfaCoImsy9/IG1husL2N8oXSBoDXA1MtP0PSXeStlAuKOdMC4KWJ55IgqBxbASuLGWtPS5nXB1J2g+7U9276AXBIUs4kiBoELafJK2bbJT0B2A9KbPqFlLm46dJ29xuqtlIEBwCRNRWEARBUIl4IgmCIAgqEY4kCIIgqEQ4kiAIgqAS4UiCIAiCSoQjCYIgCCoRjiQIgiCoRDiSIAiCoBLhSIIgCIJK/BfrRN0lrcA1vAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f97a11baf28>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "seaborn.lineplot(x = \"Year\", y = \"Annual\", data = donnees)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'0.9.0'"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "seaborn.__version__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "donnees_mois = donnees[donnees.columns[1:13]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f97a18324a8>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXAAAAEFCAYAAAD69rxNAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAIABJREFUeJzsnXmYXFWZ/z9V1VW9d6ezdKeTkJXkDRAgbEFWg6DgCo4Lor8R1BmccdRxHR3HR8UZHXTct1HGUdRx3wCRVWURBCHsJOQFspGEbJ3e99p+f7ynK0XRIdXp6qSX9/M89dStc9977nvOvfXec885934j2WwWx3EcZ+IRPdwOOI7jOAeHB3DHcZwJigdwx3GcCYoHcMdxnAmKB3DHcZwJigdwx3GcCYoHcMdxnAmKB3DHcZwJigdwx3GcCUrZ4XYgnxUfuj77kTevZCCVBuDoxhoA7t7SxlkLpwNQmyjjEzc+AcBgKgPAa0+cy/KZ1QC84u++zCw5CYC1nz2LaRd8GoCHrv0PAObUxPnluj0AfOlnj/LZt59s+5pVzft/+3guP4Dzl8xgemXM8v32fbzl9AUApLPw9kX2BOtfuszH+fXlXKeW7z+uqOMz97bkyrV+RycAy5vrALhL95BKWhmvvvQkbtnQCsAta3daPcyt5z8/+nkALvvX99LdnwTg0lMXEI9Fwv4qAHimo5+WXls/r66cjv4UAF2Daf4rlOfkY5oA+PDqxXzj7i0AVCVi3LV+t5V31RH87qFnrZzHNwOwcXcP7zt7kdVzOsvGtj4APv6jBzh6+SwAPnG+APB0a1/On989sYvj59QDsLOrn61hu4+daXV6/cYuTp5j9fDxG9dz5lLL67XLZwIwLdnGz5+xeptZFScdHhTe3TPAKXMt37k1cQA6BtJs6xwEYHZNgkUpK8PtPdN5SfZJAFqaVgJQH0uxodvqbklthMhAFwA/fDpJTzgW71piO0tNm8sfN3UA8LI5MW7abnXaXFPOScmnAEjOPsryH4xSX27toGgkQmuf2dqZCb3JDLGI7XdvX5LdPeZvY3WClY2VAGzpsuP327W7+MDxdh5/6ZEe3njsbADW7enl6FlVVj/ldj4OpLPMyFoZopsfZueCMwGY8ZcfEDvrYgC6orZNbaaXbCS01dZcT88prwegrmMT0d52K/Oe7QDE6meQmbUYgPQDNxE50/KKde0CoP/2X1J+5HG23zlLSG+zet76s19SPq3W8l1o59BgVw+1x50IwF/e/zXOuPXXVne3Xk3PdvuvTH/pKwGIxGIMrL0PgL49rey4Zx0Aj96yiaPOmAfAvBcfA8C2O9YCcMI1t0QYBf8QWVj0Y+jfzm4e1b7GCm+BjzNWhCDlOI5zIMZVC9xxHOdQkYiOy0b1iPAA7jjOlCQ28eO3B3DHcaYmQ2MTExkP4I7jTEkmQwv8oAcxRaS7lI44juMcSmKRSNGf8Yq3wB3HmZJMhhb4qAK4iNQA1wINQBz4uKpeKyILgRuBu4DTge3AharaNzp3HcdxSkN8HLesi2W088D7gdeq6onAOcAXRWSoVpYC31TVY4B24HWj3JfjOE7J8C4UiACfFZGzsYfP5gJNYd0mVX04LD8ALBzlvhzHcUrGZOhCGW0L/C3ALOAkVV0J7AIqwrqBPLs03t/uOM44wlvgUA/sVtWkiJwDLCiBT47jOGPOZGiBH1QAF5EyrIX9Y+B3IvIYsAZYX0LfHMdxxoyp/Cj9McAGVW0BTtuPzYqhBVX9wkHux3EcZ0wYz10jxTLiAC4i/wC8F3hf6d1xHMc5NEzJLhRV/Tbw7THwxXEc55AxJQO44zjOZGBKdqGMJR97y0p2dg+wdrsp2Aypscytq+DGoB5TmYhxzlGNAPz3L01xZul5y7jmMVOzOf//vYZ773wagL3ZSo59+asBePv31gBwxeuO5UIx9Ze7T57Lqyq2AfBPd8RzCj8b9/QA8Ntkmo27bTmVTFNTbtX1qqXTSd/1EwDqj7Hnk3b1JFky3dRU2qnkiAZTWzmivpJ0xgqyLajTRKIRVi6ekStjZ1DcmVaVYFtbHx9cvYSfn30hAItnVfOzW00FJrNqPk3VNkvz54/uAGDt9g7e++IlgKnzDLUq5tWVU1WTAODP924F4DUrZrNlr5VnWVMttdW2/k7dw+Z1Vr+zz1gIwF837GVzez8Ay2dUMr3Syt7R0stAqKeyMAh059MtXBTUY5bMrObujXsBeKnMIhps9gwENZzpVXQOmgLO40/s4eXH2HZbOkyp5olkBb1J83FmVTXHTbeZrl97oIuF9eZvMpwY8WiERdPKzZdYhE3MAeCMI+KkBkwtqLXf9lVZm6BzwOr/6WwZyYyp1cyo6uO+daY2kz3Gto/1trJ8pqnLPNie5JkO225RQyXphB23/vDXmVk2QLSrzbZPVNIYtHh0wM6FTW19HBXUolp6k1TFTVEHINpn2z2w3baZW19JWavJEXX315MOsj7bO/s5b4EpP+0dsLLv7UsxK2P1zOwlVMatnuJHLKUzKPE8223n1bK6csq2PGC2S46nJmkqPJH2nWSj5k+0Zpqtr25gR9xUkqJ/fZDZzYuCs5b/uh/dzrwz7dxrWLGM+CJTyRlo66H5bFO32vuQKWY1rj6T9vvuBeDoS1YR6zF/09V1JHvsnCQTChlPkB6wmcf1Z5zL7gcUgDtbeml8yupp4fl2rGuCstVo8Ra4U3I+uHrJ4XbBcaYEU3kWiuM4zoTGu1Acx3EmKN6F4jiOM0HxFrjjOM4EJTrVAriIpIHH8pIuUtXN+7FdDXxIVV910N45juOMEZFJ0Icy0hZ4X3jroOM4zoQmlogd2GicM+ouFBGJAVcCq4FyTMThO2F1nYj8HjgSuA14l6pmRrtPx3Gc0TIVW+CVIjIk0rBJVV8LvAPoUNVTRKQcuFtEbgk2q4CjgS3ATcDfAL8qgd+O4zijIjoFA/hwXSgvA44TkdeH3/WYnNogcJ+qbgQQkZ8CZ+IB3HGccUAkOlo9m8NPKWahRID3qOrN+YlhEDNbYFv423Ec57BQ6ha4iFwAfBWIAd9V1SsL1s8HfgBMCzYfVdUbRrPPUlyCbgb+UUTiACKyTESqw7pVIrJIRKLAxZhKveM4zmEnmogV/TkQYSzwm8DLsW7jS0Tk6AKzjwO/UNUTgDcB3xptGUrRAv8uJlj8YFCk3wNcFNbdD3yDfYOYvy3B/hzHcUZNiVvgq4Cn87qMfwZcCKzLs8kCQ2/iqgeeHe1ORxTAVbVmmLQM8LHwyed24OyD9sxxHGcMiZT2ZVZzga15v7cBpxbYfAq4RUTeA1QD5412p/4kpuM4U5JorPgeZBG5HLg8L+kqVb1qhLu8BLhaVb8oIqcBPxKRFaOZWu0B3HGcKclI5oGHYP1CAXs7cETe73khLZ93ABeE/O4RkQpgJrC7aEcK8ADuOM6UpMQP8twPLBWRRVjgfhPw5gKbZ4BzgatF5CigAhszPGjGVQC/4ntrOO20+fQNpgCI51Xw4qBq8qsHtrG0ydRS5h05HYB1e7q57UG72J1wVCOXXXwiAP/3yA4Wzq8HYOMWUyG54prHaXm2C4DffewczvmhKZWsXjGbT79sKQAfvcHUQKSplke3tud8eHJ3NwDXA69rMNWS6qCwUlEWJZO1WZL96Sz9QU7lmsd20BcUaOYFlZ5EWZSNeyyvWASe2mXLx8yt46an9rB2eyeXXWQD2OlMloWLrZwPbu+gY8Dq5vu/NjWiJSua2NjaC0BdRZz7nzH1ki1VcbqDos7pq+YB0NaX5LyjmwC4+pan6O02FZyVxzVRUWVqN5tCXl39KTaE5fJYlJvVGgn9bTtprl8GwN3PWN1cevIRbGgz22gkQiz0Lbb0JmmuMRWVobR7traTDvX0yjMXcnSjDavEw5zcHd0D9Kesvq5/YhctCxoAmFGd4PdPtQLwmmWmivOHTe0cM8u2T2ezVJZZHvds6+LMGZZHVbwi+JKiqToe9jGYe5FRU02Cs440hSbtM1+lEjp6gg/rdrFwhinctPelaJ/ZDEBtxI7vj7WTtyy349MfSbCp3VRl1rfYMW2oiBPc4rimahoq7HzpS2bYnjR/oMOOQ3Mt2ZSd52csms7QHf5j2zqIHG3ncU3CfKwsi5MqWwxAWevmnPpO/cwjSIT/zcygopSNRumavwqAut3rIGIZ773pOqa/5d123Po6gm2MOb1bbMcXvY5ss/0n2LoWgMWvOIGG114GQGrDo/SsuZMhIjErW81c+290PPwwtYusUTrY1s7eX37PfDjuOLq2mgrSjB2bbP8NjcRn2rm5+/fXkg7/GYBUv53zXc/YNtM+90NKQSkfpVfVlIi8G5uVFwO+p6prReTTwBpVvQ74IPA/IvJ+bEDzMlUd1dTqcRXAHXJyco7jjC3REivyhDndNxSkfSJveR1wRin36QHccZwpSWQEg5jjFQ/gjuNMSabiu1Acx3EmBVPxbYSO4ziTginVhSIiWeDHqvr/wu8yYAfwV1fdcRxnohGLT/wAPpIS9AArRKQy/H4pz5+o/oKEoO84jnPYicaiRX/GKyMNqDcAr8Te6X0J8FPgLAARWYW9SrEC6APepqoqIpdhQg412PzIF5fEc8dxnFEwGfrAR3pp+RnwpvAI6HHAX/PWrQfOCq9K/ATw2bx1JwKvV1UP3o7jjAsisWjRn/HKSN9G+KiILMRa34UvIq8HfiAiS7GnjOJ5625V1dbROOo4jlNKJoMiz8GU4DrgC1j3ST7/DtymqiuAV2NdKUP0HJx7juM4Y0MsUVb0Z7xyMAH8e8AVqvpYQXo9+wY1LxuNU47jOGPNlOtCAVDVbcDXhln1eawL5ePA70frmOM4zlgyGbpQig7g+1HjuR1T3kFV7wGW5a3+eEi/Grj64F10HMcpPUNvT5zIjN/OHcdxnDFkPHeNFIsHcMdxpiTRqdSF4jiOM5mIjuPZJcUyrkowb+kM6ivjbBtShQnqM13A8pnWBf+iJTP47Z2m4rFgwTQAls+s5u3nW/d7a88gx88x9ZJMNssf1pqKxyffeDwAA6k06aCB8fNHd/CxV5nyTTwWYUdQNXnPWYsA+NSN65lRY0o19Y01LJphaindAyki5TZLckuHqd7MrIpTW27VedeWdvZ0mjLLttZemqfZ2wfmBkWe3sE07zzL1FTqymOcFFRnegZTHDO3jke3dnDzIzsAOOeYJlbMDWosFWWsmlsHwBHLTEXmFcc3Ux4kX+rLy9gbVHa2tfXR0GR1tjEoCb3v7MVccdN6AKbNrCIe/K2vTJAOCkL1VTZ9f3lzHbs6rWwvXzqD2dOsvK943Rl09Nk+zgr1/9COLmrCnyEaiTA91FkqnaHLTNkU1IEaqxM5NZzW7kFa+6zOK0IZjm2spjzc2vbWp+kKyizz6ytpCvlGs5Z2XFMN82rN38F0lo6BdC6PwTLbR3O3Hf9svIJswo5fPFpOZXgPxlOt/ZwS6nR+Xci/fTexqKnKrFrQwPZQD/PqEjm1mxT2vTwoRQFks1nqy61ftSooNfUm0wyEE64cWLunD4ATpkepqbY6rauwMlTHo2TK7FjXp8sIh4Q3rJwDoc7a+62MkQjUJE38I1M9g/lx8z2TmkVrsMm1L7MZKkmFeignU26KVtNOO4O03mfbHX+u2UaiRHv22vbVtaQqzLZs+mwAOjZsZ1qLTTZL7dhEvM7qbuGrz4KolTkxzc677u17qDvO1JN6dRONr7wQgJ4H7yIbCpdqM0WxrkceZ9pRR9r67XuIV5vy0Nkzq2g61o5FJnPQ2r/DMqUGMZ1Dw6NbOw63C44zJfA+cMdxnAmKB3DHcZwJynh+y2CxeAB3HGdKEo1P/PA38UvgOI5zEEzZLhQRuQj4LXCUqq4vrUuO4zhjz2SYhXKwJbgEuCt8F42ITPxnVx3HmRREY7GiP+OVEbfARaQGOBM4B/gd8EkRWQ18GpuyfSRwG/AuVc2ISDfwHeA84J+wwO84jnNYmapdKBcCN6nqkyKyV0ROCumrgKOBLcBNmIzar4BqTPj4g6Vw2HEcpxRMhgB+MCW4BJNWI3wPdaPcp6obVTWNiT2cGdLTwK9H5aXjOE6JicbLiv6MV0bkmYhMB14CHCsiWUykOIu9/ztbYD70uz8EdcdxnHHDZGiBj/TS8nrgR6r6zqEEEbkDU6ZfJSKLsC6Ui4GrSual4zhOiZmKs1AuwaYP5vPrkH4/8A3gCWDTMHaO4zjjhkg0VvRnvDJSVfpzhkn7mog8CnxIVV81zPrnKfk4juMcdsZxYC6W8ds77ziOM5ZMgi6UkgTwfG1Mx3GciUAkvEN9IuMtcMdxpiYl7kIRkQuAr2Kz876rqlfux+512DMyp6jqmtHsc1wF8DeeOp+qeIxXrzD1jyG1lqp4jI1tptLzmqOb+MNjOwHYsqUdgA1tvTlFmM9/bw3f+sBZAGzvHGBxo3XBD6RsJuPV92xhMGXKHv/9xuO4b7upmiycVsmaZ01M4WWLpwOwYm49u7tMWeftp87nMzcrAPOmV7H6xaeZ00Fp5pmOfk5qNvWSZ9r7WBlUdAAe2Wp+rmi09YPpDBtaewBYPrOKzS22fMHyRk6aU8+fnm5ha0jr7k8Ri5oay7FNtTm1msWzLa/qeCynMJTJZnMKQjs6+qlM2Al6/rFWn3/e0pZT7Fkws5qnQ/0OpjPMCPl195tyy5a9PZw8fy4Am9r6SWdsJw+v3cW5q46wcu00pZ+ZVQnSWVv/6M5OGipNYaYmKP6kM1m6glpONBJhIP18ZZV4uJ3d1ZPMK28N6/ZYPdSXl1EZVHu2dtv2tYkYg6HwA+lsTmUnnc0SjdjyhqwdSwahMmP5zi5PsysoBc2ojFNbvm87gExFLa17k8GvSK48s6rKcnXdmzQfymMxCPvqHMjklIVmBmWjeDSa83FhXZwn95oiTyQNbSmzGdomAmSqTJ0p2TkwJMLDU3t7OHVOo6Vn7PhUlkVzE3XTFXVEM0MnQYqmclvuTNvxTxEhMWj1mI3u+8uXNc0nO2jnbyZp32SzRPu7LK1qGrGuXWEnVh+VsxqINNj5VHnKeQw+/ahtlkqS7bH/UsWx9t+o7erN7athxTKyKav0xKxG6hY2W/00LwQgun0P/TtsXxUz6unYZP/x4162iIoGUz1K1FZRSko5CyW8JuSbwEuBbcD9InKdqq4rsKsF/hn4ayn2O/E7gSYZf3q65XC7UFKGAr/jjDuiseI/B2YV8HR4mHEQe8jxwmHs/h34HNBfiiKMqxa44zjOIWMEXSgicjlweV7SVaqa/6zLXGBr3u9twKkFeZwIHKGqvxeRD4/c4efjAdxxnClJpCxetG0I1gf9cKKIRIEvAZcdbB7D4V0ojuNMTUrbhbIdOCLv97yQNkQtsAK4XUQ2Ay8CrhORk0dTBG+BO44zJYmU9j3f9wNLw+tEtgNvAt48tFJVO4CZQ79F5Hbs4cdRzUI5YAtcRLIi8n95v8tEZI+IXD+aHTuO4xxWotHiPwdAVVPAu4GbsdeJ/EJV14rIp0XkNWNVhGJa4D3AChGpVNU+bJrM9gNs4ziOM74p8TxwVb0BuKEg7RP7sV1din0W2wd+A/DKsHwJ9r5vAERklYjcIyIPichfRERC+p0isjLP7i4ROb4UTjuO44yWyfAyq2ID+M+AN4lIBXAcz52Evh44S1VPAD4BfDak/y9hxFVElgEVqvpIKZx2HMcZLZGyeNGf8UpRAVxVHwUWYq3vGwpW1wO/FJHHgS8Dx4T0XwKvEpE48Hbg6hL46ziOUxpKOwvlsDCSaYTXAV8gr/sk8O/Abaq6Ang1UAGgqr3ArdjTSG8Efjxqbx3HcUpFCQcxDxcjmUb4PaBdVR8LKvRD1LNvUPOygm2+iynX/1lV2w7WScdxnFJT4mmEh4WiLy2quk1VvzbMqs8D/ykiD1FwQVDVB4BO4Puj8tJxHKfUTIIulAO2wIdT1Ml//7eq3gMsy1v98aEFEZmDXSRuGaWfjuM4pWUcB+ZiGbMnMUXkrcBngA+o6vPfH+o4jnMYGc+zS4plzAK4qv4Q+OFY5e84jjMqIuN3cLJY/F0ojuNMTTyAl5Yl06uIBzWWJQ0VbOuyPqqO/hQnzzGFm4d3dPG3Zy4E4Id/3gSYIkxLr6l9/PCj5/D4blMU2dnRz9tXzQdM1QTgV28+mp+pKYfs7B5kc2tvLo+5tRUAPNlqqiknHTGNW9bvBqClN8klp9jLxk6ZW8/0CvPtwaBq0z2Y4tFdQaGmOsGyGaYi8sDWdmorrJrv3LQXgL7BNNIUFHAG00wPKjrbO/uRxhpuf6qFZXPqANjR0cfpR9o7cJqqy9nZbfs7LagGPdvRz9FBdagqHqNv0JRvtrX2kghKL1Vx87UiFqWmYt8hn9lQCcBgKkO83Gwaa8sBKItGiIZjkcxkeXSrqRW95WVLqU4897RJZrI019h2dRVxGsPy0H67BlIsDvvSvb3s7DaVo/qqOLUhr+Yau53d1D5AR38y+FtJQ8W+29y6hJWnL2UiEdMr9ynyzMj00ldmdb6zJ8n0pE16mlFp9XTXMx2cvcDOoWwsxpZ2O+5z6xI5pZ8hIsl+mqptu95kOpdeRoayCLQnoT7U1+waID0QLBIM6VcMKQzVlseYW2tl2NadYsE0O8d6y8qIBd/7g0JUXXkM0vae/5q8Op5RlSCS7Atlt/3OqYmTwdR7YgPd7E5Zvs1RaEvZvgdCvpVlMRhS4slmnhO4IokKsv09ZBOmdhMZ6Mmtj7TvJNO4BID02nts83SGSCpoEUSikLH6yfZ0QuiSyCbt+CV7+mDzBgAq5h2RU//p2byFSFCWGty2EYBovIxYhf0P0skUyZ5kzseqRvM9lihtuMp6AB8bljRUHG4XDhu3PzW5FHm6BlKH24WS0p48sM1EItvfc7hdOHx4AHccx5mgTIJ54B7AHceZkngXiuM4zkTFA7jjOM4EZRIE8IMugYh0l9IRx3GcQ0okWvxnnOItcMdxpiRTvg88vJXwQ6r6qvD7G8AaVb06KC//AHvFbBx4g6quH5W3juM4pWISvAtlrC9BLap6IvDfwIfGeF+O4zjFMwm6UMbas9+E7wcwRR/HcZxxQTYSLfozXhltH3iK514ECh+hHHrGOF2CfTmO45SOcay0UyyjDapbgKNFpByoBM4F7hq1V47jOGPNOG5ZF8tBlUBEyoABVd0K/AJ4PHw/VELfHMdxxo5J0Ad+sC3wY4ANAKr6L8C/FBqo6sK85TXA6oPcl+M4TsnJRid+r+6ISyAi/wC8F3hf6d1xHMc5RIzjlnWxjDiAq+q3gW+PgS+O4ziHjkjkcHswaib+PYTjOM7BMBVb4GNJc005O7oHWL+3jxWNVTQHpZq2viSb2k2RJJPN8qs12wC45PSFgKmmnH7ENAC2tPeTzprSSd9gmmi4yq4ICjg92TgNlaYccnJ9kg31phRz4uxqBoOcSmufqYxMr4xzaVDh0ZYeTmg2lZwZlfYEV3cyQ20iqN2URWkOajY1ibKcQtBgKkNzfSXLmmqoKd9X3UNqKV0DaTp6TSWgriLOSQsauPGxHbwp7Ld7IEVPUIXJbzC09Fj+yxprSAfJ6K6BdE5xp7Gugvbgw5A6TG8yzfGhnnZ09NPaabM8Z9SUs3ROHXu7B7h/synZzG2ooqPfxBhOaK5laZOp/px/5CwA7tzSyqY2U7WZXVNOY7Udq/n1FTwZ1I/m1dms0vqKOLq3N6SV0xRs/7ylNafaUx5UcaZV7qujZCZL/VB5qsuoCXU9mDa/0pkslQPmL7EEe3otfXpFGR1Mp6FzEx1ZU+E5ZU5tTi1nc2eSxqAAlIhG6BpIU18RI5Gx+hqoaaJlxz6hgyHfulIRYhGoKIvklICiEeiJWDlr4lCdNdWZrWFfrX3JnNpQLALpcAxj0QiDg3bghlR6kpksfTFbTpQlSQWHm2oSPNCa5YTZ1dQM7lOUyIYTIltWTjkh42SK6ek29sQacnZ9qSxVezdbndU2Eu0xZahseTWUV0PHblojpmY0vRzSId9I2y4ig1YP0Wo797u378mp8FBWRqbH1K8yqWRuRkQmiEREYlHKqq082VSS9J7tlh6Nkuqxesok7ZiVT6vNLdfMnUnvjlY7Fp0DdGxqoW7BdFJ9dnwSlIbxPL+7WMZVAB9iRWPV4XbhgHQnM0XbLgvBrxhufGzHwbgzavZ2DxzYKHDnltaibesrDo/yd0PnpqJt6yuKf6S6ouzw3HafMLu6aNv84H1AOnYfhDeHlroF08ck3+wkeJR+XAZwx3GcsSbcqE9oPIA7jjMlyUyCCO4B3HGcKcnED98ewB3HmaJkShzBReQC4KtADPiuql5ZsL4c+CFwErAXuFhVN49mnyUZhj2QOo+I3C4iJ5diX47jOKUgm80W/TkQIhIDvgm8HDgauEREji4wewfQpqpHAl8GPjfaMkz8eTSO4zgHQTpb/KcIVgFPq+pGVR0EfgZcWGBzISZyA/Ar4FwRGdW0ppIFcBFZLSLX5/3+hohcVqr8HcdxSkkmW/xHRC4XkTV5n8sLspsLbM37vS2kDWujqimgA5gxmjJ4H7jjOFOSYrpGhlDVq4Crxs6bg8O7UBzHmZJkRvApgu3AEXm/54W0YW3CK7nrscHMg6aULfADqfM4juOMG0o8Dfx+YKmILMIC9ZuANxfYXAdcCtwDvB74k6qOyotSBnBX53EcZ8KQLuE8QlVNici7gZuxaYTfU9W1IvJpYI2qXgf8L/AjEXkaaMWC/KgYdQDPV+cRkSF1nk24Oo/jOOOY4t9mVByqegNwQ0HaJ/KW+4E3lHKfpWiBF6POs7oE+3EcxykZk+BJ+tEFcFfncRxnojLl34Xi6jyO40xUJn749nngjuNMUUr9LpTDgQdwx3GmJOmp3oVSaja199FcU866PSa/NT1In82sSjAziPQ8truLC0+wJ1SH5LZOaK5lb5AlW9xQyQf/5z4A3vOGY3lkVycA84N02vqWnpyM1+5sNacdYQfxts3t1CQsv+OaTP3k2a4Mf3nGJLtedMQ0NgdZt+UVPQxE63O+AVQnYrlBkd7kILGoveIgFo1w67pdAKxe3ghAXzLNnCA31ptM56aeFRNaAAAgAElEQVQzxWMRXrNyDo9s7SAZXsDQn87kpMkqyyIMBjWr5pC2fGY1XQOWmMlmqQyyY5t2dHLk3Ppc+hB7u02Wakd7H285fQEA63Z00d1v9ffks1ZfLz6qkSXTrdIT0UhOqu2Zjj76gsRbZajHuoo4j+8yaa09vYP0BScHUhmagixeRcweEehJpklmbPz/lLnTqEkMyb1ZWmtvknjM6i4aieTqIZXOEsuYj0PHLwsMVJj6TDqTJRax/WazWSJBdm3+kzdaGRe+jJlBrm1GZVlO0q5jIENjxN7Flo3bORLPDDInyOOVReHhnba+LpqkO23nZHnwMRGL5I71QCpDJGv73dFlvhw1q5q+sK9oJEJF2C4WiTCQsrL1hPqaWxPPtQory6LsDrJ5/akM3UF+bVunpTVXRUlmre7KUwNEMd+jgz3UTZuRqxOAZMZk10LlEEkHWbZUkkiVnSPxUIZspJxIf5BJG+wnGmTH0m2m3FPdPINspcmrZZ7dQKrb6qasro7e7aYmVTvL/p+RaJSBNsurqnYalNm5MNDeTfXcmVYPFSGtrYtsOC/2Pr6JgSD319fWT1mlnZOVs0agNFQEkyB+j68A7sAjWzsOtwuOMyXwLhTHcZwJirfAHcdxJiiZSTAPxQO44zhTEm+BO47jTFCSRSo1jGeKDuAichHwW+AoVV0/di45juOMPZPhScyRvA/8EuztgpeMkS+O4ziHjHQ2W/RnvFJUC1xEaoAzgXOA3wGfFJHVwIdU9VXB5hvYaxOvFpFXAF8CeoC7gcVDdo7jOOOBqdQCvxC4SVWfBPaKyEn7MxSRCuA7wMtV9SRg1ujddBzHKS3pTPGf8UqxAfwSTGWZ8P1C3SjLgY2quin8/ulB+uY4jjNmJDOZoj/jlQN2oYjIdOAlwLEiksXUJrLAtbiEmuM4E5Sp8iTm64Efqeo7hxJE5A4seA8noabAYhFZqKqbgYtL7rXjOM4oKaWk2uGimC6US7Dpg/n8GtNzG5JQ+wVBQk1V+4B3ATeJyANAF+Av+HAcZ1yRyWaL/oxXDtgCV9Vzhkn7Wt7P50moAbep6nIRiQDfBNYcvIuO4zilZxI8xzOieeAj4e9F5GFgLVCPzUpxHMcZN0yJFvjBoKpfBr48Fnk7juOUguR4nh9YJP4uFMdxpiSToQtlXAXwW9fv5l1nLKAiqIf89OHtAFyyci53bmkF4KFn2pnXYOojmazZvetnj/Av5wsAj+1q5dXnLAYgFoEHt7QDcPbZpuaRTGdYFLZ/srWPWCSS2/81j5miyJy6hQDc9UwrK5tNfWRb5wD3bjIfXrxgCeVhBLtjwBRYdgX1FIDmmnKeau0BYO70ShbONIUf3WnqJEubamioMGWXNdva2d1l6iMXLG9kSUMV61t6eGir+d2XTDN9iSmsDKSzOfWX3qCKE41E6B40Hza09ebWN9RX5EbZhx4F3tU1wLSqeEirYmubKQylM9mcQk0kbN/dn6IrlK07mcnV0/z6SpIh32gurYKB6kSu/F1l6dxye5/l0Vhj+41FIzkFoe7BFM21tl0iKNUsaajg4V1WdzOq4jl1nu5khllllldF3PJ6tjvJ/ITVnfbGmV1t6alMFvpMxSWy7EUAzE7E6QuqP+lkJleGWBSyCTs+ZG19NlpGJGLLfalsToFoMJqgpz8d/DVVoKp4lPaQlncq5baJRSKkQqQYTGdzNv2pDJVl9qM2KEF1JzO5Ps2+VIb6citPPJrO1c/8ejvnI9kMBLWcTKKaurBdtq8sNz2uJ2kLFWURIkk71pTXkE2E/09VA9E+m1/QGY5Jdd5xLGtaQLKu2Wx7rD5nnXU6dJg6T7S6lvisJvMnGiUSDf5027k72NVL9Ww7d7PJQaK19h+MJcqomjeHfDK724gE1abq5hm0rNtjy01VzF5l/+2B9m5KyXjuGimWcRXAHZN8cxxn7MlMgmmEHsAdx5mSeBeK4zjOBMW7UBzHcSYoPgvFcRxngjLlulBE5N+ANwNpIAO8U1X/OhaOOY7jjCVTqgtFRE4DXgWcqKoDIjITSBxgM8dxnHHJeFbaKZaRtMCbgRZVHQBQ1RaAIO7wJaAGaAEuU9UdInI78Ajw4rCft6vqfSX03XEc56CZDG8jHEkAvwX4hIg8CfwB+DnwF+DrwIWqukdELgY+A7w9bFOlqitF5Gzge8CK0rnuOI5z8AymDs0gZtBU+DmwENgMvFFV2/ZjWwesA65R1XcfKO+iX2alqt3AScDlwJ7g0DuxoHxreHnVx4F5eZv9NGx7J1AnItOK3Z/jOM5Yks5ki/6Mko8Cf1TVpcAfw+/98e/AncVmPKJBTFVNA7cDt4vIY8A/AWtV9bT9bFJY8ol/z+I4zqTgEHahXAisDss/wGLoRwqNQnd0E3ATcHIxGRfdAhdjaV7SSuAJYFYY4ERE4iJyTJ7NxSH9TKBDVV3YwXGcccEhbIE3qeqOsLwTC9LPQUSiwBeBD40k45G0wGuAr4dukBTwNNadchXwNRGpD/l9BXsPOEC/iDwExNnXL+44jnPYGUlgFpHLsXg3xFWqelXe+j8As4fZ9N/yf6hqNmgLF/Iu4AZV3SYiRftVdABX1QeA04dZ1QKcvZ/N/k9V31e0N47jOIeIkQTwEKyveoH15+1vnYjsEpHmMDuvGdg9jNlpwFki8i6ssZwQkW5VfaH+cn8S03GcqcmhmoUCXAdcClwZvq8tNFDVtwwti8hlwMkHCt4whgFcVVePVd6O4zijJXXoBjGvBH4hIu8AtgBvBBCRk4F/UNW/O9iMvQXuOM6U5FDNQlHVvcC5w6SvAZ4XvFX1auDqYvIeVwE8nclSEYvyVIeph0yvsSf1W/uSLG6oAmBP1wBb9vYCcPrC6QD883lLqSizCTU9QZ0GTIHmtru3AHD20pkArNnSRs0xNtbQm8xw98a9ACTKovQNmirJto5+22/3INe37AJg3vTKnJpNIhZhY2cSMPUdgGvW7WRJUN5Z82w/bb2m0FNfGae5tvw5vv35yT285AIbqNjdU8Xa7TY556m9PcQi8Nj2Tk5bbGW7Z2NrTvUnnYEng+DDr9dsA2BFYy1Hz7L9tvQmuX2LKZls29rBjKMaAWjvM19rKsqoTJiSzF3rd9PXVJurqxmhHM9sDROFFjTklIue3tvLnqAadM26nbz+WKu/+7Z15rafX18BQNdgmoZKq6fO/mROmWh2jamxNJRH6Q/ra8tjFD7N/EznAFVxO5ZRItQkTImmJh4lkrU6LetvD/WcpmGx5XtUehObB+YDsKh/M9laO95DKjuD6SyzgyrQQCqTU6tp7U8RGbTzqT1q9VEXi1IZ/hltfYPUBfWknmSGmuBb54DlG42QU8CpS0SJ7bXzpSxmSjQ1iSgdQe1mQX0itxyLRhgMGw6pK82uKaMnqAY92zmQUw3a2dXPgvpEzgeASLSXwTLT4dmnBQU1NTPpDTZDij+V8SiZKqunbHk1pO18imxbR3bOspw/AJHUANGWzQBkBvqJTLc67dlu51XdjGaotrwoixOJPmnL0Ri9O02xqrJpFgCpnn769tjzKrU1NXSvXwdAxYx6Up12nsWqa3K+x6vtfOt6Zjd9bfYfXHDucjLJoAwVfCinNEy1JzGdQ8Bj2zsPbOQ4zqiZau9CcRzHmTR4C9xxHGeCcghnoYwZHsAdx5mSpDMewB3HcSYkU7ILRUTmAd8EjsbepXI98GFVHdyP/fuwx057R+Oo4zhOKZkMAbzol1kBiEgE+A32rtqlwDLssc/PvMBm7wOqDtpDx3GcMSCVyRb9Ga+MtAX+EqBfVb8P9npZEXk/sElEPgVcAVyA6WX+DxAB5gC3iUiLqp5TMs8dx3FGwWQYxBxRCxw4BnggP0FVO4FnsCeKFgIrVfU44Meq+jXgWeAcD96O44wnDuHrZMeMUg5irga+paopAFVtLWHejuM4JWU8B+ZiGWkLfB0mq5YjaLjNL5lHjuM4h4DJ0AIfaQD/I1AlIm8FEJEYpiJxNXAz8E4RKQvrpodtuoDa52flOI5z+MhmskV/xisjCuCqmgVeC7xBRJ4CngT6gY8B38X6wh8VkUeAN4fNrgJuEpHbSua14zjOKMlkskV/xisj7gNX1a3Aq/ez+gPhk2//deDrI3fNcRxn7MikJ/4sFH8S03GcKcl4blkXiwdwx3GmJNmJ3wD3AO44ztQk6+8DLy0XnzCX8rII8aAOsiio8Ny7tY0T5tQDcPMD2+kN6jCPqil0vPPlwg/uNeWdZU21/OxaU/746DtO4UWr5gHQVG2KJsuba3MqMTc/vpOTF9lkmbMWTufTN64HyCmh7Ojop7vf1Gy2t/WyMygBfeS02bT0WvodQdFnMJWhLSjflJdFaQwqPD+6azPvPvdIAHa2m8rI4lk17OyyV8c8vaeb3Z1Wnt88uB2A1pZeqoJyTlUiRk3CDtMjOzv584a9oRymxnJcUxV/2GiqJ2u2tLGj3dSMYrEoi4NSzzGNNgnoj0/u4b6nWwBIpzI0BxWdW+/byutXLwZgd6fle+rChpxSzJLpVXztD08BcN6xs6kM6ke/WLMVgEtOOYKHggJRVTzG2QvsWP3k0Z28NqgCDaStTrd1JYlHbfvZNQnmVNny2r1Wd619KXZ2WT2dOLua36w3f1cvbCCSsPMhbgI5HNOYpDMo3JRPX0ys15RbOqYtYeivOTQFbGAgk1vuGMjkjvFvH9/JO062c2TjXtvvvLoEDRWxXJ0PKQzVJWL0haf3hpR1BtOZnNLPYDpLRYWVfSDYtfSlSMTsfN7dk8qJCEzP9lAWtfI0BPmfTJZc/kCuHo5urGFapsvqt87ypz9DbbrbtiuvpXPQtovvfYrWSlPZqYzbflOZLNMjVs+98ToSFbacKIuTLSvP+Q7QEonTlLayEY1Bxuq0YobtNz5vCYS8MhX17PnLGgBmnryCGSccBUDro2p1n0xRVmXnWGzGbGIt9mjIYGcPg532X2q84OW2r2e2s+chU/dpPGk5m26xPMqn7VPsqT2iiVLiXShOyWlt8Xd+Oc6hIDMJHqX3AO44zpQk410ojuM4E5Px/IBOsXgAdxxnSuIB3HEcZ4IyGQYxi3qUXkSyIvLFvN8fCu//dhzHmZBks9miP+OVYt+FMgD8jYjMHEtnHMdxDhXpVLboz3il2C6UFPZSqvcD/5a/QkQWAt8DZgJ7gLcBHcCjwCJVzYhINbAeWKyqydK47jiOc/BMhj7wkbyN8JvAW0SkviD968APhlR4gK+pagfwMPDiYPMq4GYP3o7jjBem1Otkg3TaD4H3Fqw6DfhJWP4RcGZY/jlwcVh+U/jtOI4zLshks0V/xisjFXT4CvAOoLoI2+uAC4Kww0nAn0a4L8dxnDFjSrXAIadz+QssiA/xF6yFDfAW4M/Bthu4H/gqcL2qpkftreM4TomYDAH8YOaBfxF4d97v9wDfF5EPs28Qc4ifA7/EBI8dx3HGDempIuigqjV5y7uAqrzfW4CX7Ge7XwGRUfroOI5TcsZzy7pY/ElMx3GmJIfqScwwDvhzYCGwGXijqrYNY/d54JVY1/atwD8HHeL9MtJBTMdxnElBNpMu+jNKPgr8UVWXAn8Mv5+DiJwOnAEcB6wATmHfNOz94gHccZwpySEM4BcCPwjLPwAuGs4doAJIAOVAHNh1oIzHVRdKS+8gC6aVc+ocm6X4kZueBuClyxv5xh0bADhh6UwWN9r6GVWmsrO3d5CH79sGwKXvPYNvb34MgGTmJNZvtjuVxxZMAyAWifCDP1i+//OOU7g1KNS87St3MWdRAwB3BdWb85Y38uj2DgCe2tXF7m2dAER723jgWTuoj2xtB+CilXNZ2WzKN1+5YyOrgtLPi5bOZOE0GzJYEzXbRFmU8qBqc77MyinFvPmEuQBc8NFrmXvuEsBUfAZStq+XL53JH4MK0QmhPP9201N84MWLAPjr5jYSId94eRl9g7ZdRUhbOa+eJ3eZsktH9yB/Wb8bMHWeufWVAEwLdVoVj+VUh3Z1D9LXbYo71929hRPn2rNc86ZbuXZ2D/CSxTMAeHBHJ0Pc9sRu/vb4Zqv/UHd3bdjL2UfODPW0lU+eLwBUJ8zH2kSM7zz8LACr5k2jKh6UieJR9gTFnQeetbzePKuNTJn5ksnUU5MIyjnd29lVMQcADSpKD+/o5PT5dnwHUhmiERuaOba5jsZBq4fBOvNrTiX8aavVUzKTpb7C8i3raaGs0spZWWbbH5no5ek+GyKaWRmjLWHrW/vsv1cTj9IV1HL6kqncsYgmO9iVtHyHlI8WVHewh1nmYzpDNChTzapKsDNryjl93XZMFm24h9TKVwJQ8eyjzIjaXznZJDz2tNXPyxZbebMAO+3/U1VeTbTfyhapm0kmqBzNiwe/BnuI1JuKUnraHGIdO8hn4LF7KJtj5xtHnUnjeTb8Faufwe7fXwtA4wUXAJDatoEtN9xlx+TUs6haasc6uWsrDWefZ/tos7qf8ZpLKLvlNwDsfmA99zxt/9sjd7cTTYxNmMokB8ck32FoUtWhitwJPE9aSFXvEZHbgB3YuOE3VPWJA2U8rgK4Y8HbcZyxZyQtaxG5HLg8L+kqVb0qb/0fgNnDbPqcV4+oalZEntevLSJHAkcB80LSrSJylqr++YX88gDuOM6UZCQBPATrq15g/Xn7Wyciu0SkWVV3iEgzsHsYs9cC94bnZxCRG7Gn3F8wgHsfuOM4U5JD2Ad+HXBpWL4UGO42+xngxSJSJiJxbADzgF0oHsAdx5mSHMIAfiXwUhF5Cjgv/EZEThaR7wabXwEbgMeAR4BHVPV3B8rYu1Acx5mSlCAwF4Wq7gXOHSZ9DfB3YTkNvHOkeRcVwEUkjV0Z4ti7wX8IfFlVJ/6zqI7jTEnSqUM2C2XMKLYF3qeqKwFEpBF7fWwd8MmxcsxxHGcsOVQt8LFkxF0oqro7TKm5P+hiRrE+ndXYBPRvqup3AETkI8D/AzLAjar6vCeQHMdxDgfZ9BQM4ACqulFEYkAj9pRRh6qeIiLlwN0icguwPKw7VVV7w/sAHMdxxgVTsgU+DC8DjhOR14ff9cBSbLT1+6raC7l3iTuO44wLpmwAF5HFQBqbkB4B3qOqNxfYnD969xzHccaGyRDARzwPXERmAd/GntXPAjcD/xgmnyMiy4IK/a3A20SkKqR7F4rjOOOGTCpZ9Ge8UmwLvFJEHmbfNMIfAV8K676Lvef2QRGJYKo8F6nqTSKyElgjIoPADcDHSum84zjOwTIZWuDFKvLEXmBdBgvMzwvOqnol4akjx3Gc8URmqgRwx3GcycaUnUboOI4z0ZkyXSiO4ziTjcwkeJTe30Y4zrjpygsPtwuOMyU4hG8jHDMi2eyhUWZ2HMdxSou3wB3HcSYoHsAdx3EmKB7AHcdxJigewB3HcSYoHsAdx3EmKB7AHcdxJigewB3HcSYoh/VJTBEZUu2ZG5K2A9ep6hOHzyvHcZyJwWFrgQe9zJ9hghD3hU8E+KmIfLTA9r0icsSh93LyISIzDrcPjuOUhsP2JKaIPAkco6rJgvQEsFZVl+aldQA9wAbgp8AvVXXPCPfXqKq7R2B/o6q+PO/3ycB/YXcJ/wp8D1gFPAlcrqoP5dkep6qPhuU48JFg+zjwH0Myc2F9DPg7YB5wk6renbfu46r6H3m/o8BlwOuCfTrs/9uqevswZbgS+IKqtgT/f4EJTMeBt6rqHXm2s4FPhvWfAN4T9vME8M+quiPPtj7UwUWYLmoWU2e6FrhSVdvzbB8EfgP8VFU37Ke694uIPKmqy0a4zUEfu/3kN/2FJAFFJD7MeTxTVVuGsT0ZOIJw7FR1/TA2CSAZBFMQkXOAE4F1qnrjC/lapL9N5N31ququYWym5R/HkSIi71LVb41wm+WF9SEi84FOVW0XkYXAycB6VX38YH2bTBzOLpQMMAfYUpDeHNblsxE4CdPZvBi4QkQewIL5b1S1K994GPWfCHCfiJwARIZObhE5cT++RYCVBWnfwgLcNOAvwPtV9aUicm5Yd1qe7dXYHw7sfegzgC9iAe/bwFvzbL8DVGF3IF8TkTtU9QNh3d8A/5Fn+79Yff0n8HqgE/gz8HEROVZVv17g8ytVdehu5r+Ai1X1fhFZBvwE+zPk+/x7oBq4Dfgx8Io8n/Nf0vIL4E/AalXdCbkLwKVh3cvybBtCnd0mIjuxY/ZzVX22wFdEpAu7GIAdA4CqoXRVrcuzHZNjl3/RFJGjgWuAeBAruVhV/5pnew4mblIRLlSXq+rmsPoW9p0DiMiLsXOgHTuX7wYaRCQJ/K2qbs3z935gNdAmIh8GXosJonxARM5W1X/Ny/cMTFQlA7wdO18Wh4vAG1X1njzbldixrMcuZgDzRKQdeJeqPpjnQ4uI3I4dr1+/UDAXkQ8UJEWAfxWRCgBV/dLztxqWW4D5efl+FHgnMCAiXwA+hNXbFSLyvyPId9JyOAP4+4A/ishTwNDJOx84Enh3gW02CEfcAtwSWrUvBy4BvgDMKrBv4fkXhrnAg1iAWBzS7gfuYF+wyGdawe/4UOtHRD6nqr8CUNU/hpMrn/z8zgVOUdWkiNwJPFJgu0pVjwv5fgP4loj8JpSt0K+TVPVtYfkuEblXVT8R8n0YKAzgZSJSpqopoFJV7w8+Pyki5QW2TUMXgNB6+lxI/7qIvKPAdmHeekKeO4HPicjbC2zbVPVDwIdE5KxQrgdF5AmsVX5Vnu33sXr/8FCrUEQ2qeoins9YHbv8i+Z/YXcfN4rIKuArwOl5tp8HzlfVtUHU+1YR+VtVvXcYv74CvExV94jIIuBLqnqGiLwUuzDnX/RiqtoWli8GzlLVvnBH9SB2FzHEl4E3AjXYBfgiVb0rXOC+DpyRZ3s18M78i1CokxdhdX98XvITwedLgM+LyF1YML9WVfsKynYFdoFZm1fuGFBbYIeIfK0wLRDh+cftb4GjsQbOZmBxqL9q4K/sUwWbshy2AB4k15Zht7L5g5j3q2rh678iBdsmgeuA64Y0Nwv4MPBSLBA8BvsNBE9gJ/RThRmIyNaCpH4ReRnWesmKyEWqek1oWRX6Wy8ifxP8Lh+6vVbVrIgU9lkl8sqVAi4XkU9iLdyaAtukiCxR1Q3hDzoYthsYJl+w1uUN4Y9/k4h8FevOeAkW8PPJHw/54QusA9giIv8C/CAv0DZh3TuF9ZY7dqr6Z+DPIvIe7PhcDFyVt/69InISNg5yDfAN9rXICxmrY5fPnKHAr6r3iUhlwfqEqq4N638VLkq/CeM7hX7H8rr9ngEWhO1uFZGvFNh2isiK0E3QAlQAfdj/tfBYxPPO8T2qelfI98Fh/K0uDN7B9t4QFPNJqur1wPUhn1cDbwK+KSI3q+qb82yPwe4uqoErVLVXRC5V1SsK9wW8DfggMDDMuksKfqfDhWswlH9v8LdHRIbZfOpxWGehhFb1vUWYXvwCefQOk/ZFEfk58OXwZ/4kwweCT7H/gdz3FPz+B6zFlQHOx4Scr8YuOn9fYHsHdsID3CsiTaq6K3QzFPaLrhGRC1T1pjz/rxCR7cB/F9h+GOuKGMRaOJdATmj6+sICqOrXReQx4B+BZdjxXop1C/xHgfm1IlKjqt2q+vGhRBE5Eusrzudi4KPAHSFwZ4Fd2EX1jYVuDONXGrgpfArXPSAi52F3YXdgwWs4PsXojt33gWeBywtsF4vIddiFZ56IVOWdY/EC26SIzB7qRgot8XOxY7GkwHaNiPwvdmF+DXA7QGiAFEoW/gPwYxF5BBtbWBPuso4FPltgm18H/1qwLlHw+0YR+T12gR66yB2BdekVHov8C28f1jX2C7Hxj4vyDVX1GeANInIhdhfyZfbP/cDjqvqXwhUi8qmCpAdF5CfYheGPwA9E5CasAbLuBfYxZZj0r5MVkddgep0LVXX2MOuXY3cAf1XV7rz05wTVkHYU1m9fjO2pQCb0OR8NXIANvtxQhM8/VNW37mddBJgxNED2QrbDbHsWdsfzmKreMoy/T6hqZ2hxfZQwcAZ8VlU78mzfC/y2oN92f/sciW0Ca+U9q6p/EJG/xW7PvwD8jz5/oHAx1uWRGxQEfqKqncPkvSTYDg3+PgX8uNA2tMrzeVBVu8KF6vWq+s082/OAPar6SEEe9cC7VfUzeWlx7EJ/NNaN9j1VTYe6blTVLQV5xLBulaEL7zbg5sK+6HB+/6GwIRPK+zpV/XxB+ssZfuruDQV2H1LVwu6lAxJa8p8CTlXVs4dZPx3oH67hNYxtGfAGrIHwK+BUrNHyDPBNVe0ZqX+TjUkfwAHCn2SJqj4uIm9T1e+H9PcC/4Tdjq/E+juvDeseVNX8Qaj3Au8C1hdh+0msj74MuBULmrdj3QY3F/yxrytwNwKcg7XUUNXXvIAtWGvkebbB/j5VXRWW/y6U9RosMPxOTXR6yHYtcLyqpkTkKqAX+9OcG9L/Js82f1bQT7BZQc+bcTGM7QvOIBKRH2N1VoUN9lUDvw0+RFT10jzb9wKvAu7EBlsfCtu8FhuQu/1gbKcaMoLZWSOxHaEPM1R1b6ltpwJTQlIt3AIOTTu6AhuwAWsRnaSq3WJTlH4lIgtV9as8fxDq74GTi7R9PRbky4GdwLzQsv0CNvjymTzbI7DBn+9iLY0INjvki8MUZTjbU/ZjC8+95X8n+wbRvoB1XV2Ztz4a+uAJ5Ry6IN0lIoX95YWzgj4t+58VNJIZRMeq6nGh5bUd64NOi8j/8fzB378HVob1XwJuUNXVIvIdbDrjCQdjKyOYIvlCSMFUxpHYikgddtc4N/j607x131LVd+X9Hsn0z8LZWTDM7Kz92A47kyvY5u5ARWQadsc0NG32/Zo3TVH2M7U13KEUTm0dzjYd7tSeYztVmZQBXEQe3c+qCNCU9zs61BWiqptFZDUWmBfw/KA8EttU6OftFZENQ7fpYUCmcIrkScA/A4gUaXkAAAMtSURBVP+GDbo+LCJ9+zk5R2ILEBWRBqyfNDLU8g2DQKkC2/y7k0dE5GRVXSM20JwssB3JrKCR2EbDn7Maa4XXA63YhbCw/xns/E2H9TWhbM+EfRysbdFTJGUEUxlHYos1MJ4Cfg28XWyGy5tVdQB4UYHt1RQ//bPY2Vkjtf0s+/rQv4A1Wl6NdVl9h+f2mY9kautIbKckkzKAY0H6fKCtID2CzQMeYpeIrFTVhwFC6/pV2IMexxZsOxLbQdk3+HXSUGJo3T0ngIfg9mUR+WX43sV+jstIbAP1wAOh3FkRaVbVHSJSw/MvOn8HfFVEPo79ee8RGwDeGtblM5JZQSOx/V+siyqGXaR+KSIbsaD1swLb7wL3i8hfgbOAz0FuQLfwIZaR2I5kiuRIpjKOxHaJqr4uLF8jIv8G/Cn0dxcykumfxc7OGqltPier6tAF6csicmnB+pFMbR2J7ZRksgbw64GaoWCbj9jDCUO8FXhOSzScLG8Nt9ccpO3ZobU0FHSHiGMtueehqtuwkfxXYg/o7JdibVV14X5WZbD+33zbDuCycPu+iDBwpsM8pcfIZgUVbauqXxabPYSqPiv/v507RmkgCKM4/mIjNpbBA8h8CJ5BLD2AteAtrAVBAhYi2AgWHiOksrS1+moPYbkWOyHZjYaZEGQn8//BNMsrZiEMZObNmr2p3Xp5cfePXvbRzGaSTiQ9eLzBF/9lnG2aVV5FMqfKmJPdN7O9+W/H3e+sbSW9a7Vauq7+2Wm3eHo7KysraWztZZ6RpEMzG3m8RarVplBOtTUnW6UqDjGBVHHL6Ubt1sM4Pp5XJO99ccFGcWvj091XqpIWu+YbZieSpu4+6+UuJD159zMTt5ImvtSKis+P43wv/3jPte2snKy1h/bLnuNZy1Gc21Uvf65utfVL7eH669I5THa2Sk3TMBiMhBFCuN6lbAjhIIRwuu3sEN6tlsH3wIF0v90sLDbr7t+++CjU1rI5c/in7M7a1T1wYCMZDSayA8rWigUc6EptMJEdVrZKLOBAV2qDieywslWihQIAheIQEwAKxQIOAIViAQeAQrGAA0ChWMABoFA/t1Mw5t1fJBMAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f979f133ef0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "seaborn.heatmap(donnees_mois.transpose(), cmap = \"RdBu_r\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "donnees_mois = donnees[donnees.columns[1:13]]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ACP \n",
    "\n",
    "Ici, il est préférable de ne pas normer, ni même centrer les données."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PCA(copy=True, iterated_power='auto', n_components=None, random_state=None,\n",
       "  svd_solver='auto', tol=0.0, whiten=False)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "from sklearn.preprocessing import scale\n",
    "\n",
    "pca = PCA()\n",
    "pca.fit(donnees_mois)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>Dimension</th>\n",
       "      <th>Valeur propre</th>\n",
       "      <th>% variance expliquée</th>\n",
       "      <th>% cum. var. expliquée</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Dim1</td>\n",
       "      <td>1.099885</td>\n",
       "      <td>87.0</td>\n",
       "      <td>87.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Dim2</td>\n",
       "      <td>0.060922</td>\n",
       "      <td>5.0</td>\n",
       "      <td>92.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Dim3</td>\n",
       "      <td>0.026101</td>\n",
       "      <td>2.0</td>\n",
       "      <td>94.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Dim4</td>\n",
       "      <td>0.018159</td>\n",
       "      <td>1.0</td>\n",
       "      <td>95.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Dim5</td>\n",
       "      <td>0.014002</td>\n",
       "      <td>1.0</td>\n",
       "      <td>97.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Dim6</td>\n",
       "      <td>0.012945</td>\n",
       "      <td>1.0</td>\n",
       "      <td>98.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Dim7</td>\n",
       "      <td>0.007977</td>\n",
       "      <td>1.0</td>\n",
       "      <td>98.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Dim8</td>\n",
       "      <td>0.007745</td>\n",
       "      <td>1.0</td>\n",
       "      <td>99.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Dim9</td>\n",
       "      <td>0.004904</td>\n",
       "      <td>0.0</td>\n",
       "      <td>99.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Dim10</td>\n",
       "      <td>0.003739</td>\n",
       "      <td>0.0</td>\n",
       "      <td>99.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Dim11</td>\n",
       "      <td>0.003501</td>\n",
       "      <td>0.0</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Dim12</td>\n",
       "      <td>0.003233</td>\n",
       "      <td>0.0</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Dimension  Valeur propre  % variance expliquée  % cum. var. expliquée\n",
       "0       Dim1       1.099885                  87.0                   87.0\n",
       "1       Dim2       0.060922                   5.0                   92.0\n",
       "2       Dim3       0.026101                   2.0                   94.0\n",
       "3       Dim4       0.018159                   1.0                   95.0\n",
       "4       Dim5       0.014002                   1.0                   97.0\n",
       "5       Dim6       0.012945                   1.0                   98.0\n",
       "6       Dim7       0.007977                   1.0                   98.0\n",
       "7       Dim8       0.007745                   1.0                   99.0\n",
       "8       Dim9       0.004904                   0.0                   99.0\n",
       "9      Dim10       0.003739                   0.0                   99.0\n",
       "10     Dim11       0.003501                   0.0                  100.0\n",
       "11     Dim12       0.003233                   0.0                  100.0"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "eig = pandas.DataFrame(\n",
    "    {\n",
    "        \"Dimension\" : [\"Dim\" + str(x + 1) for x in range(12)], \n",
    "        \"Valeur propre\" : pca.explained_variance_,\n",
    "        \"% variance expliquée\" : numpy.round(pca.explained_variance_ratio_ * 100),\n",
    "        \"% cum. var. expliquée\" : numpy.round(numpy.cumsum(pca.explained_variance_ratio_) * 100)\n",
    "    },\n",
    "    columns = [\"Dimension\", \"Valeur propre\", \"% variance expliquée\", \"% cum. var. expliquée\"]\n",
    ")\n",
    "eig"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(3.25,8,'8%')"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEjCAYAAAA/ugbCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAIABJREFUeJzt3Xm8VVXdx/HPZVAcUZzBnJB+pU0Wij2a5pyKpuZYTimWlbNP+WhPOZSlPaahpU8OKRZOKeWYFSihqaikgqZfMxKVB8EBFZGLXLjPH2tdPd4u9xzg7HO57O/79eLF2fucvdZvn3vO+e211t5rN7W2tmJmZuXVo6sDMDOzruVEYGZWck4EZmYl50RgZlZyTgRmZiXnRGBmVnJOBCUQERtExNsR0bOrYylaRIyNiGH58Vci4k91Lr9HRNwaEV+rZ7lFiojnI2Ln/PiMiLiyq2OqlD+bm3R1HGXWq6sDsA+KiLuBhyV9v936LwK/BNaX1LIoZUp6AVi5flF2D5JGAiPrXOwPgXskXV7nchtC0o+6Oob2JJXus7m0cYtg6TMCODQimtqtPwwYuahJICKc7OtI0hmShnd1HOC/rdWPP0hLn98D/wt8DhgHEBGrA0OBIXl5T9KR6UDgTeAqSWfl5zYC/gUMA84Eno+Iw/O63pJaIuKrwHeA9YFXgPMl/TJv/3ngN8BFwGnAfOAMSVfn51fIde8PrAZMAnaRNCcitgYuBDYDpgAnShrb0U5GRH/gEmA74G3gIkkX5+fuAp6WdGpevgF4R9JREXEkcAzwGCk5TgO+JWlMB3UcCQyTtG1e3iXXuR7wa+DjwK8lXRkRZwGbSjq03fvY9p71zfu2B7AAuBo4U9L8/PqjgG8D6wIPA1+TNKWDmNrK/TpwFtAE/FTSBfn5rYDhwEeBOcAtwCmS3s3PtwLHASeRvr8bd1DHYaS/0co55srn3tvPiliOAs7Jrz8dmABcBWwA/EbScRXbL3Q/c2zfAE4F1iK1xo6T1BoRm+YyPwXMA8ZIOqhiu0GSnsvv8yXA7sA7wBXAjyQtaPt7Ag8BRwNvAN+U9If274EtGrcIljKS5gA3AYdXrD4QeEbSE3l5dn5+NWBP4BsRsU+7orYn/Zjs1kE1M0iJZVXgq8BFEfHpiufXBfoCA0hfuF/kZARwAfAZ4D+AfqSEsiAiBgB3kn6A+gH/CdwSEWu1rzwiegC3A0/kOnYCToqItliPAg6LiB0j4ivAVsCJFUUMAf4JrElKdqMiol8H+1lZ55rAKOC/83b/BLbpbJt2rgFagE2BLYBdST9Kbd12ZwD7kX4A7wOur1LeDsCgXM5pbX34pMR7co7xs6T35pvttt2H9B5s1r7QiNgMuIyUJPsDa5ASfmeG5FgOAn4GfBfYGdgcODAitl+E/RwKbAl8gvS5bfub/gD4E7B6jueShcRyCemztwnpM3w46TNaGatI789PgKs6aD3bInIiWDqNAPaPiD55+fC8DgBJYyVNkrRA0kTSl3H7dmWcJWl2TiwfIOlOSf+U1CrpL6Qv6OcqXjIPOEfSPEl3kY7YI/+AH0U60p8qab6kByTNBQ4F7pJ0V47rz8CjpCPo9rYE1pJ0jqR3JU0mHfkdnON7mXRkOYJ0dHy4pFkV288Afpbju5H0w7Bn528pewBPSbpZ0jzSD97LVbaBtOPr5O1Pyu/pDFKL6eD8kmOBH0t6Onfd/Qj4VERs2EmxZ+eyJpFaF4fkfZ8g6SFJLZKeJ40Ltf/b/ljS6x39bUkttTskjct/l++RWjCd+YGkZkl/Ih1kXC9phqSppB/7LRZhP8+T9EYel7qX1AKA9JnaEOif67q/fRD5ZIaDgdMlzcr7/1NSUmszRdIVuSU2gtS6W6fK/lkV7hpaCkm6PyJeBfaJiEdIR8T7tT0fEUOA84CPAcsBywO/bVfMiwsrPyJ2Jx1Jf5h0MLAiqYunzWvtxiLeIXUbrAn0IR1Nt7chcEBE7FWxrjfpx6Cj1/aPiDcq1vUk/ei0uZ10dKgOfjSmSqqcLXEK6ei3M/2peE9yd8VC36MO4u0NTIuItnU9KsrbEBgeET+t2KaJ1Nr5t+6hrLLuKaRuKiLiw6TunMGkv0svUlfNwrZtr/1+zo6I1zp5PcD0isdzOlhuG8ytZT8rk+s7Fdt+h9QqeDgiZpK6w37VLo41Se9z5Xs2JZff5r3yJb2T/x4ebF5CTgRLr2tJLYEA/iip8st5HfBzYHdJzRHxM9KXqFKH08pGxPKkfufDgVslzYuI35O+0NW8CjSTxiaeaPfci6T+9mNqKOdF4F+SBnXymnOBp4GNI+IQSZVdEAMioqkiGWwA3FalzmnAh9oWcnfChyqen0364W2zbrt45wJrLmSw/kXg3HyWUq0+BDyTH28A/F9+fBlp/OMQSbMi4iTSUX6lzqYMnkbqEgQgIlYkdQ/Vw+LsJ/BeK++YHNO2wOiIGCfpuYqXvcr7LYe/53UbAFOXKGqryl1DS69rSf20x1DRLZStAryek8BWwJcXody2FsQrQEtuHexay4aSFgC/Ai6MiP4R0TMiPpuTy2+AvSJit7y+T0R8PiI66p9+GJgVEadFxAr59R+LiC0BImI7Ur/w4cARwCV5DKLN2sAJEdE7Ig4g/fDdVSX8O4HNI2K/fLbNCXzwx/5xYLt8zUVf0qBp235PI3Wf/TQiVs3XEgxs6zsnDe6fHhGb5/j75rg6872IWDFv81Xgxrx+FeAt4O2I+Aipi2xR3AwMjYhtI2I50iBwvb7ni7Of5NceUPFZmElKZh/ossrdPTcB50bEKrnL6RTSZ8sK5ESwlMr9ow8AK/HvR7vfBM6JiFnA90lfnlrLnUX6EbyJ9IX8cgfld+Y/Sd1IjwCvA+cDPSS9CLQNJr5COnr8Nh18xvIXfiip//hfpCPBK4G+EbEqKQkel8ch7iOdbXJ1xaDgeNLg5quklsP+kjrt/pD0KnAAqUvttbz9Xyue/zPpx3giqSvmjnZFHE5Kon8nvW83k/qnkfS7/D7cEBFvAU+SznrpzF+A54AxwAW5fx7S+/tlYBZp3OTGjjdf6H4+BXyL1GqclmN9aVHK6KTsxdnPNlsC4yPibdLn7cQ8NtTe8aTW2WTgftJ+tO9Csjpr8o1prDtpf0roEpY1lnR6ZMOutG1/Wmqj6jXrjFsEZmYl50RgZlZy7hoyMys5twjMzErOicDMrOScCMzMSs6JwMys5JwIzMxKzonAzKzknAjMzErOicDMrOScCMzMSs6JwMys5JwIzMxKzonAzKzknAjMzErOicDMrOS6xc3rhwwZ0jpgwIDqLzQzs/c89dRTr0paq9rrukUiGDBgAKNGjerqMMzMupWImFLL69w1ZGZWck4EZmYl50RgZlZyTgRmZiXnRGBmVnJOBGZmJedEYGZWck4EZmYl50RgZlZy3S4RzJ3f0q3KNTNb2nWLKSYqLd+zF58ZcUXdy51wxDF1L9PMrDvodi0CMzOrLycCM7OScyIwMys5JwIzs5JzIjAzKzknAjOzknMiMDMrOScCM7OScyIwMys5JwIzs5JzIjAzK7lC5xqKiJOBYUArMAn4KrAecAOwBjABOEzSu0XGYWZmC1dYiyAiBgAnAIMlfQzoCRwMnA9cJGlTYCZwdFExmJlZdUV3DfUCVoiIXsCKwDRgR+Dm/PwIYJ+CYzAzs04UlggkTQUuAF4gJYA3SV1Bb0hqm/z/JWBAUTGYmVl1RXYNrQ58EdgY6A+sBHyhqPrMzGzxFNk1tDPwL0mvSJoHjAK2AVbLXUUA6wNTC4zBzMyqKPKsoReArSNiRWAOsBPwKHAvsD/pzKEjgFsLjMHMzKoocoxgPGlQ+G+kU0d7AJcDpwGnRMRzpFNIryoqBjMzq67Q6wgknQmc2W71ZGCrIus1M7Pa+cpiM7OScyIwMys5JwIzs5JzIjAzKzknAjOzknMiMDMrOScCM7OScyIwMys5JwIzs5JzIjAzKzknAjOzknMiMDMrOScCM7OScyIwMys5JwIzs5JzIjAzKzknAjOzknMiMDMrOScCM7OScyIwMyu5qjevj4g+wFDgc0B/YA7wJHCnpKeKDc/MzIrWaSKIiLNJSWAsMB6YAfQBPgycl5PEqZImFhynmZkVpFqL4GFJZy7kuQsjYm1ggzrHZGZmDdRpIpB0Z/t1uRWwnKS3JM0gtRLMzKybWqTB4ogYBvweuCUiflxMSGZm1kidJoKI2Lvdqp0lfUHSLsAexYVlZmaNUm2M4OMRcTRwpqTHgYkRcSXQCviMITOzZUC1MYJzI2Jd4JyIaAK+B6wCrOAzhczMlg1VryMAZgMnAYOAy4FHgZ8UGZSZmTVOtTGCHwK3AHcAO0jaG3gcuCsiDm9AfGZmVrBqZw0NlbQrsBNwOICk24BdgdULjs3MzBqgWtfQkxFxObAC8Je2lZJagOFFBmZmZo1RbbD40Ij4ODBP0jMNisnMzBqo2lxD20q6v5PnVwU2kPRk3SMzM7OGqNY19KWI+AlwNzABeIU06dymwA7AhsCphUZoZmaFqtY1dHJE9AO+BBwArEeahvpp4JedtRbMzKx7qHodgaTXgSvyPzMzW8b4DmVmZiXnRGBmVnK1TDGx2CJiNeBK4GOkieqOAgTcCGwEPA8cKGlmkXGYmdnC1dQiiIgVI+J7EXFFXh4UEUNr2HQ4cLekjwCfJA0y/xcwRtIgYExeNjOzLlJr19DVwFzgs3l5KvDDzjaIiL7AdsBVAJLelfQG8EVgRH7ZCGCfRYzZzMzqqNZEMFDST4B5AJLeAZqqbLMx6bqDqyPisYi4MiJWAtaRNC2/5mVgncWI28zM6qTWRPBuRKxA6ucnIgaSWgid6QV8GrhM0hak6aw/0A0kqbWtTDMz6xq1JoIzSVcXfygiRpL69r9TZZuXgJckjc/LN5MSw/SIWA8g/z9jkaM2M7O6qSkRSPozsB9wJHA9MFjS2CrbvAy8GBGRV+0E/B24DTgirzsCuHWRozYzs7qp9ayhJmB34DOS7gBWjIitatj0eGBkREwEPgX8CDgP2CUi/gHsnJfNzKyLLPQ6gojYFnhQ0nzgUmABsCNwDjCLdOeyLTsrPN/wfnAHT+20uAGbmVl9ddYiWABclh8PkfQtoBkgXwC2XMGxmZlZAyw0EUh6gNQSAJgXET15/6yhtUiJwszMurlOxwhy1w7AxcDvgLUj4lzgflJ/v5mZdXM1zTUkaWRETCD17TcB+0h6utDIzMysIWo9a2gD4B3gdtLpn7PzOjMz6+ZqnX30TtL4QBPpVpUbk2YR3byguMzMrEFq7Rr6eOVyRHwa+GYhEZmZWUMt1o1pJP0NGFLnWMzMrAvU1CKIiFMqFnuQ5gz6v0IiMjOzhqp1jGCVisctpDGDW+ofjpmZNVqtYwRnFx2ImZl1jVq7hm6nk/sGSNq7bhGZmVlD1do1NBlYF/hNXj4EmA78voigzMyscWpNBNtIqpxF9PaIeFTSyUUEZWZmjVPr6aMrRcQmbQsRsTGwUjEhmZlZI9XaIjgZGBsRk0lXF28IfL2wqMzMrGFqPWvo7ogYBHwkr3pGUrWb15uZWTfQaddQROyY/98P2BMYmP/tmdeZmVk3V61FsD1wD7BXB8+1AqPqHpGZmTVUp4lA0pn5/682JhwzM2u0ThNBuzmG/o2kC+sbjpmZNVq1rqFVqjxvZmbdXLWuIc8xZGa2jKt1rqFNgOHA1qRB4geBkyVNLjA2MzNrgFqvLL4OuAlYD+gP/Ba4vqigzMyscWq9snhFSb+uWP5NRHy7iIDMzKyxak0Ef4iI/wJuIHUNHQTcFRH9ACS9XlB8ZmZWsFoTwYH5//bzCx1MSgybYGZm3VKtcw1tXHQgZmbWNWoaLI6IH0REz4rlVSPi6uLCMjOzRqn1rKFewMMR8YmI2AV4BJhQXFhmZtYotXYNnR4Ro4HxwExgO0nPFRqZmZk1RK1dQ9sBFwPnAGOBSyKif4FxmZlZg9R61tAFwAGS/g7v3Z/gHt6/UY2ZmXVTtY4RfLYtCQBIGgVsU0xIZmbWSLUmgjUj4qqIuBsgIjYD9ikuLDMza5RaE8E1wB9Jcw0BPAucVERAZmbWWDW3CCTdBCwAkNQCzC8sKjMza5haB4tnR8QapOkkiIitgTdr2TBfiPYoMFXS0IjYmDRn0RqkaxEOk/TuIkduZmZ1UWuL4BTgNmBgRPwVuBY4vsZtTwSerlg+H7hI0qakaxKOrrEcMzMrQE2JQNLfgO2B/yBNPLe5pInVtouI9YE9gSvzchOwI3BzfskIPOhsZtalau0aahsXeGoRy/8Z8B3ev/fxGsAbuSyAl4ABi1immZnVUa1dQ4ssIoYCMyR5TiIzs6VYYYmAdMHZ3hHxPGlweEfSfY9Xi4i2lsj6wNQCYzAzsypqnWuoKSIOjYjv5+UNImKrzraRdLqk9SVtRLqBzT2SvgLcC+yfX3YEcOtiR29mZkus1hbBpcBngUPy8izgF4tZ52nAKRHxHGnM4KrFLMfMzOqg1sHiIZI+HRGPAUiaGRHL1VqJpLGkWUuRNBnotDVhZmaNU2uLYF6+MKztgrK1yFcZm5lZ91ZrIrgY+B2wdkScC9wP/KiwqMzMrGFqvUPZyIiYAOwENAH7SHq6ymZmZtYN1HrW0NakuYJ+IennwNSIGFJsaGZm1gi1dg1dBrxdsfx2XmdmZt1crYmgSVJr24KkBSzC9BRmZrb0qvXHfHJEnMD7rYBvApOLCcnMzBqp1hbBsaSZR6eSJoobAnytqKDMzKxxaj1raAZpmggzM1vG1JQI8gVkxwAbVW4j6ahiwjIzs0apdYzgVuA+YDS+V7GZ2TKl1kSwoqTTCo3EzMy6RK2DxXdExB6FRmJmZl2i1hbBicAZETEXmEeaZqJV0qqFRWZmZg1R61lDq1R/lZmZdUc1Xx0cEasDg4A+beskjSsiKDMza5xaTx8dRuoeWh94HNgaeJB0H2IzM+vGah0sPhHYEpgiaQdgC+CNwqIyM7OGqTURNEtqBoiI5SU9A0RxYZmZWaPUOkbwUkSsBvwe+HNEzASmFBeWmZk1Sq1nDe2bH54VEfcCfYG7C4vKzMwaptNEEBGrSnorIvpVrJ6U/18ZeL2wyMzMrCGqtQiuA4YCE4BW8oVkFf9vUmh0ZmZWuE4TgaShEdEEbC/phQbFZGZmDVT1rKF8i8o7GxCLmZl1gVpPH/1bRGxZaCRmZtYlaj19dAjwlYiYAszm/UnnPlFYZGZm1hC1JoLdCo3CzMy6TK3XEUwBiIi1qZh0zszMur9aJ53bG/gp0B+YAWwIPA1sXlxoZmbWCLUOFv+ANOPos5I2BnYCHiosKjMza5haE8E8Sa8BPSKih6R7gcEFxmVmZg1S62DxGxGxMjAOGBkRM0hnD5mZWTdXa4vgi8Ac4GTSZHP/BPYqKigzM2ucapPO/QK4TtJfK1aPKDYkMzNrpGpdQ88CF0TEesBNwPWSHis+LDMza5Rqk84NB4ZHxIbAwcCvImIF4HpSUni2ATGamVmBahojkDRF0vmStgAOAfYhXUdgZmbdXK0XlPUCdie1CnYCxgJnFRaVmZk1TLXB4l1ILYA9gIeBG4CvSap66mhEfAi4FliHdBObyyUNz3c7uxHYCHgeOFDSzCXYBzMzWwLVuoZOBx4APippb0nX1ZIEshbgVEmbka5K/lZEbAb8FzBG0iBgTF42M7MuUm2weMfFLVjSNGBafjwrIp4GBpCuSfh8ftkIUjfTaYtbj5mZLZlaLyhbIhGxEbAFMB5YJycJgJdJXUdmZtZFCk8EeWqKW4CTJL1V+Vy+DWZr0TGYmdnCFZoIIqI3KQmMlDQqr56eL1Aj/z+jyBjMzKxzhSWCiGgCrgKelnRhxVO3AUfkx0cAtxYVg5mZVVfr7KOLYxvgMGBSRDye150BnAfcFBFHA1OAAwuMwczMqigsEUi6n3ST+47sVFS9Zma2aBpy1pCZmS29nAjMzErOicDMrOScCMzMSs6JwMys5JwIzMxKzonAzKzknAjMzErOicDMrOScCMzMSs6JwMys5JwIzMxKzonAzKzknAjMzErOicDMrOScCKyurrnmGvbcc0+GDh3KKaecwty5czn11FPZa6+9uPDC929Ud+mllzJ69OgujNTM2jgRWN1Mnz6da6+9lltuuYU77riD+fPnc+ONN9KnTx9uv/12Jk2axKxZs5gxYwYTJ05k55137uqQzQwnAquz+fPn09zcTEtLC83NzbS2ttLc3MyCBQtoaWmhR48eXHzxxRx//PFdHaqZZU4EVjfrrLMORx11FDvssAPbbrstK6+8MkcccQT9+vVj3333ZYcdduCFF15gwYIFbL755l0drpllRd683krmzTffZMyYMYwZM4ZVVlmFE088kVtvvZXvfve7773m2GOP5eyzz+ayyy7jmWeeYZtttuHAAw/swqjNzC0Cq5sHHniA9ddfn379+tG7d2923XVXHnvssfeeHz16NJtvvjnvvPMOL7zwAsOHD+ePf/wjc+bM6cKozcyJwOqmf//+PPHEE8yZM4fW1lYefPBBBg4cCMC8efMYMWIEw4YNY+7cuTQ1NQFpTGHevHldGbZZ6blryOrmk5/8JLvtthv77rsvvXr14qMf/SgHHXQQACNHjmTfffdlhRVWICJobm5mr732YrvttmPVVVft4sjNyq3nWWed1dUxVHXttdee1bdvX1ZaaSUmTJjAU089RXOfPqw3bQa9Wuaz/Nx3WfPV15nXuzdrvvoaq771NnOX7826L79Cz/kL6NM8lzVem8nc5ZdjnemvstLs2bT06sU601+hx4JWVnxnDv1em8naa6/N6NGjmTp1Kssttxzjxo0DYPLkyTz22GPvPT99+nSampq477776NmzJ5J44okn3nv+tdde49133+WBBx6gT58+TJw4kUmTJr33/Jtvvsns2bN58MEHP7BPbc/Pnj2bmTNnMn78ePr27ctDDz3Es88+y+qrr84999xDc3Mz06dP55FHHmGNNdZg3LhxPP/886y00kqMHTuWlpYWXnrpJSZMmNDwfVqwYAFf//rXiQj69+/Peuutx+jRoxk0aBDrrrsu48ePZ7XVVqNv374MHjyYoUOHLvX7tCz+nbxP5din3/72t9OOP/74y6v9xja1trbW/5e7zvbbb7/WUaNGvbf8mRFX1L2OCUccU/cyzcy6UkRMkDS42us8RmBmVnJOBGZmJedEYGZWck4EZmYl50RgZlZyTgRVzJ3f0q3KNTNbVL6grIrle/by6apmtkxzi8DMrOScCMzMSs6JwMys5JwIzMxKzonAzKzknAiWMj5d1cwarctOH42ILwDDgZ7AlZLO66pYliY+XdXMGq1LWgQR0RP4BbA7sBlwSERs1hWxlF2jWyCur371FdnKcwuyXLqqRbAV8JykyQARcQPwReDvXRRPaTW6BeL66ldfUXUtrD5bdnXVGMEA4MWK5ZfyOjMza7AuuUNZROwPfEHSsLx8GDBE0nELef0rwJQGhmhmtizYUNJa1V7UVV1DU4EPVSyvn9d1qJYdMTOzxdNVieARYFBEbExKAAcDX+6iWMzMSq1LxggktQDHAX8EngZukvRUV8RiZlZ2XTJGYGZmSw9fWWxmVnJOBGZmJdet7lAWEfOBSUBvoAW4FrhI0oKIGAwcLumERSjvOOAkYCCwlqRXC65vJDAYmAc8DHxd0rwC67sq19cEPAscKentouqrKPdi4ChJK7dbX+/9uwbYHngzrzpS0uMF1dUE/BA4AJgPXCbp4gL37T5glby4NvCwpH0KrG8n4H9IB4dvk97L5wqsb0fgAmA5YAJwNDC3znV0+P3Of8vhwPFAM/B/eZ+Lqu8jwNXA1rmuNwrev68Ap5G+97OAb0h6orOyulUiAOZI+hRARKwNXAesCpwp6VHg0UUs76/AHcDYBtU3Ejg0P74OGAZcVmB9J0t6K5d3IWmAvnJOp3rXR/5Ar76Qp+teH/BtSTc3oK4jSac8fyR/cdcusj5Jn2t7HBG3ALcWWR/pc/hFSU9HxDeB/ybtc93ri4gewAhgJ0nPRsQ5wBEF7NPCvt+7A4OA2cDOpKSwV4H1vQ6ckJ+/SNIFBe/fv4DtJc2MiN2By4EhnRXUrQaLI+LtyqPMiNiEdCrqmqQjw/+UNDQizgI2BjYBNgBOJmXj3Umnq+7V7kj8eWBwBy2CQurLZZ0MrCnpuw3YvybgUuB5SecXVV+eQ2o06VTgf3TQIqh3fdcAd3SUCAqo62Hgy5VHyUXWV1HOqqSLKTdsS+oF7Z9IR6TjI+J0YBVJZxRRH7Aa8JCkgbmszwGnA9s14vsdEb8k/XheIWnlvO+fB1Yoor6K9e8CZ0i6oMj9a1fn6sCTkjqduaFbjxHkuYp6kprO7Q0EdgT2Bn4D3Cvp48AcYM+urC8iegOHAXcXXV9EXA28DHwEuKTg+o4DbpM0rbN66rl/wLkRMTEiLoqI5QusayBwUEQ8GhF/iIhBDdg3gH2AMZVJoKD6hgF3RcRLpM9mp7MBL2F9rwK9cusRYH8+eIFpvfZpYTqc4qa7/p5UcTTwh2ov6taJoIo/5Kw5ifRmt/3oTgI26uL6LgXGSbqv6PokfRXoT7pe46Ci6ouI/qT+806TTb3qy49PJyW4LYF+pH7RoupaHmiWNBi4AvjVYtZVa31tDgGuX4K6aq3vZGAPSeuT+rMvLKo+Sa2ki0gvyi2tWaRxl7rVsQSxLzP1RcQOpERQ9XvRrRNBblrNB2Z08PRcAEkLgHn5wwewgMUcG6lHfRFxJrAWcEoj6suvmQ/cAHypwPq2ADYFnstN1RUjosNulHrtn6RpklolzSX9eG1VVF2ko8ZR+fHvgE8UuW+5jDXzPt3ZWV1LWl9ErAV8UtL4vP5G4D+K3D9JD0r6nKStgHGkkxnqWkcnOpzipjv+nnRS9ieAK0njPq9Ve323TQT5w/u/wM8r3qSlur6IGAbsBhyS/8CF1RcRTRGxadtjUhPzmaLqk3SnpHUlbSRpI+AdSZsWVV8uY738fxOpC+XJouoCfg/skB9vTwc/XHWuD1KXyR2Smjt7UR3qmwn0jYgP5+VdSC3IouprGwwmd+edlsurax39uVLVAAAC1klEQVSduA04PNezNemss5YC6/s3Re5fRGxAOmg5TNJCP6eVuttZQytExOO8f3rZr1mCJmxEnAB8B1gXmBgRdynPiFpEfaQ//BTgwYgAGCXpnILqawJG5MHGJuAJ4BvtXlPv/aum3vWNzF+oJuBx4NgC6zov13cy6VTDYe2eL+K9PJiF99XXrT5JLRFxDHBLRCwgJYajiqov+3ZEDCUdjF4m6Z6IaNT3+y5gD+ALwL2kUzpHF1VfRKxLOiOoN3B+RPyYdFA2oqD9+z6wBnBp/p1pyV2aC9WtzhoyM7P667ZdQ2ZmVh9OBGZmJedEYGZWck4EZmYl50RgZlZy3e30UbOaRUGzq9YhrgckdXrBllkj+fRRW2ZVTpQW78/2+FdJZ3ZtZGZLFycCW2YVMEPnZ0gXAK1MmjjtSEnTImIsMJ505fFqwNGS7ouIzUlTXyxH6ob9kqR/tMWVr4j+Sa6nFfihpBsj4vPAWbmOj5Hm6z+0EVe8Wjl5jMBKY0lme8wzxl4C7C/pM6RJ586t2L5XnjfnJKCtxXEsMFxpjv3BpPmKKu0HfAr4JGle/P9pmzaDNHfTScBmpAS1zeLut1k1TgRmSbXZHoN0dP7nPA3Cf5MmK2vTNiHdBN6fHfJB4IyIOI10P4E57ercFrhe0nxJ04G/kGZShXRHspfynFSPU8wMl2aAE4GVyBLO9tgEPCXpU/nfxyXt2n77XH7bDJvXkVoYc0hz/e+4COHOrXj8XplmRXAisFKow2yPAtaKiM/m8nrnMYDO6twEmKx0b+Nb+fepq+8j3eymZ45vO9K9rM0ayonAlmUrRMTjEfEUaXbJPwFnL05Bkt4lTQt9fkQ8QequqXYK6IHAk7kr6WOk01cr/Q6YSJoZ9h7gO5JeXpz4zJaEzxoyMys5twjMzErOicDMrOScCMzMSs6JwMys5JwIzMxKzonAzKzknAjMzErOicDMrOT+H/rEYNEEJ1ZeAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f979bd6d198>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "g_eig = seaborn.barplot(x = \"Dimension\", \n",
    "                y = \"% variance expliquée\",\n",
    "                palette = [\"lightseagreen\"],\n",
    "                data = eig)\n",
    "g_eig.set(ylabel = \"Variance expliquée (%)\")\n",
    "g_eig.figure.suptitle(\"Variance expliquée par dimension\")\n",
    "plt.axhline(y = 8, linewidth = .5, color = \"dimgray\", linestyle = \"--\") # 8 = 100 / 12 (nb dimensions)\n",
    "plt.text(3.25, 8, \"8%\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>Dim1</th>\n",
       "      <th>Dim2</th>\n",
       "      <th>Year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-1.012091</td>\n",
       "      <td>-0.337605</td>\n",
       "      <td>1850.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.471109</td>\n",
       "      <td>-0.376655</td>\n",
       "      <td>1851.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.519207</td>\n",
       "      <td>-0.452940</td>\n",
       "      <td>1852.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.638864</td>\n",
       "      <td>0.080396</td>\n",
       "      <td>1853.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0.571057</td>\n",
       "      <td>-0.094629</td>\n",
       "      <td>1854.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Dim1      Dim2    Year\n",
       "0 -1.012091 -0.337605  1850.0\n",
       "1 -0.471109 -0.376655  1851.0\n",
       "2 -0.519207 -0.452940  1852.0\n",
       "3 -0.638864  0.080396  1853.0\n",
       "4 -0.571057 -0.094629  1854.0"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "donnees_pca = pca.transform(donnees_mois)\n",
    "# Transformation en DataFrame pandas\n",
    "donnees_pca_df = pandas.DataFrame({\n",
    "    \"Dim1\" : donnees_pca[:,0], \n",
    "    \"Dim2\" : donnees_pca[:,1], \n",
    "    \"Year\" : donnees.Year\n",
    "})\n",
    "\n",
    "# Résultat (premières lignes)\n",
    "donnees_pca_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f979bab1470>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "g_pca = seaborn.scatterplot(\"Dim1\", \"Dim2\", data = donnees_pca_df)\n",
    "for x, y, v in zip(donnees_pca_df.Dim1, donnees_pca_df.Dim2, donnees_pca_df.Year):\n",
    "    g_pca.text(x, y, int(v))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## DBSCAN\n",
    "\n",
    "La banque mondiale fournit un grand nombre de données, dont des indicateurs de gouvernance au niveau mondial (voir [ici](https://data.worldbank.org/data-catalog/worldwide-governance-indicators)). Le code ci-dessous importe les données du fichier [WGI_Data.csv](https://fxjollois.github.io/cours-2017-2018/analyse-donnees/WGI_Data.csv) (que vous devez donc télécharger) pour les importer. Les informations concernant la définition des indicateurs et leur source se trouvent dans le fichier [WGI_Definition and Source.csv](https://fxjollois.github.io/cours-2017-2018/analyse-donnees/WGI_Definition%20and%20Source.csv)."
   ]
  },
  {
   "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>Country Name</th>\n",
       "      <th>Country Code</th>\n",
       "      <th>Series Name</th>\n",
       "      <th>Series Code</th>\n",
       "      <th>Value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>AFG</td>\n",
       "      <td>Control of Corruption: Estimate</td>\n",
       "      <td>CC.EST</td>\n",
       "      <td>-1.56373512744904</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>AFG</td>\n",
       "      <td>Government Effectiveness: Estimate</td>\n",
       "      <td>GE.EST</td>\n",
       "      <td>-1.2221063375473</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>AFG</td>\n",
       "      <td>Political Stability and Absence of Violence/Te...</td>\n",
       "      <td>PV.EST</td>\n",
       "      <td>-2.74916744232178</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>AFG</td>\n",
       "      <td>Regulatory Quality: Estimate</td>\n",
       "      <td>RQ.EST</td>\n",
       "      <td>-1.32791936397552</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>AFG</td>\n",
       "      <td>Rule of Law: Estimate</td>\n",
       "      <td>RL.EST</td>\n",
       "      <td>-1.61642956733704</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Country Name Country Code  \\\n",
       "0  Afghanistan          AFG   \n",
       "1  Afghanistan          AFG   \n",
       "2  Afghanistan          AFG   \n",
       "3  Afghanistan          AFG   \n",
       "4  Afghanistan          AFG   \n",
       "\n",
       "                                         Series Name Series Code  \\\n",
       "0                    Control of Corruption: Estimate      CC.EST   \n",
       "1                 Government Effectiveness: Estimate      GE.EST   \n",
       "2  Political Stability and Absence of Violence/Te...      PV.EST   \n",
       "3                       Regulatory Quality: Estimate      RQ.EST   \n",
       "4                              Rule of Law: Estimate      RL.EST   \n",
       "\n",
       "               Value  \n",
       "0  -1.56373512744904  \n",
       "1   -1.2221063375473  \n",
       "2  -2.74916744232178  \n",
       "3  -1.32791936397552  \n",
       "4  -1.61642956733704  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wgi_one = pandas.read_csv(\"WGI_Data.csv\")\n",
    "wgi_one.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Control of Corruption',\n",
       " 'Government Effectiveness',\n",
       " 'Political Stability and Absence of Violence/Terrorism',\n",
       " 'Regulatory Quality',\n",
       " 'Rule of Law',\n",
       " 'Voice and Accountability']"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "[s.replace(\": Estimate\", \"\") for s in wgi_one[\"Series Name\"].unique()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": true
   },
   "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>Series Code</th>\n",
       "      <th>CC.EST</th>\n",
       "      <th>GE.EST</th>\n",
       "      <th>PV.EST</th>\n",
       "      <th>RL.EST</th>\n",
       "      <th>RQ.EST</th>\n",
       "      <th>VA.EST</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Country Name</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",
       "      <th>Afghanistan</th>\n",
       "      <td>-1.563735</td>\n",
       "      <td>-1.222106</td>\n",
       "      <td>-2.749167</td>\n",
       "      <td>-1.616430</td>\n",
       "      <td>-1.327919</td>\n",
       "      <td>-1.086220</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Albania</th>\n",
       "      <td>-0.397567</td>\n",
       "      <td>-0.000573</td>\n",
       "      <td>0.257720</td>\n",
       "      <td>-0.347468</td>\n",
       "      <td>0.188962</td>\n",
       "      <td>0.159191</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Algeria</th>\n",
       "      <td>-0.691359</td>\n",
       "      <td>-0.544695</td>\n",
       "      <td>-1.139282</td>\n",
       "      <td>-0.849319</td>\n",
       "      <td>-1.171843</td>\n",
       "      <td>-0.877241</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Samoa</th>\n",
       "      <td>1.230341</td>\n",
       "      <td>0.531472</td>\n",
       "      <td>1.192534</td>\n",
       "      <td>1.328246</td>\n",
       "      <td>0.136647</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Andorra</th>\n",
       "      <td>1.230341</td>\n",
       "      <td>1.860201</td>\n",
       "      <td>1.396047</td>\n",
       "      <td>1.564011</td>\n",
       "      <td>0.871409</td>\n",
       "      <td>1.197540</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Series Code       CC.EST    GE.EST    PV.EST    RL.EST    RQ.EST    VA.EST\n",
       "Country Name                                                              \n",
       "Afghanistan    -1.563735 -1.222106 -2.749167 -1.616430 -1.327919 -1.086220\n",
       "Albania        -0.397567 -0.000573  0.257720 -0.347468  0.188962  0.159191\n",
       "Algeria        -0.691359 -0.544695 -1.139282 -0.849319 -1.171843 -0.877241\n",
       "American Samoa  1.230341  0.531472  1.192534  1.328246  0.136647       NaN\n",
       "Andorra         1.230341  1.860201  1.396047  1.564011  0.871409  1.197540"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wgi_bis = wgi_one[[\"Country Name\", \"Series Code\", \"Value\"]][wgi_one.Value != \"..\"]\n",
    "wgi_ter = wgi_bis.astype({\"Value\": float})\n",
    "wgi = wgi_ter.pivot_table(values = \"Value\", index = \"Country Name\", columns = \"Series Code\")\n",
    "wgi.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "wgi = wgi.dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": true
   },
   "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>Series Code</th>\n",
       "      <th>CC.EST</th>\n",
       "      <th>GE.EST</th>\n",
       "      <th>PV.EST</th>\n",
       "      <th>RL.EST</th>\n",
       "      <th>RQ.EST</th>\n",
       "      <th>VA.EST</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>202.000000</td>\n",
       "      <td>202.000000</td>\n",
       "      <td>202.000000</td>\n",
       "      <td>202.000000</td>\n",
       "      <td>202.000000</td>\n",
       "      <td>202.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>-0.036816</td>\n",
       "      <td>-0.032889</td>\n",
       "      <td>-0.041644</td>\n",
       "      <td>-0.028521</td>\n",
       "      <td>-0.026560</td>\n",
       "      <td>-0.010150</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.995842</td>\n",
       "      <td>0.998423</td>\n",
       "      <td>1.000286</td>\n",
       "      <td>1.002046</td>\n",
       "      <td>1.003657</td>\n",
       "      <td>0.999537</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-1.813440</td>\n",
       "      <td>-2.259862</td>\n",
       "      <td>-2.914053</td>\n",
       "      <td>-2.370821</td>\n",
       "      <td>-2.331930</td>\n",
       "      <td>-2.132314</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-0.752065</td>\n",
       "      <td>-0.696937</td>\n",
       "      <td>-0.623982</td>\n",
       "      <td>-0.762459</td>\n",
       "      <td>-0.738538</td>\n",
       "      <td>-0.811747</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>-0.271538</td>\n",
       "      <td>-0.170732</td>\n",
       "      <td>0.045975</td>\n",
       "      <td>-0.162339</td>\n",
       "      <td>-0.171410</td>\n",
       "      <td>0.090408</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.671235</td>\n",
       "      <td>0.655952</td>\n",
       "      <td>0.815968</td>\n",
       "      <td>0.629157</td>\n",
       "      <td>0.653908</td>\n",
       "      <td>0.922974</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2.295205</td>\n",
       "      <td>2.208584</td>\n",
       "      <td>1.961483</td>\n",
       "      <td>2.043331</td>\n",
       "      <td>2.180700</td>\n",
       "      <td>1.581641</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Series Code      CC.EST      GE.EST      PV.EST      RL.EST      RQ.EST  \\\n",
       "count        202.000000  202.000000  202.000000  202.000000  202.000000   \n",
       "mean          -0.036816   -0.032889   -0.041644   -0.028521   -0.026560   \n",
       "std            0.995842    0.998423    1.000286    1.002046    1.003657   \n",
       "min           -1.813440   -2.259862   -2.914053   -2.370821   -2.331930   \n",
       "25%           -0.752065   -0.696937   -0.623982   -0.762459   -0.738538   \n",
       "50%           -0.271538   -0.170732    0.045975   -0.162339   -0.171410   \n",
       "75%            0.671235    0.655952    0.815968    0.629157    0.653908   \n",
       "max            2.295205    2.208584    1.961483    2.043331    2.180700   \n",
       "\n",
       "Series Code      VA.EST  \n",
       "count        202.000000  \n",
       "mean          -0.010150  \n",
       "std            0.999537  \n",
       "min           -2.132314  \n",
       "25%           -0.811747  \n",
       "50%            0.090408  \n",
       "75%            0.922974  \n",
       "max            1.581641  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wgi.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Classification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DBSCAN(algorithm='auto', eps=0.8, leaf_size=30, metric='euclidean',\n",
       "    metric_params=None, min_samples=3, n_jobs=1, p=None)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.cluster import DBSCAN\n",
    "\n",
    "db = DBSCAN(eps = .8, min_samples = 3)\n",
    "db.fit(wgi)"
   ]
  },
  {
   "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>Series Code</th>\n",
       "      <th>CC.EST</th>\n",
       "      <th>GE.EST</th>\n",
       "      <th>PV.EST</th>\n",
       "      <th>RL.EST</th>\n",
       "      <th>RQ.EST</th>\n",
       "      <th>VA.EST</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Labels</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",
       "      <th>-1</th>\n",
       "      <td>-0.036795</td>\n",
       "      <td>-0.055243</td>\n",
       "      <td>0.112498</td>\n",
       "      <td>-0.053595</td>\n",
       "      <td>-0.083211</td>\n",
       "      <td>-0.576135</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-1.494299</td>\n",
       "      <td>-1.685430</td>\n",
       "      <td>-2.371651</td>\n",
       "      <td>-1.723911</td>\n",
       "      <td>-1.551087</td>\n",
       "      <td>-1.491839</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.541357</td>\n",
       "      <td>-0.464884</td>\n",
       "      <td>-0.383015</td>\n",
       "      <td>-0.482340</td>\n",
       "      <td>-0.398705</td>\n",
       "      <td>-0.242662</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.103227</td>\n",
       "      <td>1.046090</td>\n",
       "      <td>0.879165</td>\n",
       "      <td>1.098989</td>\n",
       "      <td>0.987419</td>\n",
       "      <td>1.040816</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-1.348473</td>\n",
       "      <td>-1.490840</td>\n",
       "      <td>-0.634753</td>\n",
       "      <td>-1.572616</td>\n",
       "      <td>-2.203282</td>\n",
       "      <td>-2.119010</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Series Code    CC.EST    GE.EST    PV.EST    RL.EST    RQ.EST    VA.EST\n",
       "Labels                                                                 \n",
       "-1          -0.036795 -0.055243  0.112498 -0.053595 -0.083211 -0.576135\n",
       " 0          -1.494299 -1.685430 -2.371651 -1.723911 -1.551087 -1.491839\n",
       " 1          -0.541357 -0.464884 -0.383015 -0.482340 -0.398705 -0.242662\n",
       " 2           1.103227  1.046090  0.879165  1.098989  0.987419  1.040816\n",
       " 3          -1.348473 -1.490840 -0.634753 -1.572616 -2.203282 -2.119010"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wgi_db = wgi.assign(Labels = db.labels_)\n",
    "wgi_db.groupby(\"Labels\").mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f979829ea90>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9798286dd8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "seaborn.heatmap(wgi_db.groupby(\"Labels\").mean(), cmap = \"BrBG\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ACP"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PCA(copy=True, iterated_power='auto', n_components=None, random_state=None,\n",
       "  svd_solver='auto', tol=0.0, whiten=False)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pca = PCA()\n",
    "pca.fit(wgi)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "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>Dimension</th>\n",
       "      <th>Valeur propre</th>\n",
       "      <th>% variance expliquée</th>\n",
       "      <th>% cum. var. expliquée</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Dim1</td>\n",
       "      <td>5.004603</td>\n",
       "      <td>83.0</td>\n",
       "      <td>83.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Dim2</td>\n",
       "      <td>0.470234</td>\n",
       "      <td>8.0</td>\n",
       "      <td>91.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Dim3</td>\n",
       "      <td>0.317695</td>\n",
       "      <td>5.0</td>\n",
       "      <td>97.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Dim4</td>\n",
       "      <td>0.114488</td>\n",
       "      <td>2.0</td>\n",
       "      <td>98.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Dim5</td>\n",
       "      <td>0.051994</td>\n",
       "      <td>1.0</td>\n",
       "      <td>99.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Dim6</td>\n",
       "      <td>0.040607</td>\n",
       "      <td>1.0</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Dimension  Valeur propre  % variance expliquée  % cum. var. expliquée\n",
       "0      Dim1       5.004603                  83.0                   83.0\n",
       "1      Dim2       0.470234                   8.0                   91.0\n",
       "2      Dim3       0.317695                   5.0                   97.0\n",
       "3      Dim4       0.114488                   2.0                   98.0\n",
       "4      Dim5       0.051994                   1.0                   99.0\n",
       "5      Dim6       0.040607                   1.0                  100.0"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "eig = pandas.DataFrame(\n",
    "    {\n",
    "        \"Dimension\" : [\"Dim\" + str(x + 1) for x in range(6)], \n",
    "        \"Valeur propre\" : pca.explained_variance_,\n",
    "        \"% variance expliquée\" : numpy.round(pca.explained_variance_ratio_ * 100),\n",
    "        \"% cum. var. expliquée\" : numpy.round(numpy.cumsum(pca.explained_variance_ratio_) * 100)\n",
    "    },\n",
    "    columns = [\"Dimension\", \"Valeur propre\", \"% variance expliquée\", \"% cum. var. expliquée\"]\n",
    ")\n",
    "eig"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(3.25,16,'16%')"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f97936aaba8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "g_eig = seaborn.barplot(x = \"Dimension\", \n",
    "                y = \"% variance expliquée\",\n",
    "                palette = [\"lightseagreen\"],\n",
    "                data = eig)\n",
    "g_eig.set(ylabel = \"Variance expliquée (%)\")\n",
    "g_eig.figure.suptitle(\"Variance expliquée par dimension\")\n",
    "plt.axhline(y = 16, linewidth = .5, color = \"dimgray\", linestyle = \"--\") # 16 = 100 / 6 (nb dimensions)\n",
    "plt.text(3.25, 16, \"16%\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "wgi_pca = pca.transform(wgi)\n",
    "# Transformation en DataFrame pandas\n",
    "wgi_pca_df = pandas.DataFrame({\n",
    "    \"Dim1\" : wgi_pca[:,0], \n",
    "    \"Dim2\" : wgi_pca[:,1], \n",
    "    \"Pays\" : wgi.index,\n",
    "    \"Labels\" : db.labels_\n",
    "})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "scrolled": true
   },
   "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>Dim1</th>\n",
       "      <th>Dim2</th>\n",
       "      <th>Labels</th>\n",
       "      <th>Pays</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3.787545</td>\n",
       "      <td>-0.937147</td>\n",
       "      <td>0</td>\n",
       "      <td>Afghanistan</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.016315</td>\n",
       "      <td>0.286759</td>\n",
       "      <td>1</td>\n",
       "      <td>Albania</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2.059633</td>\n",
       "      <td>-0.341831</td>\n",
       "      <td>1</td>\n",
       "      <td>Algeria</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-3.389095</td>\n",
       "      <td>0.124413</td>\n",
       "      <td>2</td>\n",
       "      <td>Andorra</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2.436206</td>\n",
       "      <td>0.227125</td>\n",
       "      <td>1</td>\n",
       "      <td>Angola</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Dim1      Dim2  Labels         Pays\n",
       "0  3.787545 -0.937147       0  Afghanistan\n",
       "1  0.016315  0.286759       1      Albania\n",
       "2  2.059633 -0.341831       1      Algeria\n",
       "3 -3.389095  0.124413       2      Andorra\n",
       "4  2.436206  0.227125       1       Angola"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Résultat (premières lignes)\n",
    "wgi_pca_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f97936bef28>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(16,8))\n",
    "g_pca = seaborn.scatterplot(\"Dim1\", \"Dim2\", \n",
    "                            hue = \"Labels\", size = \"Labels\", sizes = (100, 100),\n",
    "                            data = wgi_pca_df, palette = \"Accent\")\n",
    "for x, y, v in zip(wgi_pca_df.Dim1, wgi_pca_df.Dim2, wgi_pca_df.Pays):\n",
    "    if (abs(x) > 4) | (abs(y) > 1.5):\n",
    "        g_pca.text(x, y, v)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f979369c5c0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "wgi_var = pandas.DataFrame({\n",
    "    \"Dim1\" : pca.components_[0, :], \n",
    "    \"Dim2\" : pca.components_[1, :],\n",
    "    \"Var\" : wgi.columns\n",
    "})\n",
    "g_pca = seaborn.scatterplot(\"Dim1\", \"Dim2\", data = wgi_var)\n",
    "g_pca.axvline(x = 0, linewidth = .5, color = \"dimgray\", linestyle = \"--\") \n",
    "g_pca.axhline(y = 0, linewidth = .5, color = \"dimgray\", linestyle = \"--\") \n",
    "\n",
    "for x, y, v in zip(wgi_var.Dim1, wgi_var.Dim2, wgi_var.Var):\n",
    "    g_pca.text(x, y, v)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Variation de `eps`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     eps  nb_clusters  nb_outliers\n",
      "0   0.05            0          202\n",
      "1   0.10            0          202\n",
      "2   0.15            0          202\n",
      "3   0.20            0          202\n",
      "4   0.25            0          202\n",
      "5   0.30            1          198\n",
      "6   0.35            5          182\n",
      "7   0.40            8          167\n",
      "8   0.45           10          148\n",
      "9   0.50           12          128\n",
      "10  0.55           10          117\n",
      "11  0.60            6           93\n",
      "12  0.65            6           66\n",
      "13  0.70            4           50\n",
      "14  0.75            4           40\n",
      "15  0.80            4           32\n",
      "16  0.85            3           23\n",
      "17  0.90            3           16\n",
      "18  0.95            2           10\n",
      "19  1.00            2            9\n"
     ]
    }
   ],
   "source": [
    "res = []\n",
    "\n",
    "for e in [v / 20 for v in range(1, 21)]:\n",
    "    dbe = DBSCAN(eps = e, min_samples = 3)\n",
    "    dbe.fit(wgi)\n",
    "    t = pandas.crosstab(dbe.labels_, \"Freq\")\n",
    "    if len(t[\"Freq\"]) == 1:\n",
    "        nboutliers = t[\"Freq\"]\n",
    "    else:\n",
    "        nboutliers = t[\"Freq\"][0]\n",
    "    res.append({\"eps\": e, \"nb_clusters\": len(t) - 1, \"nb_outliers\": list(t[\"Freq\"])[0]})\n",
    "res = pandas.DataFrame(res)\n",
    "print(res)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f9791f09390>"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9791f07a58>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = seaborn.lineplot(\"eps\", \"nb_clusters\", data = res)\n",
    "ax2 = ax.twinx()\n",
    "# en rouge les outliers\n",
    "seaborn.lineplot(\"eps\", \"nb_outliers\", data = res, color = \"red\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}