1388 lines
226 KiB
Plaintext
1388 lines
226 KiB
Plaintext
{
|
||
"cells": [
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "395d61ec",
|
||
"metadata": {},
|
||
"source": [
|
||
"# TP1 du module 4 : le travail sur les données.\n",
|
||
"\n",
|
||
"Dans ce TP, nous allons explorer plus en détails le jeu de données du Titanic. Objectifs :\n",
|
||
"* Analyser des statistiques pour décrire les données\n",
|
||
"* Produire des visualisations pertinentes pour la compréhesion des données.\n",
|
||
"* Nettoyer le jeu de données\n",
|
||
"* Préparer les données pour qu'elles soient prêtes à être fournies à un algorithme d'apprentissage."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"id": "5117092f",
|
||
"metadata": {
|
||
"ExecuteTime": {
|
||
"end_time": "2025-09-16T10:06:44.308342Z",
|
||
"start_time": "2025-09-16T10:06:44.305868Z"
|
||
}
|
||
},
|
||
"source": [
|
||
"# Ajoutez ici les imports de librairies nécessaires\n",
|
||
"import numpy as np\n",
|
||
"import pandas as pd\n",
|
||
"import matplotlib.pyplot as plt\n",
|
||
"import seaborn as sns\n",
|
||
"\n",
|
||
"from sklearn.preprocessing import OneHotEncoder\n",
|
||
"import skitlearn"
|
||
],
|
||
"outputs": [],
|
||
"execution_count": 4
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "fde8da96",
|
||
"metadata": {},
|
||
"source": [
|
||
"Commencez par recharger le jeu de données depuis un csv dans un dataframe Pandas. Rappel de l'adresse à laquelle vous pouvez le trouver : https://raw.githubusercontent.com/datasciencedojo/datasets/master/titanic.csv"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"id": "33fba6ca",
|
||
"metadata": {
|
||
"ExecuteTime": {
|
||
"end_time": "2025-09-16T10:06:44.431049Z",
|
||
"start_time": "2025-09-16T10:06:44.315956Z"
|
||
}
|
||
},
|
||
"source": [
|
||
"titanic = pd.read_csv('https://raw.githubusercontent.com/datasciencedojo/datasets/master/titanic.csv')"
|
||
],
|
||
"outputs": [],
|
||
"execution_count": 5
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "205f765d",
|
||
"metadata": {},
|
||
"source": [
|
||
"## Exploration du jeu de données\n",
|
||
"\n",
|
||
"Commencez par répondre au question suivante. Prenez le temps de bien analyser vos réponses, afin de mieux vous familiariser avec le contenu du jeu de données.\n",
|
||
"\n",
|
||
"1. Combien de données dans le jeu de données Titanic ?"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"id": "4ee3884e",
|
||
"metadata": {
|
||
"ExecuteTime": {
|
||
"end_time": "2025-09-16T10:06:44.443097Z",
|
||
"start_time": "2025-09-16T10:06:44.440519Z"
|
||
}
|
||
},
|
||
"source": [
|
||
"print(\"Nombre de données : \", len(titanic))"
|
||
],
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Nombre de données : 891\n"
|
||
]
|
||
}
|
||
],
|
||
"execution_count": 6
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "74f32328",
|
||
"metadata": {},
|
||
"source": [
|
||
"2. Combien d'attributs compte le jeu de données ?"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"id": "80eeccc3",
|
||
"metadata": {
|
||
"ExecuteTime": {
|
||
"end_time": "2025-09-16T10:06:44.467871Z",
|
||
"start_time": "2025-09-16T10:06:44.465137Z"
|
||
}
|
||
},
|
||
"source": [
|
||
"print(\"Nombre d'attributs : \", len(titanic.columns))"
|
||
],
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Nombre d'attributs : 12\n"
|
||
]
|
||
}
|
||
],
|
||
"execution_count": 7
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "819573a7",
|
||
"metadata": {},
|
||
"source": [
|
||
"3. Identifiez quelles colonnes contiennent des données discrètes, et lesquelles contiennent des données continues."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"id": "87aa38a1",
|
||
"metadata": {
|
||
"ExecuteTime": {
|
||
"end_time": "2025-09-16T10:06:44.497752Z",
|
||
"start_time": "2025-09-16T10:06:44.488952Z"
|
||
}
|
||
},
|
||
"source": [
|
||
"print(titanic.info())"
|
||
],
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"<class 'pandas.core.frame.DataFrame'>\n",
|
||
"RangeIndex: 891 entries, 0 to 890\n",
|
||
"Data columns (total 12 columns):\n",
|
||
" # Column Non-Null Count Dtype \n",
|
||
"--- ------ -------------- ----- \n",
|
||
" 0 PassengerId 891 non-null int64 \n",
|
||
" 1 Survived 891 non-null int64 \n",
|
||
" 2 Pclass 891 non-null int64 \n",
|
||
" 3 Name 891 non-null object \n",
|
||
" 4 Sex 891 non-null object \n",
|
||
" 5 Age 714 non-null float64\n",
|
||
" 6 SibSp 891 non-null int64 \n",
|
||
" 7 Parch 891 non-null int64 \n",
|
||
" 8 Ticket 891 non-null object \n",
|
||
" 9 Fare 891 non-null float64\n",
|
||
" 10 Cabin 204 non-null object \n",
|
||
" 11 Embarked 889 non-null object \n",
|
||
"dtypes: float64(2), int64(5), object(5)\n",
|
||
"memory usage: 83.7+ KB\n",
|
||
"None\n"
|
||
]
|
||
}
|
||
],
|
||
"execution_count": 8
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "3daaaf5c",
|
||
"metadata": {},
|
||
"source": [
|
||
"**Réponse :** \n",
|
||
"\n",
|
||
"* Données discrètes : PassengerId, Survived, Pclass, Name, Sex, SibSp, Parch, Cabin, Embarked\n",
|
||
"* Données continues : Age, Fare, Ticket"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "59733c69",
|
||
"metadata": {},
|
||
"source": [
|
||
"4. De la même manière, identifiez les colonnes de données qualitatives, et celles de données quantitatives.\n",
|
||
"\n",
|
||
"**Réponse :** \n",
|
||
"\n",
|
||
"* Données qualitatives : Survived, Name, Sex, Ticket, Cabin, Embarked\n",
|
||
"* Données quantitatives : PassengerId, Pclass, Age, SibSp, Parch, Fare"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "69c4bee2",
|
||
"metadata": {},
|
||
"source": [
|
||
"5. Affichez les statistiques de base sur les colonnes quantitatives du dataset. \n",
|
||
"Quelles informations pouvez-vous en retirer ? Pour chaque attribut, cherchez au moins une information pertinente que vous pouvez déduire de vos observations."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"id": "82ebfbb6",
|
||
"metadata": {
|
||
"ExecuteTime": {
|
||
"end_time": "2025-09-16T10:06:44.540637Z",
|
||
"start_time": "2025-09-16T10:06:44.518122Z"
|
||
}
|
||
},
|
||
"source": [
|
||
"titanic.describe()"
|
||
],
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"text/plain": [
|
||
" PassengerId Survived Pclass Age SibSp \\\n",
|
||
"count 891.000000 891.000000 891.000000 714.000000 891.000000 \n",
|
||
"mean 446.000000 0.383838 2.308642 29.699118 0.523008 \n",
|
||
"std 257.353842 0.486592 0.836071 14.526497 1.102743 \n",
|
||
"min 1.000000 0.000000 1.000000 0.420000 0.000000 \n",
|
||
"25% 223.500000 0.000000 2.000000 20.125000 0.000000 \n",
|
||
"50% 446.000000 0.000000 3.000000 28.000000 0.000000 \n",
|
||
"75% 668.500000 1.000000 3.000000 38.000000 1.000000 \n",
|
||
"max 891.000000 1.000000 3.000000 80.000000 8.000000 \n",
|
||
"\n",
|
||
" Parch Fare \n",
|
||
"count 891.000000 891.000000 \n",
|
||
"mean 0.381594 32.204208 \n",
|
||
"std 0.806057 49.693429 \n",
|
||
"min 0.000000 0.000000 \n",
|
||
"25% 0.000000 7.910400 \n",
|
||
"50% 0.000000 14.454200 \n",
|
||
"75% 0.000000 31.000000 \n",
|
||
"max 6.000000 512.329200 "
|
||
],
|
||
"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>PassengerId</th>\n",
|
||
" <th>Survived</th>\n",
|
||
" <th>Pclass</th>\n",
|
||
" <th>Age</th>\n",
|
||
" <th>SibSp</th>\n",
|
||
" <th>Parch</th>\n",
|
||
" <th>Fare</th>\n",
|
||
" </tr>\n",
|
||
" </thead>\n",
|
||
" <tbody>\n",
|
||
" <tr>\n",
|
||
" <th>count</th>\n",
|
||
" <td>891.000000</td>\n",
|
||
" <td>891.000000</td>\n",
|
||
" <td>891.000000</td>\n",
|
||
" <td>714.000000</td>\n",
|
||
" <td>891.000000</td>\n",
|
||
" <td>891.000000</td>\n",
|
||
" <td>891.000000</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>mean</th>\n",
|
||
" <td>446.000000</td>\n",
|
||
" <td>0.383838</td>\n",
|
||
" <td>2.308642</td>\n",
|
||
" <td>29.699118</td>\n",
|
||
" <td>0.523008</td>\n",
|
||
" <td>0.381594</td>\n",
|
||
" <td>32.204208</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>std</th>\n",
|
||
" <td>257.353842</td>\n",
|
||
" <td>0.486592</td>\n",
|
||
" <td>0.836071</td>\n",
|
||
" <td>14.526497</td>\n",
|
||
" <td>1.102743</td>\n",
|
||
" <td>0.806057</td>\n",
|
||
" <td>49.693429</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>min</th>\n",
|
||
" <td>1.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>1.000000</td>\n",
|
||
" <td>0.420000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>25%</th>\n",
|
||
" <td>223.500000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>2.000000</td>\n",
|
||
" <td>20.125000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>7.910400</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>50%</th>\n",
|
||
" <td>446.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>3.000000</td>\n",
|
||
" <td>28.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>14.454200</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>75%</th>\n",
|
||
" <td>668.500000</td>\n",
|
||
" <td>1.000000</td>\n",
|
||
" <td>3.000000</td>\n",
|
||
" <td>38.000000</td>\n",
|
||
" <td>1.000000</td>\n",
|
||
" <td>0.000000</td>\n",
|
||
" <td>31.000000</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>max</th>\n",
|
||
" <td>891.000000</td>\n",
|
||
" <td>1.000000</td>\n",
|
||
" <td>3.000000</td>\n",
|
||
" <td>80.000000</td>\n",
|
||
" <td>8.000000</td>\n",
|
||
" <td>6.000000</td>\n",
|
||
" <td>512.329200</td>\n",
|
||
" </tr>\n",
|
||
" </tbody>\n",
|
||
"</table>\n",
|
||
"</div>"
|
||
]
|
||
},
|
||
"execution_count": 9,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"execution_count": 9
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "a5ad3de1",
|
||
"metadata": {},
|
||
"source": [
|
||
"**Observations sur les statistiques :** \n",
|
||
"* Il y a malheureusement eu plus de personnes ayant péri que de survivants. La proportion est d'environ 1/3 de survivants.\n",
|
||
"* Il semble y avoir eu une majorité (plus de la moitié) de voyageurs en 3e classe.\n",
|
||
"* Les âges sont très variés, avec un pic autour de la trentaine.\n",
|
||
"* Une majorité de passagers voyageaient sans frère, soeur ou conjoint à bord\n",
|
||
"* De même, peu de passagers voyageaient avec des parents ou des enfants.\n",
|
||
"* Les prix des tickets semblent très variables, avec une majorité de prix assez faibles et éloignés du prix maximum. Cela est cohérent avec la majorité de voyageurs en 3e classe."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "46384afa",
|
||
"metadata": {},
|
||
"source": [
|
||
"6. Sur une figure avec 6 sous-figures, proposez un histogramme pour visualiser la répartition des valeurs sur les attributs suivants : Survived, Pclass, Sex, Embarked, Age, Fare. Pour chaque figure, quelle(s) observation(s) pouvez-vous faire ?"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"id": "94ddbac5",
|
||
"metadata": {
|
||
"ExecuteTime": {
|
||
"end_time": "2025-09-16T10:06:45.173695Z",
|
||
"start_time": "2025-09-16T10:06:44.560697Z"
|
||
}
|
||
},
|
||
"source": [
|
||
"f, axes = plt.subplots(2, 3, figsize=(20, 10))\n",
|
||
" \n",
|
||
"# Histogramme pour la survie\n",
|
||
"sub1 = axes[0, 0]\n",
|
||
"sns.histplot(titanic['Survived'] , color=\"skyblue\", ax=sub1, bins=2)\n",
|
||
"sub1.set_xticks([0,1])\n",
|
||
"\n",
|
||
"# Histogramme pour la classe des passagers\n",
|
||
"sub1 = axes[0, 1]\n",
|
||
"sns.histplot(titanic['Pclass'] , color=\"skyblue\", ax=sub1, bins=3)\n",
|
||
"sub1.set_xticks([1,2,3])\n",
|
||
"\n",
|
||
"# Histogramme pour le genre des passagers\n",
|
||
"sub1 = axes[0, 2]\n",
|
||
"sns.histplot(titanic['Sex'] , color=\"skyblue\", ax=sub1, bins=2)\n",
|
||
"\n",
|
||
"# Histogramme pour le port d'embarquement\n",
|
||
"sub1 = axes[1, 0]\n",
|
||
"sns.histplot(titanic['Embarked'] , color=\"skyblue\", ax=sub1)\n",
|
||
"\n",
|
||
"# Histogramme pour l'âge des passagers, regroupés par dizaine\n",
|
||
"sub1 = axes[1, 1]\n",
|
||
"sns.histplot(titanic['Age'] , color=\"skyblue\", ax=sub1, binwidth=10)\n",
|
||
"\n",
|
||
"# Histogramme pour le prix du billet\n",
|
||
"sub1 = axes[1, 2]\n",
|
||
"sns.histplot(titanic['Fare'] , color=\"skyblue\", ax=sub1)"
|
||
],
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"text/plain": [
|
||
"<Axes: xlabel='Fare', ylabel='Count'>"
|
||
]
|
||
},
|
||
"execution_count": 10,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
},
|
||
{
|
||
"data": {
|
||
"text/plain": [
|
||
"<Figure size 2000x1000 with 6 Axes>"
|
||
],
|
||
"image/png": "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"
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data",
|
||
"jetTransient": {
|
||
"display_id": null
|
||
}
|
||
}
|
||
],
|
||
"execution_count": 10
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "fd730025",
|
||
"metadata": {},
|
||
"source": [
|
||
"**Observations :**\n",
|
||
"* Plus de décès que de survivants\n",
|
||
"* Majorité de personnes en 3e classe\n",
|
||
"* Plus d'hommes que de femmes\n",
|
||
"* Grande majorité de personnes ayant embarqué à Southampton\n",
|
||
"* Ages très répartis, pic autour de la trentaine (corrobore les statistiques)\n",
|
||
"* Beaucoup de billets à bas prix, valeurs très étalées."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "620fbbba",
|
||
"metadata": {},
|
||
"source": [
|
||
"7. Sur un même graphique, représentez, pour chaque genre, le nombre de personnes ayant survécu. Que pouvez-vous en déduire ? Le genre d'un passager vous parait-il pertinent pour qu'un modèle d'apprentissage puisse prédire si ce passager a survécu ?"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"id": "f58cb499",
|
||
"metadata": {
|
||
"ExecuteTime": {
|
||
"end_time": "2025-09-16T10:06:45.449518Z",
|
||
"start_time": "2025-09-16T10:06:45.221430Z"
|
||
}
|
||
},
|
||
"source": [
|
||
"sns.catplot(x =\"Sex\", hue =\"Survived\",\n",
|
||
"kind =\"count\", data = titanic)"
|
||
],
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"text/plain": [
|
||
"<seaborn.axisgrid.FacetGrid at 0x203836716a0>"
|
||
]
|
||
},
|
||
"execution_count": 11,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
},
|
||
{
|
||
"data": {
|
||
"text/plain": [
|
||
"<Figure size 565.361x500 with 1 Axes>"
|
||
],
|
||
"image/png": "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"
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data",
|
||
"jetTransient": {
|
||
"display_id": null
|
||
}
|
||
}
|
||
],
|
||
"execution_count": 11
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "4f61683e",
|
||
"metadata": {},
|
||
"source": [
|
||
"**Observation :** Il y avait plus d'hommes que de femmes à bord du bateau, mais les femmes ont été plus nombreuses à survivre. Ainsi, si la majorité des hommes ont péri, chez les femmes la tendance est inversée. Le genre est donc un attribut très pertinent pour prédire la survie ou non, car très discriminant. Il ne sera pas suffisant, mais pourra améliorer la prédiction s'il est combiné à d'autres attributs."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "412e2e95",
|
||
"metadata": {},
|
||
"source": [
|
||
"8. En vous basant sur une visualisation, observez-vous une corrélation entre certains attributs ? Que pouvez-vous en déduire pour un futur modèle d'apprentissage ?"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"id": "48d8ee62",
|
||
"metadata": {
|
||
"ExecuteTime": {
|
||
"end_time": "2025-09-16T10:06:45.719113Z",
|
||
"start_time": "2025-09-16T10:06:45.488383Z"
|
||
}
|
||
},
|
||
"source": [
|
||
"# Filtrer les colonnes numériques, car titanic.corr() échoue sinon.\n",
|
||
"# Ajouter des annotations (optionnel mais utile) pour rendre les corrélations lisibles.\n",
|
||
"# Commenter ou analyser brièvement la heatmap après coup, pour répondre à \"Que pouvez-vous en déduire\".\n",
|
||
"\n",
|
||
"# 1. Sélection des colonnes numériques uniquement\n",
|
||
"numeric_df = titanic.select_dtypes(include='number')\n",
|
||
"\n",
|
||
"# 2. Calcul des corrélations\n",
|
||
"correlation_matrix = numeric_df.corr()\n",
|
||
"\n",
|
||
"# 3. Affichage de la heatmap\n",
|
||
"sns.heatmap(correlation_matrix, annot=True, cmap='coolwarm')\n",
|
||
"plt.title(\"Matrice de corrélation des attributs numériques du Titanic\")\n",
|
||
"plt.show()"
|
||
],
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"text/plain": [
|
||
"<Figure size 640x480 with 2 Axes>"
|
||
],
|
||
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAkkAAAHxCAYAAAB58SixAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjYsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvq6yFwwAAAAlwSFlzAAAPYQAAD2EBqD+naQAA4FBJREFUeJzsnQV0FFcXx/+xjQuBBIgRkmDB3d3d3aFIKVIBihQrbqVQaAkUt+LuUNw1aCACERIgCXG379y37GYtWDchyXd/58zZ3Zk3s/Nmnvzn3vve6GRkZGSAYRiGYRiGUUJX+SfDMAzDMAxDsEhiGIZhGIbRAIskhmEYhmEYDbBIYhiGYRiG0QCLJIZhGIZhGA2wSGIYhmEYhtEAiySGYRiGYRgNsEhiGIZhGIbRAIukPAjP/5l34Xv3edcnN1yv3HAOTN6Hy1HehEVSNtC/f3+UKlUKvXr1yjLNDz/8INJMmjTps459584dDB8+/KPp/vjjD3H8r0FQUJD473379iEv4+3tjTJlymD69On/+dq+fv1a3LdXr17J1zVp0uSz7782+ZplRJXk5GTMmzcPhw8fVrr+vXv3/ui+VM4oH1TuCLqmdG21wZ9//ol169YhP/H999+jWrVq8PX1/WjaGzduiGtLn3kZbeRDVl8+tFC5U/0vTXVfW6iWfUb76GfDMRlSn7q6uH//vqggRYoUUdoWHx+Pc+fOfdFxd+/e/UmNW/fu3VG/fv0v+g8ms1GsXr06pk2b9p+PdfXqVVy4cEFp3cqVK2FmZvafj50fePv2LTZt2oT58+fL1504cQL37t376L6NGjXCzp07YWtrq/XzWr58OUaPHo38wvPnz3HmzBn89ddfcHV1/Wj6smXLimvr5uaG/3dU21Rqi/fs2SOujwyJRAInJyela6ap7muL7Cz7jBQWSdmEu7s7fHx8REM/aNAgpW0kkIyNjWFhYZFt/0/CTFWcMZ8HWfsKFy4MAwODbCsjzH/H2tpaLMzHoc705MmTsLe3/6T0JOIrVaqU7eeVF1BtUy9duiQ+NV2fnLpmXPazH3a3ZRMmJiZo2LChEEmqHDt2DC1btoS+vrJGfffuHWbNmoXGjRujXLlyqFGjBr777jslN8L+/fuF2VbmzpK5tjZs2IBWrVqhYsWK2Lt3r0ZXyoEDB9C5c2eRhp5Ali5dKtwcik+ZI0aMQJUqVcRC/x0YGPjRvJ46dQodOnRAhQoVxPG9vLzU0kRGRgq3VZ06dVC+fHn06NED165d++ix/fz8xJM8XQuy6tD5KVrSYmJihPWhWbNm4rjt2rUTT3eKkAmcXDkDBw4U5zh16lS5Sfyff/4R15vye+XKFZH+9u3b6NevHzp16iSu088//yzuTVakpaVhzZo14r/p+NRAkqv1+vXrYjvdp8mTJ4vvTZs2lbvYVN1tn5qXFStWYOHCheJa0v8NHToUL1++/OB1TEpKEseuW7cuKleuLM6H1qkiyzuVEbrmqnlPT0/HsmXLxHlQGaVPKkcpKSkf/H+yXvTp00f8N+1HZXXbtm1iG5Vhui4EnRcdk8ovWdoIuk/0W/ad1nfp0kXknb5n5XKgJ2y6f5SO7v2TJ08+6mpU/S+C/kP2PTExETNnzkSDBg3k+fiYO47uMT0oUb2kek/7dezYERcvXvys85HVdWpTRo0aJcoZlQFyCcbGxmLKlCmoWrWqWLd48WKlGBi614sWLRL/S+fQvn170Q59aj1RdFNRuab6S2WkdevWOHv2LJo3by4/z6xcWxSGQIsiZI1p27atuCZ0r+gYVJ9kUNn76aefRLmlOkHnT+3Yx6B6TfmkfFB5Dg4OVtr+Kdf7S1HMf1Z1n8oR1ZsWLVqIvFP7M3jwYDx9+vSzyo2msk9WK2p/qHzUq1dPtLvR0dH/KU//z7BIykbatGkjd7nJoMaMCjl1gIpQg0YCgDrq8ePHi4aXxAEJiRkzZog01DCS8LKxsZF3ADKoYg8bNkw0hNSgqEIdEnV4ZD6nRp985Fu2bMGcOXPE9hcvXoiKFR4eLjrguXPnCoFEMSG0Liv+/fdfjB07VlTUVatWiUZzwoQJSmmogaaGlxpTss7Q/9MT2TfffPNBofTmzRv07NlTCADqmKjhDwsLE8ci0UUNDXW8FMdCx6LOgjoJatxXr16tln9qZClNt27d5OvpXOi6UENCHfitW7dEw2RkZITff/9ddDw3b97EgAEDxP9pYsmSJeK4dK5///03Zs+eLc5v3LhxSEhIEPfp22+/lf8f3UdVPicvmzdvFuKRRA/dv0ePHok8fAi6J7t27RJljPIVFRWFjRs3KqX5lLyvXbsWO3bsEAJ6/fr1onxQWSX3TVacP39epKeyR/misuro6Ihff/0Vnp6ewrohE0R0neg7uTZk94nKOv2WQdeDOnkSi9R5aILqHB2H4m9+++03kV/qoFU7yw8hc6PQeci+k4ig+kvXm/JNHR/VOerIPgTdI0pPdYXqiZ6eHsaMGSPO63P55ZdfULJkSXHNa9euLVyCdI503yjP1PFSOZQ9oFHbQtefhAN1xLQflXWqi6qCI6t6IuPx48eifJKFia4/lVkqJ1QvPxcPDw/hyqY80D3t27evKF+K7m0qt/RQRA+PtI2sr3TtZQ8gmti6datoM6mtpHyQmNOGy/xLyKruT5w4UZQZaoepHpGQohg8EoSK4vZzyw15KaiOFyxYUNRh6kvoAYXuNfNlsLstmysIudUUXW6nT58WBZg6QNWYDEpLDQAFVRI1a9ZEQECAvIEmXzeZVsnvLTPnUnwTQeKka9euGs+Dnv6pgpGFQiaKCOrAjx49KqwAVIHp/6njlMXJUONF+1CDm1UnTMelpzUSMITMZ09PSTIOHjworEvUSVODRdCTOHVaJDCy6mDoXMjSRVYyEoZE6dKlRcdMnStZ1Mj6RY0/Nfqy/09NTRWNI4k+Kysrsd7Ozk40GDJkT7nUyJM1QAadd/HixUUDTg0SQedMT7t0ntSQq0L3jhohxadkQ0ND0Zg9e/ZM3Cu6dwQFgjs4OKgdg54IPzUv5KaldbLzozJCwiMiIgIFChRQOzY1vuRiIaEpC4SmY5PQIJfw5+SdRBM90crKGlmbqNyYm5sjK+g/yMJIgk8G5ZHKN90H+g+6LgRdJ5kbUubaUHVdUP2gzl7Gw4cP1f6TrBGysinLB5VlejD4mKCUIftfOg/Zd8o/PYTQNSEoD2Q1pjr9IchKSPdYVg5oH7JwUGefldDLCrp3JP6IEiVK4MiRI+L/ZQMMatWqJcT23bt3RbtAMTHkGiILID24yY5B9Z/qHz2wyazaWdUTGVQ2qA0iUUPtEGFpaan2YPQx6HrIHixI9BFk9aAyTr/p/lLe6HqTwKN7JytvlEb236qQwKDjUj5JvMmOSw+nVLdyGrpWqnWf2rS4uDiRT9n9oHzROS5YsEAITll797nlhtoB+h9qz3V0dMQ6ulYkpOm4hQoVysHc5w/YkpSN0JMdmbAVXW4kSqjhkhVgGRT7QhYCEk9kOiWLEjXo1NApusSyQtbJaIKsRGQNIpO4IuSmoQpIMTdU6aii0jlTx0wLiSXqkKiR1QRZF+jJktxVilD+FCFrEVV6siTIjk2dGO1HT0pZPRXRSD7qnGQNhqzDoqclekqkBpRiK2SiQga5/sh6RULqY9dHcT11GrQPHZsaW9m5ktWDglxl7jhVSFyQdYtcA+SuIkFx6NAhse1T7h3xOXmhJ32ZiJFdE9n5a4LOiVAc8UUDCxQb2U/NO4kC+k7iksQzCSBqtMkNkBVkeaDGnzoGut/k5qHO9nOuz6eWdRl03jKBRFAZorJE1rL/AuWfxD5ZbcliQdZW6sQVrbof6yw/5Z59CMUyIuv0FPNKbQsJF+pgZfWP1tG9ld1XWqg8hIaGChH9qdeWyhIJLEWRQh29aujAx6CAfGo/6BxUz4lQLG/U8ZMlhVxz1NGTyCX3lCbIwkpt3cfapK8JXTuyDtF1I2s5tb0k4GSDeRTrxOeUG7qe5FImQanYv9D/0EMSC6Qvgy1J2QxVTnKbkfmfrAvUYMmeAlWhjpVcAyEhIeJpiRosEi2fAj1hZAW5fogPPe1SGuq8VOMUiKwCA0ncUIeqar1QHWlBx6bGmESSJmgbNeqazkmT1UXx/xUFlAxZY6Doh8/q+iiup/RkdSOzPi2q0P3TBFkyyB1An2RVoVEt9ET+OXOjfE5e6D8UIcFD0LlndWxC9T4p/t+n5p0Ej6mpqRCCZIUgCyI98dNTMVkwNEHikdwfZPanxrtYsWJya+mXzB3zobIuQ1OHQOWf6tZ/gaxh1FFRXSW3Ki0kWshKR1bOrFC9Z7JOLKt79iE0jYj8WP2n65yVsCBLqEwcfezaUllSbQ9IIH3MkqbpnIispjOhcyLI+kVWq+PHj4uOnso6xVyRq1ZT8PmnlPXcAFn2yHVLoo7qE5Ud2bVXrBOfU25k7fHn3gvmw7BIymbIrUSVgKxJVAmo0yd3haYnNHpCIpcNWXjIskRQvANZVP4LslF0qsHH5J6hJw9q5MldQo2PohtDRlZPiSTkqNFSjUeQNYAy6NjOzs6iU9VEVkKI9tMUME1Ck/YhYeXv769RdBGaXE8fgu4TNULkGpW5UxRRbbAIMpGTcKCYLLISuri4iGtCwZPUqH8q2s6LIrJ96T7JxJvqffrUvFPeyO1GCz2xUz6pEyPXIj39a3KDkPuGOgNyn1JZozT0JEwWmexCk3WSrqWsg5d1NmTRlFnlyNL1MejcKcaEFopvoqd/cu9QLAnd/y/lS8/nU6B6RG0PWao1QaL1c8qSpvgjxeudVUdO+aFyptgmUZtAbUNWIpfOnVx5tFAZorhGut70UEKDJTSdH6EaR6naJmXn9f4Y5B6XuRDJokpWTzofigeTjZj7Ekg803FU20yyRJO1ilzOMpc98+mwuy2boUaVKgN1mPQ0pKkDkpmfqVGhzkYmkKgCy1xdsgZHZjX4HKjjpsZDdW4mihWiJzmKSSJXG7lO6ImS3Dm0kJijjo3iqDRB1gXq9Gh0m+LTDwVzK0LHpid4esKRHZsW6lTJZaPoOlKErA3kAlKs9NT4kSihzplGu1FckupcOvSUTy5ERRfEpzYyFA9DjbHieZKlhEz+miaio7TUAFNwM1mQZPdHNgLlU++btvOiiMzCozrSUrE8fGreKTZKFtdG95NGmZFgIksUCUZNkMinYGJynchElOr10VQGvqSsK7qYqTOSQeWPri2dgyy/hOKgCk0PI4rnQO4MclFSoC1BgpPyTnX6cwLCNfGp5/MlUP2j2EWqo4r3lmLgKG6L3FyfCsUp0r1TdPdQG6U4qEFTXkhEKY5KpQ6byjW5mxTPiR7IyJpOIQdUHxRHCFM7Rm5OepjL6nqT4CpatOgHy3p2X++PlWNyOZNwobaXXGkywSYTSF86MzcJUGq/VfNK94v+S2adYz4PtiTlAOQTphEHVFlkQYqqyDpBMiNTUCw1KvRkIRtOT40cVWx6AqMnORIJnxKbQchGRNCxqWMjvz91IjQ6hRp5smLQqAvqAOk8KbiXBBAFjJOLhNJlxY8//ijiccilSEGYdFzV0VjUkVL8BlmpRo4cKRoxaljJrUPxLFnNQ0RWDRp9Q6KIzovS0cgccndQ0DF1uNu3bxdPZRSzQNYlEmjkCqLz+ZJ5qCg/1KCQZYDigUioUqdIYk3TqDQKdKb7QnmmBp4WEsSyofuyzkR2LiQ4ybqoOpEfXSNt50XRUkD3hlwX1CFSuSGBTEHln5t3EnO0jp70SSBTJ0eB9dQRZ+WWpbJNgcTkbqV7R3F2ZAWgzkF2fWSB32QlpGtDnagszxSYTL/piftTofJL1h4KqKd8UOAqPUVTWSWo86XRgRTsTJZbElEkGGSWDhl0DnS+FMtEol02OpTKIlkPqbzTtByfG3ytyqeez5cem+4b3UNa6Po+ePBA1GuKL/qceXaofFK5pHOkekkPCCRqFKHrQnWczl9m3SCLiaIllh7aaH+6LySuSbxSWaLflJ7cT1QmqLyQKKc0JChIYFDbR+2BJmhfslxSGaa2lgZl0AhjGpGZU9dbFdW6T2WI2glyVQ8ZMkTEIFFsKI0CVRyM8yVQ20HlnuoyTWFCfQXdH3pQpxGRzOfDlqQcgJ58qKLQU3lWs9xSI0EVlp526WmJAl3pSVU2NFr2lEOdKfniqbH6lPlCZJAYomOSRYAaGLIQ0f/QUFSCGiUSZdTI0DqqbOSeoIaDrABZQR0HiR1q4KgzJ2FFvnZFyNRPx6agdGoY6H/J+kQNmWwOEU1QQ0vCgWKcaM4QSkvraGZmEnbU6FJwOwVpUuNKjQNdJ5q+gEThl0AjYSiokp4w6RrQtSCRSUJA0wRx1JCT+Z+e/mjIP6Wnp1wShdTgyoKm6f5SOaAgb5piQZXsyIsiFBMkCzam+0RP/iRYPzfvlEfaj8SbLCCb9vuQkKY0JHIofofKLblMyF1C+8muD3WmJKJJlNN5knWTyh1ZF+jef+6rQcgqRtMGUKwQ5YM6WCpLMkFA4pbuA1ksSBiSK4rOTzWejvJKHTOdE3Wk9KBBdZCEInVwsqHy9D//hU89ny+BHs5IlJLFi8QKiQLZdAAknD8HstRQXSaRSLGV1D6ojhakMkPlgYQ0ddZUhum/VdsR2p/uLYkHur7UNlAbQWVUJpqp/SMhR3WCrjeJHSq/VI6ygkbrUb5IHFE9IssK3becut6qqNZ9emih79Rm0vnJRiVS/af2V1YnvgRqP+iBTebSo+tGD5Sy0cfM56OTwW/dYxiGYf4DZD0i8aINQc8wuQm2JDEMwzAMw2iARRLDMAzDMIwGWCQxDMMw/wkaBMCuNkabUEA7xZdpGlUsg6awodhDinmkAU8UP6htWCQxDMMwDJNrSEpKEkH/irPBq0KjACnongYP0ehAGm1Lg5L+y+hATbBIYhiGYRgmV0Dz9fXo0UNpnjNN0NshaKoPGr1Ko8ZpNnzZxM3ahEUSwzAMwzC5gps3b4ppE2Qvds8Kmr+NpoyQTcZJn/TqHZr6QZvwZJIMwzAMw2RrfFGyysusaTJgTa8xopdnfwo0jx+95UARmiz5Qy66L4FFUi7hqEEp5GdSrj5Bfic5NfPN2/kVK5NPf4VFXiQsRvPs7/kJM6PPf6luXkNXN/9P/9euin6e6ZP8fhstnxhZxn+dV4tm61cVWfRbVYz9V1gkMQzDMAyjhI6B9h76KKBa9eXpmqxInwPFI6kKIvptZGQEbcIiiWEYhmGYbEOShWvtv0Avgqd30ylCv7X9ahkO3GYYhmEYRgldfR2tLdkBzY1E7zqVvVmNPull1LRem7BIYhiGYRhGCR0DXa0t2oKCtenl3ESrVq0QHR0tXqBM0wbQJ8UptW7dGtqERRLDMAzDMLneklSvXj0xPxJhZmYGDw8P3LlzB126dBFTAqxZswYmJibQJhyTxDAMwzBMrnzdzYd+V6hQAfv370d2wiKJYRiGYZhsG92Wl2GRxDAMwzCMEtkVcJ3X4JgkhmEYhmEYDbAliWEYhmEYJdjdJoVFEsMwDMMwSrC7TQq72xiGYRiGYTTAliSGYRiGYZTQ0WNLEsEiiWEYhmEYJXRZJAnY3cYwDMMwDKMBtiQxDMMwDKOEji5bkggWSQzDMAzDKKGjx44mgkUSwzAMwzBKcEySFJaKDMMwDMMw/9WS1KRJE7x69SpzZ319ODo6olevXhg0aBD+H7hx4wYGDBig9jbiT92eW9CVGKDejX14NG423l28idzMq5dPsH/DLLwO9EZhezd0HjIDDsXLZpn+0onNuHh0PRITYlGhZit0HDAVEkNjsS0lOQkHNs3Go1unYWBgiAZtB6NBm8HyfQ9tnocrp7YqHY/2r9Oib7blL9j/CQ5vmoU3Qc9ha++G9gNnwt456/xdPbkJl4+vR1JiLMpVb4W2/X6R5y/8jT+ObJmNAO97MDa1RK1mfVGvzVD5vke3zcP101uUjkf7U7rsgK73rnVz4XnjDAwkhmjSfhCath+oMW3gi6fYuXY2ggO8UdTRFT2HTYOTi/Q6pKen4fCOFbhx/iCSkxLgXqkeug2ZDAurQvB+fAsrZg3ReMxZf56CdaGiyAlC/J/g6JaZePPqOWzt3NC2/0zYOZfLMv3105tw9cQ6JCXEwr16a7Tp8wsM3t/Hp3dPY9eqMUrpy1RtgR6jVsi3/7tvGaLevUYRx9Jo3WcqihbLusx8ab3bu/59vXNwQ9eP1bvjm3H+6HqRH6p3nQYq17v9G2fjIdU7iSEathmMhm0z652MsNf+WDqpE+ZvvCdfN29cM0SEBaulbdF1NJp3GaWVvNL57dswBw9uSs+vUdvBaNROc58W9OIp9q6bhZBAbxRxcEXXoTPg+L6cKnJmvwdCX/uj97fz5OtiosLF/zx/cFX8T7UGHdG65zjo6eU+pw7HJEn57DszZcoUtGnTRnxPTU3F9evXMXXqVFhZWaFTp06fezjmK6BrKEHlLUthXq4kcjvJifHYsHgkKtVth+7D5+HG2Z3YsGQkfl56EhIjE7X0D2+ewpl9q9Dr24UwsyiIXWum4Ng/S9Fp4C9i+7Edi/HqxWMMn7xBNLy7PCbDqpAdKtRoKba/CfZFqx4/oFqDzLJsaGyWfflLiseW30agYu326PzNPNw6txNbfxuJHxafhMRQPX+Pb53CuQOr0G3EQphaFMK+vyfj1K4laNd/GtLT07Fl2UjYFy+PUbP2CsG0a/V4mBcojIq124n9Q4N90Lzbj6hcL2fyd2DrUgT4PcaY6X/jXVgItq6aCmuboqhcq4VSuqTEeKyePwrV6rVFv1FzcPn0Lqye/x1m/HEMhkYmOH1gHe5ePYEhPyyBqUUB7NkwH1tWTsF3v6xB8VKVMHfNOaXjrV/2E0zNrHJMINF93L58BMrXbIeOQ+bj9vl/sH35SIxdcErjfXxy+yTOH1yJLsMWift4cP1knN6zGG36ThfbQ4N9UbJiY7Qf+Kt8H30DQ/H59pU39q0Zj3YDZsHRrQqun94o/a/5p+Qi6z/nJzEe6xaPROU67dBzxDxcP7tT/J78m+Z69+DmKZzatwq9v10Ic8uC2OkxBUd3LEXnQdJ6d3THYgS9eIwRU6T1bufqyShA9a6mtN4RkeEhWL9kFFJTkpSOPXb2LiGSFev4id0rUK1+R2iLw9uWINDvEb79Zb04vx1/TUEBm6KoqHB+snL696KRqFK3HXp9OxfXzuzCukXfYvLvJ0Q5lXH3ylGc3LMKVepJ652MbSt/ho6ODsb8ug3xsVHYtnIijEzM0azTcOQ22N32he42c3Nz2NjYiKVo0aLo3LkzateujVOnTn3uoZivgFkZV9S5sgsmrk7IC3jeOA4DiRHa9p6AwvauaN9/MgyNTPHg5kmN6a+c3IJ6LfujTOVGcHQtjy5DZuL2hX3C+kAN/83ze9G+32TYF3dHuerN0LDdUFw7tV2+/9tXfmKbuZWNfJE9DWcHD9/nr2XPCbC1c0WbPpMhMTbBoyzyd+30ZtRuMQClKjWGg0t5dBw0C3cvSfMXFx2Gok6l0WHADBQs4oySFRvCxb0WArzvyPcPDfaDnXPO5I86lGtn96HboElwdHFHxRpN0bTDYFw8sUMt7d2rJ8WTdaf+P6GIgwu6DvoZRsamuHdd2q6kpaWhy8CJcHOvhqIOrmjUui98vaTWBn19A2FRki3PH90U1qjeI2Yip3h887gQMc17TISNnSta9Z4iyumTWyc0pr9xZgtqNR8ghBCJWhI89y7vQ0pSgtgeFuILW/sSMLO0kS9GJhZim+/jK7Cxc0PFOp1gbeuEpl1+RGxUKEJDfLSWn/vXj8PAwAjt+kjrXYf+k2FkZArPG5rL5eUTW1C/VX+4V5HWu65DZ+KWQr27cW4vOvafDIfi7ihfvRkatRuKK6cz692j22fw+y/doW8gUTu2mYU1LKxsxEJ15fT+v9C+70QUsLHXWjml8+s0MPP8Grcbgisn1cvp/WsnxDm07zteXJeOAybB0DjzuqSlpWLPul+x02MaChZ2VNo3NSVZCMiuQ35BEQc3uJSuigo1W+DFs7tayQeTi2OSyO1mYGCA2NhYTJ48WYimcuXKoVWrVjhz5ow83bFjx9CyZUuUL19eWKMUt23evBmNGzcW27p06YLbt2/Ltz1//hz9+/dHhQoVxP7btm2Tb/vjjz/w008/YcaMGahSpYr477Vr18q309P1kiVLULNmTbH8+eefaN68uXCLEdHR0ZgwYYLYt169epg9ezYSExPFNkpDLkY6dtWqVbFmzRq1vFOef/zxR1SuXFmc28OHD5GbsW5QA+Hnb+BKvZ7ICwT4PIBzqSri6YugT+eSVeDvfV8tLT1t0tNg8dLV5Ouc3CoiLTUFIQHPEBzwDOlpqShWspJ8Ox0rwPeBKCeJ8bGIjniDQkWccyh3QJCvJ5xKKOevmFsVBPpqzt+rF4/gXCozfw6u0vy9DvSCuZUteo5aJhrtjIwM+Hvfhf+z23AuXUOkJfcj5a9g4ZzJ3yv/56LTIEuPDNfSdO8eiuutyEtvT7iUVr4OtN/L557id5vu3wqRJXNZXD27DyXKZl4HGXQtju78Ay07D4eZRQHkFEF+9+FUoqrS+Tu6VUagn+b7GPzyIYqVrC5f5+CSeR9llqSs7pOJmZWwCAZ430VGejruX9knrIEFbJyytd4Vo3rnk3W9c9FQ74KzqHfFS1UR/yErB0/vXUTLbmOEkPoQF45uEGKpWoPOWsur7PycFc+PyqnC+ckI8PEU567cHlWGv7e0nJIgpLZm3OwdKFaiotK+JAD7jl6IQkWKid+vA33w+M45uJbJLAe5bcZtHS0teZn/5AhNSUnBuXPncOXKFcybNw9z587FixcvsH79ehgbG+Pvv/8WrrgGDRogJiYGEydOxK+//irEyokTJ4S4uHjxIoKDg7Fo0SKsXLkSbm5uQjB9//33YltycjKGDRsmLFYkYPz8/DBt2jSYmprK3XsnT55Enz59sH//fpw+fRqLFy9Gs2bNULx4cXh4eODAgQNYunQprK2tMXPmTAQGBsrzQOdH+dixYweSkpIwZ84ccY6UH4JisOgc9u3bJ4Sg4r4ECSg6p61bt+Ldu3eYNGkScjMBHupPR7mZmMhQEYekiJllQbwJ9FZLmxAXI0z1FgVs5evI10+dCsVu6OjowsTcCvr6EqVj0T7xsZF49zZQNHr/HvTAM89LMDW3Qv1WA1FVwfWWHfmjOCRFTC0L4m2Qev4S46PFuZIYUsyfsZkVot+9UUq7dHxTRIWHoFTFRihbrYW846X8XTi8Gt4PL4n96rYcpOR60ybREaHiGpKlRwY9SaekJCEuNhLmFtYKacNQxNFVaX8Ly4IIDlS2jhzdtQon9qyGiakFfpitHFtF3L12EvFxMajfqhdyErLk2NiVUFpHbrTQV592H3Xfl1MSsSRww1+/gO/jy7h0zEMIIfdqrdC40xjo6UtQtnobPLv/LzYs6AMdXT1RrvuMWy1i0LRFdGSosHYoQvfudZD26x1ZiroPk7oVfZ9kHR9JVqkrp7ah65CZ0NXV3pijGHk5lSjlVfH8FK8LxWepXZf35dTY1AJjZmU+xGfFqlkD4ed1W8R41W3RG7kRHS1e4/8rkUSigMQKQRYXIyMjDBw4EB06dBAxSoMHD0bJktJYlyFDhmD37t0IDw9HRESEECNFihSBvb292FaqVCkYGhoKIUKNt52dHRwcHIRAIqsSqfjDhw+jYMGCYh3h7Ows0pOQkokkiof6+eefoaenh2+++UZYkh49eiRE0vbt28W+ZCUiFixYgNatW4vvAQEBwpp18+ZN4UYkKG90XLKIyaBjFismVf+KIomE3/Hjx8W5lC0rDdwbNWqUEFmMdqCGUc8gs5MlqDFLTU1WS5uSLHVVKDZ2BO1Ppm5N22TmfdoeGvyCmgbY2hVH3RZ94ed1C3vXzxBP6eSayw5SkhOhp+JiyDJ/SVILp/4npO89eoXouA9t/hXHty9A235TERYizZ9NURfUat4PL71u4eDG6cLy5F61udbzlpycqH6u7++l7H58OK16vmo0aI/yVRvhzMH1WDV3OKYsPQBjk8yYqqtn9qBOky6QSIyQk9C90VMQg1mdv0ibLL2PJHgUod9pqcmICg8WZZl+dx/5OyLCgnBi+1yxHwVoJ8RGIDYqDK37ThMWKIp/OrhhCkZM3wdTi4LayY/4fw35UblvsrSa69b7epfx4Xr3qXhePw5DQxOUr6Hdsqqx7L0/X7VympSoMS+a7vOH6DxoMuLjorF/4zxs/WMChk5Y9cXnz+QykTR27Fi0aCF9MiWBQ7FJJE4IEhckOnbt2iWsK48fP5bHE5QpUwaNGjUSIorES9OmTdG9e3dhcSIBQ8Kqffv2cHd3l28jNx4dx8vLS7izZNDxZP9JkLBS/E1WJhJsZNl5+/atcOHJcHFxgaWl9InL19dXCDGydClC6/z9/ZWOrwmymtG5lC5dWr5O8b+Yz4esOOcOZbo1HV0rIC0lRSkNNUgGEvU4Gllgq2qDRftT3A25BVS3yRpBiaERqtTviDJVGoknYKKoUymEhbzE9bP/aE0kXTjsgYtHMvPn4EL5S/68/GlMrywK7ItLR1W1TknGHo8JaNlrAirV7YhSlTLzV8SxFMJev8TNf//JFpFkoKFTTX1/L1XjoDR1wPRbonIdbIpIXUr9R8/DtG+bwfPmGdRq1EnuhvP1uovuQ6cgu7l0dDUuHVW4j8UrCPeS6vmr3hfF+0iCSBH6TemtCtlj4vLrMDK1FA+PRZzKCGvS/r8nomWvSTizZylsHUqiRhPpiMT2A37Fql/aiJimem2GfVF+zh70wL8HM/Pj5KY5P5ri17Kqd6mfWO8+FQoOr1irldZHgtEoV011StP5ibQa8qLpPn8Iu2LSPqPXiDn4/ZeeeBf6CtZairHSFjy6Tcpnlzay6sisKqqQO+3evXvo2LEjevfuLQRUz57S2Beq7OT6evDgAc6ePSvcYmTloYUEFFmcyKJD7jtybZH7iz5J7FCc0fTp0lEfmiA3mCpksiaRJfuuuo0ggUMWpL1796rtX7hwYXh6esrF4KcikagHHjKfTq2mPcXwYRnnj6xDTFSYUpqYyDCYWxVS25c6f2qwabutnYtYRzExZDKnAGW67/ExkWKdrKGlJ3Jq4CgolsqoTEDIoGBqnyfS+DVtUL1xT5SrkZm/S0f/FuegCP2m81XF+H3+aLuNQv4S3ueP1gf43Id71WZK50+dXVJCHEzNC6jlj47j9/Q6sgNL68KIU7ne0ZHS621sIrXcyrCythXbFImODIdlAel9fnTnAhyKl4aVdWHxm4K8C9o6IC46Up7+qedVFLS1h51T9o/arNawF8pWk1qkicvH6T6GKqWJiw6FmYb7aGIqu4+hKFRUeh8pJka4dixt5PdaEQoGJ/dPQlwUgv0fo2az/kpukcKOpYUF6kup3bQnKirUu3OH16ndD6qHX1LvoKHeUVpZvfsUSIj4Pr2Fxu2/gbaxsLZVK6dZnZ+lta3Ypgj9tiigfp9VoZjHp56XxIg5mbuwsIPUxRwXHZHrRBKPbpOiNacjBTAfOXIEy5YtE9YmCo6OiooS26hzIqvNwoULRfD1Dz/8gKNHj4rRcZcuXRLCigRUrVq1hJuL4pUoPujOnTvC6kQWG7LmkDij5f79+9iyRT0eQRULCwvY2trKLVoydxkFaxN0bHKZiaDE98cmFyLFR1Ec0scgqxQJNMVg7SdPnnzhFWRkDS4FNsoWCn70974nF7YiIPn5XREYqgo1PI4u5fDyeeZokQDv+yLeg6xC9PRG3yn4UgaNLHFwKSf2PbXnD6ydrzzfTnCAl7zh11b+ChYuJl8c3SohwEc5fzTHkaOr5vyRhchfYbRaoI80fzRXTkRoEP5ZOVbEtcjP/+VjmJpbC4F0dt8KbFikPDfN6wAv4X7LDhycS4lO56X3A/k6P6+7KOZaVi2mxLlERbx47ql0Hfye3YNziQri9/4tS3DzwmF5+sSEOISG+KOwQ+a50/8UL5Vpcc5OSMRYFy4mXxxdKyHQV+U++twT7jBVSNTYOZcXgdcyKFBf7/199Hl0CYvG1pSPdCNeBzwV/0n3kmKZKHBbEYphsrLRbPHWVr2jelXsA/VOcZQWDaygcmmXVb17flfs86mxRSGBz5GemgInV2l50Cb2789PFnwtzu+Z5vOjdufl8/tK1+XF83sar4sqyckJ2LpivAhYlxH04gl0dfVgUzTnBoswX0kkkQWFXGc0FUBQUJAQP7LYHBIcJFjIOkSjy0ionD9/XsQWkXuN4ppWrVolrEm0Lwmo+Ph4EbNEsU4kXMiSRELrwoULIkCcLFqfAo2KW7FiBa5duybcdrJYIxJGrq6uqF+/PsaPHy8sXCSmaDv9N53vxzAzMxNWM4pjIqsTjYaj4HNGe5Sv3hIJ8TE4vGU+3rzyEZ8UpyR76qU4DQp+llGrWW9cOLoej2+fQaDvQ+zf+CtqNO4mzP60VK3fUUxMSdsozcVjG8SUAQRNG0DBlLR/+JsAXDvzD+5ePqg02aS2KVu9JRLjY3Bs+3y8feUjPmnOHZm1STV/NZr0FhNJPrlzBkF+D3F48yxUa9hd5M3epTzsnMti/7qp4ljPPS/g5M4laNh+hNiXpg14+ey22P/d2wDc/HcH7l85iLqtsyd/dE41GnYQE0T6+zyC582zOHt4Exq26Se2k6WC4kGISrWaiwDgvRsXIiTIV3zSfa5cWzpPTYOWvXD20AY8vnsRIYE+2PzHJBQq4igmlZRB62n6gK+BezXpfTyxY54QMPRJIqds9dby+6hoaareuDeunlwHr7tn8OrFQxzdMgtVGnQX8xw5ulaGvsQIhzb9grDXfvB+eBGndy9G3VZSKwqlu3txNzyvHsS7N/7C/RYZHoxKdbQXgE/zhlG9O0T1LshHfKrWOwpillGnubTe0VB+qlv7NvyKmgr1juY0ookpaRuluXh0A+q1yrSGfQya0NLa1lHjFAH/FTq/6g06Yu+6XxHg+xAPb53F+SMbUb+1rJyGyuPIKtZsIe7zwc0L8DrIR3yK61JLeT4lTdCovPI1mmH/xrliQko/rzvYtWY66rXsAyOFuLrc5G7T0dKSl9HXpkiiUWVkLSIrD1l+vv32W/z+++94+vQp2rVrJ4br03D81atXC5FDo9tkAdUkfEhAkbCiAG46FokYggKxabQZxTxRkHbfvn0xYoS04f8YFCBOcUljxowRcUvDhw8X0wvIXHRkNaIRbTRjOLnnSDT98ot0ArRPgUbakUiiWCuKdSJRRteA0Q7UeAz+6U/sWz8LN87tRlGnkhg8YbV8QjsK5ty9ZioWbpVa8CrVboOI0FciPcUOlKveHG16jZcfr13fn7F/w69YM2+QmMStedfRIg1B87v0G7sMp/euFFYlmoel96jFKFaiUvblz9gM/X74C4c2zcTt87tEnFD/Hz3kExDSPEr7103B7I1Pxe8KtdoiMixYpKdYJvdqzdGihzR/9ETaZ+wqHNk6G2vm9BaNPwVo12ou7YxoXqVe3/2Os/v/EFYlin3pNnIxnNyyz/rSZeAE7Fw7R8yITS62Nj1GoVJNqTtw6vDG6DtqtogpouDrEZNWCkFFwdd2xUpi5OQ/5RP01W/ZG0lJCdj59xzERkegdIXaGDHxD6UnfYpJolFvXwMK7u8zdjWObJmJuxd3obBDKfQZl3kfH988JoKrZ6yTDvEvV7MtIsNf4ciWGcKVRLNpN+8+QX6sfj/8jZP/zMfaX7tBYmSKqg17ok4r6czp5Wq0EUL68jEPRNOM205lMHDCRq0Fbcvq3ZDx0np3/V9pvRuqUO/uXzuOXWumYvG2zHpHcTU0EzXVu/LVm6Nt78x6177fz0I4rZ4rrXc0Wzal+VRio8PFyLHsokP/iUIk/TV7sDi/lt2+Q4X3AeKzvm2EniPnoEbDzuK6UJA1zYV07exu4dr9ZuJqpYkkP0TPEXNwcPNCeMyTCt5qDTqgbe8fkBvh0W1SdDJUA3byGTSNAM3ZRMP/CQrmphgniovKKiD7a3DUoBTyMylX878bMjk1bz8xfQpWJqnIz4TFqMc35jfMjJTn/smP6Orm625N0K5K9r7K5GG7xlo7VvkjyjPi5yVy3wtjtMzOnTtFcDi51MjFtnz5cjECLTcJJIZhGIZhch/53p5GsUxkkqeX8Pbo0UMM76f4J4ZhGIZhsh7dpqulJS+T7y1JNJSfYp0YhmEYhvk08nrAtbbI95YkhmEYhmGYLyHfW5IYhmEYhvk8eHSbFBZJDMMwDMMowe42KSwVGYZhGIZhNMCWJIZhGIZhlGBLkhQWSQzDMAzDKMEiSQq72xiGYRiGYTTAliSGYRiGYZTg0W1S+CowDMMwDJMrZtxOSkrClClTUK1aNdSrVw/r16/PMu3p06fRunVrVK5cGb1798bjx4+hbVgkMQzDMAyjFpOko6Xlc1i0aBEePXqETZs2YcaMGVi5ciVOnDihls7b2xs//fQTRowYgYMHD6JMmTLie0JCghavAoskhmEYhmFyAfHx8di9ezemTp2KsmXLonnz5vjmm2+wbds2tbRXrlyBm5sbOnXqBCcnJ/z4448IDQ2Fj4+PVs+JRRLDMAzDMGoxSTpaWpKTkxEbG6u00DpVvLy8kJqaKtxnMqpWrQpPT0/xcnpFrKyshCC6c+eO2LZv3z6YmZkJwaRNOHCbYRiGYZhsmwLAw8NDuM0UGT16NMaMGaO0jixBBQoUgEQika8rVKiQiFOKjIyEtbW1fH2bNm3w77//ok+fPtDT04Ourq74H0tLS2gTFkkMwzAMw2QbI0aMwODBg5XWKQohGRRPpLpe9lvV8hQRESFE1fTp01GxYkXs2LEDkydPxv79+1GwYEGtnTu72xiGYRiGybbAbYlEIlxhiosmkWRoaKgmhmS/jYyMlNYvWbIEJUuWRN++fVGuXDnMnj0bxsbG2Lt3r1avA4skhmEYhmGyLSbpUylcuLCwEFFckgyyFpFAsrCwUEpLw/1Lly4t/03uNvodHBwMbcIiiWEYhmGYr06ZMmWgr6+P+/fvy9dRYHb58uWFCFLE1tYWvr6+SutevHgBBwcHrZ4TxyTlElKuPkF+xqCOO/I7R2ddRX6nfl3t+fpzI0YS5RE0+RFrk0Tkd4oZ+CP/457v3t1mbGwshvTPnDkT8+bNw9u3b8VkkvPnz5dblczNzYVlqUePHpg0aZJwtdFoOJo6gKxInTt31uo5sUhiGIZhGCZXvJZk8uTJQiQNHDhQxC7RCLgWLVqIbTQDNwmmLl26iNFtcXFxYkTb69evhRWKJqDUZtA2oZORkZGh1SMyX8SBW2nIz/w/WJIOsCUpz/P/YEmys9TujMS5kf8HS5JjiextU4NGd9fasRxW7kZehS1JDMMwDMMoo5Pz7rbcCIskhmEYhmG+ekxSboRFEsMwDMMwuSImKbfBV4FhGIZhGEYDbEliGIZhGEYJdrdJYZHEMAzDMIwS7G6TwleBYRiGYRhGA2xJYhiGYRhGCXa3SWGRxDAMwzCMEiySpLC7jWEYhmEYRgNsSWIYhmEYRhkO3BawSGIYhmEYRgkdfi2JgKUiwzAMwzCMBtiSxDAMwzCMEjxPkhQWSQzDMAzDKMGj26SwSGIYhmEYRhm2JAn4KjAMwzAMw2iALUkMwzAMwyjB7rZcKJJSUlKwevVqHDhwAG/evEGhQoXQsmVLjBkzBmZmZlr9rz/++AM3b97Eli1boG1KlSqFzZs3o2bNmshuXr18gv0bZuF1oDcK27uh85AZcCheNsv0l05sxsWj65GYEIsKNVuh44CpkBgai20pyUk4sGk2Ht06DQMDQzRoOxgN2gyW73to8zxcObVV6Xi0f50WfZHb0JUYoN6NfXg0bjbeXbyJvIRjYT30a2UKe1t9hISmYcuJWAS8TtOYVmIA9GxmiiqlJKARu3e8krHrTBySUtTTjulhjtj4dGw4Eoec5nXAExzfOgNvXz2HjZ0bWvebhaLFymWZ/uaZjbh2ah2SE2JRplprtOw1DQaGxvC8ug9HNk5W30FHB1M9vLBlSX8EPFe/3xXqdEH7QfORnQT7P8HhTbPwJug5bO3d0H7gTNg7Z10Xr57chMvH1yMpMRblqrdC236/yOti+Bt/HNkyGwHe92BsaolazfqiXpuh8n2PbpuH66eV2y7an9JlB9Q2bF2zAHeunYXE0AgtO/YXiyb8/bywZfU8vPL3gZ2TC/qPnAJnV3exLSMjA4d2euDSmQNISkxA2Uq10HfYJJhbFhDbI8LfYse6xXj68BYkEkNUr9sCXfuNhoHEEDlJcnIyVvy1BpeuXoOhxBDdO3dE9y4dNaa9fus2Nmzehlchr1G0SGEM7t8HdWrWkG+/ePkq1m3eivDwdyjrXho/jhmFwra2yO3o6LCjKdeJpCVLluDq1auYM2cOHB0dERgYiLlz58Lf31+IJ20yZMgQ9O+vuZLnFZIT47Fh8UhUqtsO3YfPw42zO7FhyUj8vPQkJEYmaukf3jyFM/tWode3C2FmURC71kzBsX+WotPAX8T2YzsW49WLxxg+eQMiwoKxy2MyrArZoUKNlmL7m2BftOrxA6o16CQ/pqGxdsWrNtA1lKDylqUwL1cSeQ0SPWN7WuDGoyQhZhpWMcTYHhaY8lcEkjUIHxJIzkX1seyfaCADGNTODD2amWLLcWUhVN1dggpuElx9kIicJjkpHv+sGI5yNduj/eAFuHthB3b+MQKj5p6GxFC9nHrdOYmLh1ei49DFMLUoiMMbJuPs3sVo1Wc63Ku1gWvZ+vK06Wmp2PrbQJQo30j87vbtH0hLzbxQr154Yv+a71G1UZ9sz+OW30agYu326PzNPNw6txNbfxuJHxaf1JjHx7dO4dyBVeg2YiFMLQph39+TcWrXErTrPw3p6enYsmwk7IuXx6hZe4Vg2rV6PMwLFEbF2u3E/qHBPmje7UdUrpczdXHXpt/x0vcJJvzqgfDQEKxbMQMFbYqiWp1mSulI+CyfMxY1G7TGkDEzcf7kXiyfMw4L/joEQyNjXDi1F5fOHMSw7+fC1NwSWz3mYeOqXzFmyjIhoP5cNAGmZhaYNHcd4mKjsGHlLOjq6qLHoB+Qk3is34Tn3r5YMvdXvHkbikXLVqCwrQ0a1KujlM7vxUvMmrsQw4cMRI1qVXH77j38On8xVv22CK4uxfH4qRfmLv4NY0YOQ8Xy5eCxbiPmLFyKP5YuzNH8MF9OrpKK+/fvx7hx41C7dm04ODiIz5kzZ+LcuXN4+/atVv/L1NQUVlZWyMt43jgOA4kR2vaegML2rmjffzIMjUzx4OZJjemvnNyCei37o0zlRnB0LY8uQ2bi9oV9SE5KEILr5vm9aN9vMuyLu6Nc9WZo2G4orp3aLt//7Ss/sc3cyka+yJ58cwtmZVxR58oumLg6IS9SvYwhUlIysOffeLwOT8PO0/FITM5AtdKan6TT0oHtp+KEpSngTRoueybBzVH52cfESAfdmpjgRXAqvgZPbh0TloCm3SaiUFFXNO85FRIjUzy9c0Jj+pv/bkaNpgNRokJj2DlXQJt+s+B5ZS9SkhJEeTeztJEvD28cIvMEGncZL/Y1NrWSbzMxt8b5/ctQq+U3sHMun615fPi+LrbsOQG2dq5o02cyJMYmeJRFXbx2ejNqtxiAUpUaw8GlPDoOmoW7l6R1MS46DEWdSqPDgBkoWMQZJSs2hIt7LQR435HvHxrsBzvnnKmLJHzI8tN76AQUcy2DKrWaoFXngfj32E61tDcvnxT3usfA72Hn6ILeQ8fDyNgEt66ell6nO1dQvV4LlCpXFQ7F3MRxnj6UWv5ev3oJv+cPMXjMTNg7uaKkexV06v0tblzSXE6yi4TERBw/dQajhg9FCTdX1KtTCz26dsaBI8fU0p69cBGVKpZH5w7tYG9XFB3btRFi6MLlK2L77n0H0KxxQ7Rr3RKODvb4bsQ3eBcRgaioaOR6yN2mq6UlD6Ob22b4vH79uniSklG5cmUcPXoUBQoUQJMmTbBv3z75ths3bgjXFhEUFCS+r1q1CtWrV8fkyZNRvnx5cTwZsbGxYt3t27eFu40sSfRf9evXx969e+Xp6ImmQYMGOHjwoPhN6bt06YIKFSqgffv2OHlSueFbuXKlEHTkXtu9ezdyigCfB3AuVUU+Myp9OpesAn/v+2pp09PTEOj3CMVLV5Ovc3KrKJ66QwKeITjgmXgqL1ayknw7HSvA94G4RonxsYiOeINCRZyRm7FuUAPh52/gSr2eyIu42OvDJ0hZzPgGpcDFQbPRd/vJOPi+T1/QUhc1y0rwzF95/+5NTXD9YRJCwr6OSCJrjqNbVaVy6uhaBa98NZfTkJcP4VQys5zau1RCWloK3gR5KaVNiIvEtRNr0bjLT9A3kKgd68HVfUiIi0KdlsOQ3QT5esKphHJdLOZWBYFZ5PHVi0dwLpWZRwdXaV18HegFcytb9By1DIbGpqIt8ve+C/9nt+FcWurCIVc51cWChXOmLga+fI601FS4laooX1eiTCX4eT9SaqsJEjm0TfE6uJWuBL9nD8Rvsh49vH1JuNWSkxJx89JJOBWXtuGWBQrhh+krYWlVUOmYCfGxOZBLhTy8eInU1FSULSM9L6Kcexl4PfdWy2+Lpo3xzUB1j0RcXLz49Hz4GPVq15KvJ3fctvVrYGlpgbwwT5KOlpa8TK5ytw0YMAArVqzAmTNn0LBhQ9SpUwf16tWDm5vbJx/j7t27QvBQYY6KisLp06dRq5a0kJ4/fx7W1taoWrUqrl27JtaRKbdVq1YiXdeuXcW6+/fvIzIyEk2bNkVoaChGjBiBH374QYgp2jZp0iQULFgQ1apVw86dO0X80cKFC1GkSBHMmjULOUVMZKiIQ1LEzLIg3gR6q6VNiItBakoSLApk+sL19PRhYmaFqHevhf/ZxNwK+voSpWPRPvGxkXj3NlA0eP8e9MAzz0swNbdC/VYDUVXB9ZYbCPDYgbyMpZkugkOV44+i4zJgZ6P3wf0GtzNFnQpGCItMw5HLMfL1pYvpo6SjAWb+HSninL4GsVGhIg5JEXKjvX2lXk4T46NFmTOzzCynulROTa0QE/FaKe2d8zuEoChTtZXacUhcXDvxN2o0GyCsVjlRFykOSRFTy4J4G5R1HuncFeuisZkVot+9UUq7dHxTRIWHoFTFRihbrYVYFxrsK+rihcOr4f3wktivbstBSq43bRIZEQYzCyvoGxjI11lYWos4pbiYKHk8EREVEQY7R1el/S2srPEqwFd879BzOFbM/R7jv2kFXV09IYymLNgotpmYmqNc5Ux3FrXhZK0qUz4zvicnCH8XAUsLCxgo5LdAAUsRpxQdEwMrS0v5+mKOjkr7vvQPwD3PB2jfuiViY+MQExuLtLQ0/DxtlhBfpUuVwLhvR6BQIWUhyORecpXE++6777B48WIhNnbt2oWxY8eqWXk+xsCBA+Hk5ARnZ2e0bdtWiB9qMAmyALVu3VrtnTSU7sqVK8LSJEtHIo2Cxbdt2ybEWr9+/VCsWDF07NgRPXv2xKZNm0RaOk/6z8aNG6NMmTIiniqnINO8nkJFJkjkpKYmq6VNSU6Qb1eE9k9NSUZKcqLaNtnTOW0PDX5BzxawtSuOIRNWo3qjrti7fgYe3TqTDTn7/0VioIPUNGl5lZGSlgH9D2sknLieiPkboxAelY5xPS1AJZz26dfaTFibUr6OEUmQmpwAPdVypy9BmoZymposjZlStQzpqZRrqtP3L+9Gtcb9NP6n/7MbiI54jcr1eyAnoPqjp3LOWdbFJM151JS+9+gV6Pf9nwgJ9MLx7QvEurAQaV20KeqC/j96oFqDbji4cTqe3JG6tLKjnVEUSITB+3NPSVE+36SkRPW0lK/36cLeBovA77FTfsfEOWtRoKCtiDvSxO7Ny0UQeOe+3yEnSUpKUhJIhOw3DS7KCnKhzZq/COXcS6NOrRpISJS2uavW/C1cbrOnT0FKSiqm/jpXzSKVW0e36WhpycvkKksS0aFDB7FERETg8uXL2Lp1K6ZOnSp3q30Me3t7+XcSLrSvp6en2P/SpUvC6qNKpUqVYGNjgwsXLgjBdOrUKUyYMEFs8/PzEzFR5PaTQRWlePHi4ruvr68QdzLI6mVioh6oqQ3IinPu0Br5b0fXCkhTqbTUyBpI1GMT9A2kMS2qjTDtT7EM5AJQ3SZr2KhRq1K/I8pUaSQsT0RRp1IIC3mJ62f/EfFLzJfRpo4xWtfJvF8UN6Svp9yoGOjpaAzaViQkTGp98jgQgyVjCqCEkz7KukjgH5KKxy8+srOWuXJsNa4c95D/ti9eQU0Q0W+K4VFFT1ZOUzSlz7xOIf4PERPxBu7V22o8B6+7J+FaroGIUcoOLhz2wMUjmXXRwYXqYvLn1UWN6ZWviX1x6QjA1inJ2OMxAS17TUCluh1RqlJmXSziWAphr1/i5r//wL1qc2gbGumaqtLOpCi0DUppJRrSpiaLdCRs1y2fju4Dx6Fi9QZi27cTFmLi8LbCTedSsrySQDpzeDtGjF8gYpdyEonEQE0MyX4bGmqODYyIiMTEaTOF+Jk+eaLwUOjpSZ9sWrdojuZNpAMLJo//Ht37DcbTZ89Rtkxp5Gp4dFvuEkleXl5i6D+5sgiKQaL4H5oCoEWLFkqxRTLIjKmKYiEmsUJCiSxDsikFKK5IE23atBHpyFpEAq1RI2mhJt80ncfIkSOV0uvrZ146maVK0zZtUqtpTzFsX8b5I+sQExWmlCYmMgzmVoXU9qUGlRpn2m5r5yLWpaWlClcaBX1SHuJjIsU6Mv0TsVFhotE2MrEQ1jdZoyyDAlR9ntzIlrz+v3D+biJuPU2S/25VyxiWZsoiycJUB1Fx6k+eerpAxRISPHmRIoK7iZi4DMQmZMDMRBc13CWwMNXFH+OtxTaZNapKaUOMWfIu2/JUpWEvMWxfBsUNxamU09joMCWXmgxyq1E5peBlCvImKFYuPi5SBGPL8H10ScQt0fB4TdD2Bu1HI7uo3rgnytXIrIuXjv4t6osi9JvqlirkHqM80nYbhbqY8L4u0voAn/twr9pMqa5RzFJSQhxMzQuo1UU6jt9T9TZSG5C1JzZauW2IjgyHRGIkXGRKaa1tEBUZrrQuOiIcVgUKISYqAu/CXsPROXPUqXWhIjAztxIj5mQiadvahTh/Yg+++X4OqtVuipymUMGCiIqOFv2LTOiQCDI0lMDMVN11GxYWjvFTp4vvS+fPkbvjyGVHfQEFbMugdRYW5ngbGoayZXIsS8x/INdIRSqQGzZswJMnT5TWSyQSGBkZiVgiMnnGxWUObaYpAj4GWYbIQkRxTiSEPpSOXG4klChA3NhY+gRIFiOagoDEk2w5e/YsDh8+LLaXKFECDx8+lB+HAsijo7Nn5AI1jIWKFJMvxUpUhL/3PblIE0Gez++KgGxV6MnG0aUcXj6/K18X4H1fxHuQVciuWGnxPcDHU779xbO7cHApJ/Y9tecPrJ0/ROmYwQFecsHFfBnxiRkIjUiXL36vUuFqr2zqd3MwEOtVods+uL0ZKrhlpre20IWZiQ5eh6Vh8dZoEYv06zrp4umdLBb6np2Q9cbatph8ocDrIF/lchrkcxf2LurllII8izqXR6BP5kiuIL/7onMu7JD55B384gEcXKto/P/4mHeIDAuEg1tVZBdUFwsWLiZfHN0qIcBHOY80x5Gjq+a6SBYif4XRaoE+0rpYxLE0IkKD8M/KsSI4W0bwy8cwNbcWAunsvhXYsChz/jLidYCXcL9lB47FS0JPXx9+zzLbOe+n9+Hs5i7yoggJHV8vT6Xr4ON1X6w3NbcQLsbgQD95+pjoCMTGRqGQrVRIHNzpgQsn92LET/NRs7506pGcxrV4cSFunng9k6979PgpSpVwU8svjYSbNONX6Oro4LcFc1CooPSBhCCBVcLVRcQiKbrkoqNjUKRwHpgnid1tuUsklS1bVlhvRo0aJQQIiQ0Kkp4xY4YImCNrEo1M27NnD54/fy5Gtq1fv/6jx6VRajR9wMdEEsUT2draCvcexS3J6NOnDx49eoRly5bh5cuX4tx+++032NnZie0Uq0QuPBJXdF7k3lOtSNlF+eotkRAfg8Nb5uPNKx/xSfEDFd9bmyhOggJKZdRq1hsXjq7H49tnEOj7EPs3/ooajbsJdxstVet3FBNT0jZKc/HYBjFlgLg+lRvBz+u22D/8TQCunfkHdy8fVJpskvnv0GSQxkY66NncBEUL6YlPiUQHt99bmwz0pZYlIj0DuHgvEZ0bmcDNQR9ORfQwvJMZPJ+nIDgsDe+iM8UXLUnJGWKh7zkJBVYnJkTj9M65Yn4f+qRyKrM2UTml4G4ZVRv2wfWT6/Ds3hkEv3yAE9tmolL9HmIySRmhwd4oVFSzG4a2kaXGqpADcoqy1VsiMT4Gx7bPx9tXPuKT5k6SWZtU62KNJr3FRJJP7pxBkN9DHN48C9Uadhf10N6lPOycy2L/uqniWM89L+DkziVo2H6E2JemDXj57LbY/93bANz8dwfuXzmIuq2zpy4aGhqjTqN22Lx6Hl54P8bdG+dw8uAWNGvXWx6sTSPVCJo3KT4uBjvWLRFiiD4pTokmhSShW69JBzHn0rPHdxDk74O/f/8FriXLC8FF6Y/s+hutuwwSI+TouLIlJzEyMkSLJo2wfNVqMaLtyrUb2LX/oBjmT9AQfopbInbs2oOQ168x8Yex8m20xL5/mKdJKPcfPiqmBPAPDMTi3/+Aa3FnlC5ZArke6sd0tbTkYXQyVH1FX5GEhAQxaeSJEycQHBws3GU0uu2nn34SooSEEw3tv3fvHlxcXIQLjEadPXv2TGyj0Whk5aE5lhQhFx7FJR0/fvyDM27TOgrIpgktyYIlg37TRJckggoXLozBgwcLcSRj48aN+Pvvv5GYmIjhw4dj3bp1YpTe58y4feCW5hmVP0ag7wPsWz8Lb4P9UNSpJDoPngF7Z+nstrcv7sfuNVOxcGumde7cobW4fGKziH8oV705Og2cJp/Nljqu/Rt+xaNbp2BkYo4GbYegfqsB8n0f3zmL03tXIjTkJQrY2KNV93HiGJ+CQR3pOeUkbVOe4VrT/jk24/aBWVe1chyaHLJfa1MULaiHoLep2HoiDoFvpOWjTnlDYT0aNi9c7kLr1NAENcsZwtAAuPcsGTtOSedW0jQCjvgvM27Xr/tlo3JevXggZtwOf+0LW/tSYsbtIk7SMiGbRXvqmswn96vH14hZt6mclq7SAq36zJDH8hALv6uAbqNWKU0sqTgv06md8/D9ksuffZ5Gki8XkEF+D3Bo00wxhxHFCbUfOAN2xaR5vHtpP/avm4LZG5/K0188shZXT20SsUzu1ZqjXf/p8roYHfEWR7bOht+T60I41WzaFw3aDZcPOnl69yzO7v8D4a9fwqqQPZp1HScf/fYx7CylAcWfQ1JSArauno8718/C2MQMrToNQPP20tm9h3auIuY2IgFE+D1/hC0e8xAS9AIOxUqIGbeLuUitgDQibt/2P3Hr8kkkJyXBvVJN9P3mZzFC7tjeDdi79Q+N/79uf6YF/FMoZuCP/0JiYhKW/7kal65eh6mJCXp07YSuHduLbc3adcaE78egZbMmGDxyNAKDXqntT1MDyITT0ROnsH3XHkRGRaFiuXL4Ycy3sCmkHhLxuTiWyN42NWb5T1o7lvm4pcir5CqR9P/Ml4qkvMLXEEk5jbZEUm7mS0VSXuG/iKS8wpeIpLzGfxVJeQEWSf9ngdsMwzAMw+QS8ribTFuwSGIYhmEYRom8HnCtLVgqMgzDMAzDaIAtSQzDMAzDKMOTSQpYJDEMwzAMowy72wQsFRmGYRiGyRUkJSVhypQp4gXyNAXQh+ZDpOl/evfuLd6kQW/G0PRmjv8KiySGYRiGYZTQ0dHV2vI5LFq0SEzgTHMW0mTSK1euFHMnqhITE4MhQ4aI96XSJM/NmzfH6NGjER6u/Fqc/wqLJIZhGIZh1N1tulpaPpH4+Hjs3r1bvLmC3sJBwuebb77Btm3b1NLu379fTDg9c+ZM8bqwsWPHik8SWNqEY5IYhmEYhskVL7pPTU1F5cqV5euqVq0q3sSRnp6u9MovemMGvWVD9hJiYu/evVo/J7YkMQzDMAyj9rJpHS0t9P7V2NhYpYXWqRIaGooCBQoovRasUKFCIk4pMlL5xdz0gnt68f20adNQt25d9OjRA3fuZL40WluwSGIYhmEYRhl6T6COdhYPDw9hEVJcaJ2m97cqCiRC9ltVVJFrbs2aNbCxscHatWtRvXp1DB06FCEhIVq9DOxuYxiGYRgm215LMmLECPFieEVUxRBhaGioJoZkv42MjJTWk5utTJkyIhaJcHd3x5UrV3Dw4EGMHDlSa+fOIolhGIZhmGxDIpFoFEWqFC5cGBERESIuSV9fX+6CI4FkYWGhlJYsSC4uLkrrnJ2dtW5JYncbwzAMwzDZ5m77VMgyROLo/v378nUUZ1S+fHmloG2iUqVKYp4kRfz8/GBvbw9twiKJYRiGYZhsC9z+VIyNjdGpUycxrP/Bgwc4c+aMmExywIABcqtSYmKi+N6rVy8hkv744w/4+/tj+fLlIpi7Y8eO0CYskhiGYRiGyRVMnjxZzJE0cOBAzJo1C2PGjEGLFi3ENpqB+9ixY+I7WYz+/vtvnDt3Du3atROfFMhNLjttwjFJDMMwDMPkihfcGhsbY+HChWJRRdW9RqPk9u3bl63nwyKJYRiGYRhl+AW3Ana3MQzDMAzDaIAtSQzDMAzDKPG5L6bNr7BIyiUkp+Zv0+bRWVeR3+k0ow7yO7smX0B+Zlgvc+R3ohM/Pl9NXqf3z6HI71w+nM1/wO42AUtFhmEYhmEYDbAliWEYhmEYZdjdJmCRxDAMwzCMMp8xU3Z+hkUSwzAMwzDZ9oLbvAxfBYZhGIZhGA2wJYlhGIZhGGU4JknAIolhGIZhGGV4CgABS0WGYRiGYRgNsCWJYRiGYRhl2N0mYJHEMAzDMIwyPAWAgKUiwzAMwzCMBtiSxDAMwzCMMjxPkoBFEsMwDMMwyrC7TcBSkWEYhmEYRgNsSWIYhmEYRhke3SZgkcQwDMMwjDIckyRgkcQwDMMwjDIckyRgqcgwDMMwDKMBtiQxDMMwDKMMxyTlD5HUpEkTvHr1Sv5bX18fjo6O6NWrFwYNGvTRfUePHo0uXbogrxLs/wSHN83Cm6DnsLV3Q/uBM2HvXDbL9FdPbsLl4+uRlBiLctVboW2/XyAxNBbbwt/448iW2QjwvgdjU0vUatYX9doMle97dNs8XD+9Rel4tD+lyykcC+uhXytT2NvqIyQ0DVtOxCLgdZrGtBIDoGczU1QpJRGW4zteydh1Jg5JKeppx/QwR2x8OjYciUNeQVdigHo39uHRuNl4d/Em8hJORfQxqIMFHGwN8Co0FZsOReFlSKrGtBIDHfRtbY5q7kbiPt56nIjtJ2KQlJyBepWMMayLpdo+6ekZGDzzTQ7kBEhJTsLWNQtw59pZSAyN0LJjf7Fowt/PC1tWz8Mrfx/YObmg/8gpcHZ1F9syMjJwaKcHLp05gKTEBJStVAt9h02CuWUBsf1NSAC2rVkAHy9PmJpZommbnmjVeWCO5XHvhjnwvHkGBhJDNG47CI3baW5fg148xe51vyIk0BtFHFzRfeh0OLqot0mn93sg9HUA+nw7V74uPjYK+zbOw9P7l2AgMUK1+u3Rpuc46H6l+JgSLmaYMKoEXJxN8SIgDktWeeOZb6zGtOam+jj+T12ldZHRKWjX96r43qOjPcZ+46a0fcf+QKxa74dcCbvb8odIIqZMmYI2bdqI76mpqbh+/TqmTp0KKysrdOrUCfmV5KR4bPltBCrWbo/O38zDrXM7sfW3kfhh8UlIDE3U0j++dQrnDqxCtxELYWpRCPv+noxTu5agXf9pSE9Px5ZlI2FfvDxGzdorBNOu1eNhXqAwKtZuJ/YPDfZB824/onK9zGtqaGyWY/kl0TO2pwVuPEoSYqZhFUOM7WGBKX9FIFmD8CGB5FxUH8v+iQYygEHtzNCjmSm2HFcWQtXdJajgJsHVB4nIK+gaSlB5y1KYlyuJvAaJnp/6F8C1B4lYuy8KTaqb4Id+BTDh9zAkp2SopSeBVNzeAIs3vaPbiKGdLNGnlTk2HIrGjUcJeOiTJE+rpwv8PNgans8y12U3uzb9jpe+TzDhVw+Eh4Zg3YoZKGhTFNXqNFNKR8Jn+ZyxqNmgNYaMmYnzJ/di+ZxxWPDXIRgaGePCqb24dOYghn0/F6bmltjqMQ8bV/2KMVOWifpJaYu7uWPG0h1CMK35bQqsCtqiVoPW2Z7HQ9uWItDvMUb9sg4RYcHY/tdUFLCxQ6WaLVTyGI81i75F1bpt0efbObhyZhfWLhqFqb8fh6FRZpt098oxnNjzJ6rWk7YtMvasn4OYqHCMnrEZsdHvsGXlRJhZFkSjNgOQ0xgZ6mLxjHI4feEt5v7+DJ1aF8WiGeXRc9gNJCalq6V3djIRomjAd7fk69IVirOzoyn2HX2Fjf/4y9claDgOk7vIF/Y0c3Nz2NjYiKVo0aLo3LkzateujVOnTiE/8/DGcfG01bLnBNjauaJNn8mQGJvg0c2TGtNfO70ZtVsMQKlKjeHgUh4dB83C3Uv7kJyUgLjoMBR1Ko0OA2agYBFnlKzYEC7utRDgfUe+f2iwH+yc3WFuZSNfZFaonKB6GUOkpGRgz7/xeB2ehp2n45GYnIFqpQ01pk9LB7afihOWpoA3abjsmQQ3R+XnAhMjHXRrYoIXwZqtGLkRszKuqHNlF0xcnZAXqVnOSIihf07GICQsDduOx4j7WKOskcb0qWkZ2HI0Wlia/ENSceluAko4ScS2lFQgKjZdvtSpaAx6/t11OiZH8kLChyw/vYdOQDHXMqhSq4mw7vx7bKda2puXTworTI+B38PO0QW9h46HkbEJbl09LbY/vHMF1eu1QKlyVeFQzE0c5+lDqYUwOjIcTsVLot/IKShs54QKVeuhTPnq8H56PwfyGI8b5/ai88BJcCzujgrVm6FJuyG4fHK7Wtr7106INqlD3/EobO+KzgMmwdDYFJ43pG1xWlqqsDL94zENBQs7qu1PFqRGbQegqKMbSpStgap12sL70XV8DZrWt0Vycrqw9PgHxWP5Wl/EJ6SicT0bjemLOZog8FU83kWmyJfIqMynN2dHE/i8iFPanpCg2QqeKyDrna6WljxM3j77D0BuNwMDA2FZ+u2331CvXj1UrVoVY8eORUREhFr62NhYTJ48WYircuXKoVWrVjhz5ox8+7Fjx9CyZUuUL19eWK0Ut23evBmNGzcW28h1d/v27RzJY5CvJ5xKVIHOe7MofRZzq4JAX/WGMz09Da9ePIJzqWrydQ6uFZGWmoLXgV4wt7JFz1HLRINGZn9/77vwf3YbzqVriLSJCbGIjniDgoWd8bVwsdeHT5CymPENSoGLg2aD6PaTcfB9n76gpS5qlpXgmb/y/t2bmuD6wySEhOUdkWTdoAbCz9/AlXo9kRdxczSAd4Cy6Y9+03pNbDkaI09fyEoPtSoYwetlslo6U2MdtKlnKgRSag71PYEvnyMtNRVupSrK15UoUwl+3o+E9UcRv+cPxTbF+upWuhL8nj2Qnr+5JR7evoSI8LdITkrEzUsn4VS8lNhmZW2DkeMXwvh9/SRx9PzJPZQuVzXb8xgc8EyIG+eSleXrXEpXRoDPQ7U8vvR5AJdSlZXyWLxkZbz09hS/kxPjERzwHN/P3g7nEpnXTIaJuRXuXD4iHtyi3r3FU8/LsHcug69B2VLmePAkSmndw6fRKFfaQmN6shQFBidkebxiDlIRlVfI0NHR2pKXyXciKSUlRViQrly5gqZNm2L58uXYv38/5s2bh507dyI8PBwzZsxQ22/u3Ll48eIF1q9fjyNHjqBatWrCZZecnCz2mThxIkaMGIETJ06ga9eu+PHHHxEZGYknT55g0aJF4pjHjx8X+33//fdqjUd2EBMZCgsrW6V1ppYFEf3utVraxPhopKYkCTEkQ09PH8ZmVoh+pxy7sXR8U/w9ty8cXSuhbDWpOT002Fc0eBcOr8biHxph5bROuHf5AHISSzNdRMYoX9fouAwUMP9wMR7czhQLvisAC1NdHLmc2UiVLqaPko4GOHIl64YtNxLgsQNPx89HekLecQ8qYmmui4gYZRUTHZuGApYfvo/DOlti6Y82ohwcPK8eF0JuOyoft5/knKstMiIMZhZW0DfIFHgWltYihicuRrmDjYoIg1UBZSuEhZU13oW/Fd879BwOXT19jP+mFb7rU1+IoOE/zlf7z4kj2mLBlCFwLVUeVWs1RXYTHREGU3Mr6Otn5tHcsiBSUpIQHxupnJbapALKbRKljQyXtknGphYYN2sr7IpJxZ8q3Qb/guePbmDS4JqY+V0TWBawQcuu3+JrUNDaEGHvlMV4REQybApqtlyTpci2oCHWLK2M/RtrYeaEMihYQGrxLGBlAEsLA7RuVgS7/66JrX9WQ+/ODjmSD+a/kS9EEgmUypUri6VChQr4+eefMXDgQLRv3x67du3CDz/8gAYNGsDNzQ2zZs1CiRIl1I5RvXp1/PrrryhTpgycnZ0xZMgQIYJIIL1580aIryJFisDe3l5s+/PPP2FoaCiCxkk82NnZwcHBQQikxYsX54hISklOhJ6BtBLK0NeXIDVV/Sk7JUnaoep/Qvreo1eg3/d/IiTQC8e3LxDrwkJe0HMhbIq6oP+PHqjWoBsObpyOJ3ekroKcimUh14siKWkZ0Nf78H4nridi/sYohEelY1xPC+GOoX36tTYT1iZy2TDI2fuocs1T0gADvQ8/cR69HItf14QjLDJdxDSpPqA2rGqMMzdyNvCeLB6KAokweF/HUlKU61VSUqJ6Wqp/79OFvQ0Wgd9jp/yOiXPWokBBW2xYOUvtP7+buESkCXjxHP9sWIrsJjk5Qa3d0NOX/padu2I7oyimCNo3NVVD0KAG3oa8EEHeY2dtweAff0dIkA/+PbQOXwNDQ10kpyi348mpGZAYaO42nRyMYWKihz/+9sWMRU9QyFqCRdPLCW8TWZFkIuvn2Y+wdXcgBvQoJoK5c/XoNh0tLXmYfBG4TS60Fi2kFg8SLhSbpKenh3fv3gmhU7Zs5sgKEkpjxoxROwYFeJMLjUSVn58fHj9+LNanpaUJ4dSoUSMMHjwYxYsXFxaq7t27w9jYWLjxSpYsKQSZu7u7fBu5+7TNhcMeuHhkjfy3g0sFpKk0UiR4DCTqcUL6BoYaGzVpeuVYEPvi5cRn65Rk7PGYgJa9JqBS3Y4oVakRTMysxLYijqUQ9volbv77D9yrNkd20KaOMVrXycwLxQ3pq3Sk1LFqCtpWhOJeCI8DMVgypgBKOOmjrItExLc8fvFpjTfz5bRrYIr29U3lv31fpUC1ehjoQWPQtiLBoXQf0/Dnrkgsn2CDUsUkcrdbcTt9FLDQw/WHOWtdMzAwRGqKchmSiSMSPEppJRrSpiaLdORCW7d8OroPHIeK1RuIbd9OWIiJw9sKN51LyfLyfZzd3OX/s3bZVPQY+IOa+NJ+HpXbjbT3D1YGKjGJmgQR7StRaWM0ERrij0Nbl2D6yjPCgiQTXbvXz0aTDkOF5Ts76d/dSSwynjyPVhNEEn0dJCZp9uX2/+42MjJIVEqF1S8LnuDgptpwL2mB+4+i0KbPFUTHSJ8O/PzjYGVpgM6t7bDrYObo7FxFHhc32iJfiKSCBQuiWLFiaus/R6iQO+3evXvo2LEjevfuLYRWz57SmA+yFHl4eODBgwc4e/YsTp8+je3bt4uFBNTu3btx8+ZNnDt3Dvv27cOOHTvEZ+HChbWaz+qNe6JcjVby35eO/o3YqDClNPSbAqpVIbcaCSXabmPnItZRnEFCbKRIT+sDfO7DvWrmiBwKBqeYpaSEOJiaF5ALJBl0HL+n2RdUef5uIm49zXSdtKplDEszZZFkYaqDqDh1qx2NcqpYQoInL1JEUDARE5eB2IQMmJnoooa7RLjf/hhvLbbJrFFVShtizJJ32Zan/0fO3YrHzUeZ4qVtPVNYmimb/+i3qiuV0NMDKpcyxCPfZCQmSe9jdFy6mK7BzCSzLJQvYYhn/smIT/yw0NI2ZO2JjY4UdUnWiVOQNYkCE1Nz5bTWNoiKDFdaFx0RDqsChRATFYF3Ya/h6Jw5WtG6UBGYmVuJEXMFbe3g++wBqtRsLN9Owd8kSBISYmFuIJ0mIDuwtLZFXIx6HunhythEOY+W1oURE6ncJkVHhsFCxc2oiaCXT4VbTyaQCIpHovaHpgYgt112cuB4MP69LHV9En27OsH6vbtMBv0Oj1C31BNJKiPVKGg7OiYFNgWlx5AJJBkvA+NRKAvXHZN7yNdS0cLCAgUKFICXl5d83dOnT4XrLTExUSlom+KQli1bJqxSzZs3R1SUNJ6AnvB8fX2xcOFC4coj193Ro0fFKLpLly4JYUUCqlatWiLwm2KWkpKScOdO5qgwbUEipWDhYvLF0a0SAnzuiXOUnSvNceToqh4QSfOMkIXIX2G0WqDPfREDUcSxNCJCg/DPyrEiOFtG8MvHMDW3FgLp7L4V2LBosNIxXwd4CfdbdkEdXmhEunzxe5UKV3vlJ2Y3BwOxXhW6JIPbm6GCW2Z6awtd0bG+DkvD4q3RmPl3JH5dJ108vZPFQt8Z7RKXkIG379Lki09gCkqoBGmXcDIQQfia7iPFIlUqmdmZWFvSfdQV82TJcHVQDwbPCRyLl4Sevj78nj2Ur6OgarL2qM7tQ9YgXy9Ppfrq43VfrDc1txBWmODAzDlzYqIjEBsbhUK29gh78wp/Lhwvgrpl0LQD5hYFxJKd2BcrLcSRv7c0wJx48ewunFzKqeXR2a0CXj6/r5THl8/voZhbhY/+j2UBqRijKQBkvA32E1MHmFlIH2ayk5jYVLwKSZQvj72iUV4lSLt8GUuxXhUTYz0c31EHlctnPkiSu43ikPyDEtCuRRFs/6u62hxMAUG5N5CbA7f/D0QS0b9/fxG8TXMneXt7iwDtSpUqwcgo0/wrkUiE64wCvoOCgoT4ofgkggK3SWyRdYjikAIDA3H+/HkRi0TuNTrOqlWrhDWJ9iUBFR8fj1KlNAcmapOy1VsiMT4Gx7bPx9tXPuKT5k6SWZsoZomCu2XUaNJbTCT55M4ZBPk9xOHNs1CtYXcxjN/epTzsnMti/7qp4ljPPS/g5M4laNh+hNiXpg14+ey22P/d2wDc/HcH7l85iLqtlYVTdkKTQRob6aBncxMULaQnPiUSHdx+b20y0JdalmTzk1y8l4jOjUzg5qAPpyJ6GN7JDJ7PUxAcloZ30ZniixaamJAW+s5kL7eeJIqpF2j+IzsbPfFpKNHBjffWJrqPFJxNUGjfudsJ6NbMTAgpmvfqux5WuOeVJCahlGFva4DgtzkfXGZoaIw6jdph8+p5eOH9GHdvnMPJg1vQrF1vebA2jVQjaN6k+LgY7Fi3RIgh+qQ4pep1WwgRUq9JBzHn0rPHdxDk74O/f/8FriXLC8FV3K2smGJgw8qZYt8Hdy5j96blaNstc7LX7ILah2oNOoih+wG+D/Hw1lmcO7IRDVr3lVuKkpOleaxYswUS4mOwf/MCvA7yFZ9JSQmoVKvlR/+nWIkKKGzvgu1/TkFIoA98ntwS8zPVa9FbPlouJzl3JRRmpvoYN8xVBGXTp5GRLv69LG1TJRJdWFtJxX58Qho8H0dh7DeuKF3CHCVdzTBrojtu3H0nXGu37kWgoLUEo4e4wL6oEZrWt0Hfro7YtjcQuRaOSRLoZMgkfx7lY7NmU8D10qVLceDAATEdAMUWTZs2DZaWlkr7UjwSWYvevn0rArAp/uj333/HpEmT0K5dOyGclixZIkbAkXuPZvOm4HDi4MGDQkAFBweLAG6yRrVt2/az8rHr2pd1zkF+D3Bo00wxhxHFCbUfOAN2xaQxC3cv7cf+dVMwe+NTefqLR9bi6qlNIpbJvVpztOs/XcRKENERb3Fk62z4PbkuGsaaTfuiQbvh8gbq6d2zOLv/D4S/fgmrQvZo1nWcfPTbxzh9Tn3ahS+BOsl+rU1RtKAegt6mYuuJOAS+kVoU6pQ3FNajYfPC5S60Tg1NULOcIQwNgHvPkrHjlHRuJU0j4Ij/MuN2pxl1kNO0TXmGa03759iM27smX9DKcVzsDTCwvQXsbPQR+CYFGw9FI+C1VOTIZtEeOP21/D52a2aO2hWMYGggFcVbj0XL3W/E2mmFsXxHBB75aHaFfCrDeim7jz4FEgFbV8/HnetnYWxihladBqB5e6mAGNq5CgaPmSkEEOH3/BG2eMxDSNALOBQrIWbcLuZSWmyjEXH7tv+JW5dPIjkpCe6VaqLvNz/LZ9yOeBeK7WsW4OnDWyKOiWbcbtN1yGcLiOhEZRfSpwao7143Gw9unhYutsbtBqNhG+ms4j/0LofeI+egRkPpJLP+Pg+FoHr7yg9FnUqKGbcdiqsP46cJKQnFGbdpFBwJK5/HNyExMhEzbrfqOgp6KsHgH2PeDOks1/+VMiXMMf67EnB2MIHvyzgs/tMb3n7SkZWtmxbG1O9Lo177C/IZt0cPdUGdGgVFLNOlG+FY7uGDmDhpua7gboFRg13hVtwU7yKTsW1PIA6eCPnic7t8uCGyk/hLu7V2LJP63ZFXyfMiKb/wpSIpr6AtkZSb+RoiKafRlkjKrXyJSMprfIlIymtoSyTlZlgk5Qz5InCbYRiGYRgtksdnytYWfBUYhmEYhskVgdtJSUnifaw0MTNNsUMTPH8MigemeRJv3LgBbcOWJIZhGIZhcgWLFi3Co0ePsGnTJhHnS5NDU6wvvSosK2bOnCkGTGUHLJIYhmEYhlHmK4xKi4+PFyPF165dKyaBpoVGpW/bti1LkXTo0CHExWXfTPvsbmMYhmEYRokMHV2tLZ8KzWlIo9DJdSaDXkzv6emp8VVf9LJ6eg2YbMqe7IAtSQzDMAzDZBvJycliUYTmJ6RFkdDQUDEBtOL6QoUKiTglesWYtbXypKILFixA586dNb6PVVuwSGIYhmEYRhktTuDp4eGBlStXKq2jOQpV36OakJCgJpxkv1VF1tWrV8WbLehtGdkJiySGYRiGYZT4HDfZxxgxYoSYoFkRVTEke0G9qhiS/VZ8Swa9Vmz69OmYMWOG0vrsgEUSwzAMwzDZZkmSaHCtaYJeCk9xRhSXJHtBPbngSAjR68Fk0Mvm6RVh9HYLRYYNG4ZOnTppNUaJRRLDMAzDMF+dMmXKCHF0//59MU8SQS618uXLK71MmV42T+9aVaRFixaYM2cO6tatq9VzYpHEMAzDMMxXnwLA2NhYWIJo3qN58+aJd6nSZJLz58+XW5XMzc2FZalYsWIaLVH0blVtwlMAMAzDMAyTK2bcnjx5spgfiV4gP2vWLBHcTVYigmbgPnbsGHIStiQxDMMwDJMrMDY2xsKFC8WiyrNnz7Lc70Pb/gsskhiGYRiG+eruttwIiySGYRiGYZTIgPZGt+VlWCoyDMMwDMNogC1JDMMwDMNk22SSeRkWSQzDMAzDKMMiScBXgWEYhmEYRgNsSWIYhmEYRonPnd8ov8IiiWEYhmEYJTgmSQqLpFyClUkq8jP162p3qvjcyK7JF5Df6TG/IfIz/1jdRH7H3iF735qeG1i0qMrXPoW8D1uSBCwVGYZhGIZhNMCWJIZhGIZhlGB3mxQWSQzDMAzDKMEzbkthqcgwDMMwDKMBtiQxDMMwDKMEu9uksEhiGIZhGEYZHt0mYKnIMAzDMAyjAbYkMQzDMAyjRAbbUAQskhiGYRiGUYJfSyKFpSLDMAzDMIwG2JLEMAzDMIwSPLpNCoskhmEYhmGU4MkkpbBIYhiGYRhGCbYkSeGrwDAMwzAMowG2JDEMwzAMowSPbpPCIolhGIZhGCU4JkkKu9sYhmEYhmE0wJYkhmEYhmGU4MBtKSySGIZhGIZRgt1tUlgkKbBv3z5MnjwZc+bMQffu3ZGbSUlOwq51c+F54wwMJIZo0n4QmrYfqDFt4Iun2Ll2NoIDvFHU0RU9h02Dk0tZsS09PQ2Hd6zAjfMHkZyUAPdK9dBtyGRYWBWC9+NbWDFriMZjzvrzFKwLFUVO8DrgCY5vnYG3r57Dxs4NrfvNQtFi5bJMf/PMRlw7tQ7JCbEoU601WvaaBgNDY3he3YcjGyer76Cjg6keXtiypD8Cnt9U21yhThe0HzQfOYVTEX0M6mABB1sDvApNxaZDUXgZkqoxrcRAB31bm6Oau5F4afetx4nYfiIGSckZqFfJGMO6WKrtk56egcEz3yAvoCsxQL0b+/Bo3Gy8u6h+b3IrDja66NXMGHaF9BASnoZ/ziQg8G26xrQSfaBrY2NUctOHjo4O7j1Pwd4LCUhOUU6nrwdM7GuG3f8mwDsoDV+bwlZAq6q6sLEEwqKBE3fS8Tri4/u1rqaDmATg8uMM+ToTQ6BlFV04FwYSkoArTzPw8GXm9pxoT7d4LMTta/9CYmiIVh37o1WnfhrT+vt5YfNf8xHk7wM7J1cMHDkZzm5lxLaMjAycOLgV/x7bjbjYaFSt1Rh9h02AkbGJfPuBf9bg/Im9SEtLRbXaTdF32HjRhjO5E7anKXD06FE4OTnh4MGDyO0c2LoUAX6PMWb63+jxzS84secv3Lt+Si1dUmI8Vs8fBdfSVTBxwU4UL1kJq+d/J9YTpw+sw92rJzDkhyX4ad52xMVFYcvKKWJb8VKVMHfNOaXFtUwVVKjeJMcEUnJSPP5ZMRyOJaph6C/74OBaGTv/GCHWa8LrzklcPLwSbfr9ir4/bcIrP0+c3btYbHOv1gbjFl+WL2MWnEcB22Ko0WSA2N7t2z+UtncbtQp6+gao2qgPcgoSPT/1L4Dn/imYsToMPgHJ+KFfAbFeEySQitsbYPGmd1i48Z343qeVudh241ECxi56K19+WPIWr8NTcfq65muX29A1lKDy1t9gXq4k8hIker7tbArfV2lYuC0WfsFp4jet1wQJJKfCeli5Nw4r9sSiWBE9dG1orCaQBrUxEaIrN2CgB/Sor4vAsAxsOJ2OoLAMdK+nK9Z/iJqldFDJRb3b6VpXF+YmwPbz6ThzPx1NK+qgpD1yjJ0bl+Ol71NMnL0a/UdMwsGda3Hr6hm1dEmJCVg2exxKuFfGjKVb4Va6ApbN+V6sJ86f2oeD/6xB136jMHXBOkSEv4XHb1Pl+x/btwn/Ht+NET/NxY/TV+Dpw1s4sHNtzmX0M91tGVpa8jJ5++y1SHh4OK5du4bvvvsOt2/fRmBgIHIrJHCund2HboMmwdHFHRVrNEXTDoNx8cQOtbR3r54UTymd+v+EIg4u6DroZxgZm8oFVVpaGroMnAg392oo6uCKRq37wtfrntimr28gLEqy5fmjm8Ia1XvEzBzL65Nbx8T5N+02EYWKuqJ5z6mQGJni6Z0TGtPf/HczajQdiBIVGsPOuQLa9JsFzyt7kZKUAAOJEcwsbeTLwxuH6NEOjbuMF/sam1rJt5mYW+P8/mWo1fIb2DmXz7H81ixnhOSUDPxzMgYhYWnYdjwGickZqFHWSGP61LQMbDkaLSxN/iGpuHQ3ASWcJGJbSioQFZsuX+pUNBYG9F2nY5DbMSvjijpXdsHE1Ql5jSqlDMS1338xEW/epWPv+URxDyuXNNCYPi0tQ1iHyNIU9DYd1x4nw8U+U20UsdbF+N5mKGSVe5rrMk46SEkD/vXMQHgMcOZ+BpJTgdKOmsU8CcTOtXVRu7QOouKULURFCgAOhXRw6Ho63kQCPiHA9WcZqFUqZ/JLAufimYPoM/QnOLuWFtaf1p374+zRXWppb1w+BYnECD0HjYOdY3GxD1mJbl2RCqqzR3eiZcd+qNWgFeydXPHNuFnwvH0ZIa9eIj0tDScPbUPPQd/DvUJ1uJQsh069R8Df5ylyq7stQ0tLXib31LqvzIkTJ2Bubo4OHTrA1tZWyZqUmJiIqVOnomrVqqhfvz52794Nd3d3BAUFie0hISEYOXIkKlasiCZNmmDlypVCfGQXr/yfC1MtWXpkkKXI3/sh0tOVTfovvT3hUrqKMOMT9En7vXzuKX636f6tEFlETFQ4rp7dhxJlq6n9Z1pqCo7u/AMtOw+HmUUB5BSvXnjC0a2q0vk7ulbBK9/7amnJdRjy8iGcSmaev71LJaSlpeBNkJdS2oS4SFw7sRaNu/wEfQOpqFDkwdV9SIiLQp2Ww5CTuDkawDtA2c9Cv2m9JrYcjZGnL2Slh1oVjOD1MlktnamxDtrUMxUCKfXre2o+inWDGgg/fwNX6vVEXqN4UX34Biu7R8maVNxOs5ll17+JYjthbaGDaqUN4B2Yub+bgz6eB6Zi6Y5Y5BbsrCGsR4rQb/uCmtNbmUqtYWR1ioxT3aaDuMQMpfVvIzNQxBrQzYH+NeDlc6SlpsKtdEX5upJlKsHP+7Fae+r37BFKlKmo1B6VKF0RPs8eiN+hr1/BpUTZzLxZF4K5RQH4ej3Eq0A/xEZHokrNRvLttRu2xvhZq7I/k8wXwzFJCq62Ro0aQVdXVwidAwcOCKsSVQKKUbp37x7WrVuH1NRUIZhkIoh8zKNHj0bp0qWxf/9+hIaGYvr06WI/2j87iI4Iham5lbD0yDC3LIiUlCTExUbC3MJaIW0Yiji6Ku1vYVkQwYE+yvnftQon9qyGiakFfpi9Re0/7147ifi4GNRv1Qs5SWxUqIhDUsTUoiDevvJWS5sYH43UlCSYWdrK1+nq6cPE1AoxEa+V0t45vwPmVrYoU7WV2nHonl478TdqNBsgrFY5iaW5Ll69Ve5go2PTYF/4w1V1WGdL1KtsjNCIVBw8r96ZNqlugsiYdNx+koS8QICHulU0r2BhqiPikBSJic9A0YIffibt39IYNctKEBaVjuMKLtHLD9RF79fGzFgHYVHKIikuCbCxIPGgHkv0NgrYfVlzTFZcUgaMDHSEiJIJeAsTHejp6sDQAEjI5uxHvQuDmYUV9A0y21MLq4IiTik2JgoWlpkPhZERYbB3clHa38LKGq8CfOX7Rb4LVbJSxcVGISYmUggoUzML+Hh5Yu/WVYiJjkK12k3QfeAYGGh4UPva5HU3mbbgq/DeEnT37l00a9ZM/G7RooVwt925cwdxcXFCME2bNg2VKlVCtWrV8Msvv8j3vX79OoKDgzF79my4uLigZs2a+Pnnn7F58+ZsO9/k5EQ164esgqemJH9CWglSU5XT1WjQHhPm/4NS5Wth1dzhSIhX7mivntmDOk26CFNzTpKanAA9feXzp99pKucvTZsoPlXzS+kV80si6P7l3ajWOIvAzGc3EB3xGpXr90BOQ7FHqSox2uTWMND78CP10cux+HVNOMIi00VMk+pkuQ2rGuPMDZVHeCb77mGaultU/yP38PStJCzZEYuI6HSM6mKaq50UFHuUpqJ56LlR7wtCpoLDgZhEoEVlHXHcAmZAjZLS3OvlQA+VnJQIAwWB9MH2NEm9PSWBk5oitebWqNccR/duRHDgCyGydqz/TaxPS0lBYmK82H/PlpXoOfgHDB0zDfdvXcTODb8jN8LuNilsSXpvRTI0NES9evXE7xo1asDS0lJYhmh9SkoKypfPjEupXLmy/Luvry8iIyOFK04GmWjJRRcREYECBbTvmpJWSuXKK6ukEkOVgE+NaZMhkSinsykijf3oP3oepn3bDJ43z6BWo05yN5yv1110HyoN6M5OrhxbjSvHPeS/7YtXUBNE9Jvii1TRM5COEFHNrzR9Zn5D/B8iJuIN3Ku31XgOXndPwrVcAxGjlN20a2CK9vUzrVW+r1Kgr1IrqeOgOKUPERxKvXIa/twVieUTbFCqmETuditup48CFnq4/lAqIhnt0qKGIVrWyByd9PJ1mrCKKEICKSX1w/fw9Tup6lh/NB5zh5vDzUEvV4xiI2qX0UGd0pmdXfA7dQFDAklV4H8KJLYOXE1Hp9q6+LGzDuKTpDFJzSrpIOkLjve5UMwjtfGa21PldsZAot6eplB7aii9/x16fCMsRr+M7QE9PX00atkFjsVLwcjEFHp6ekhOTkKfb8ajdDlpf9Fr8A9Y/dtUsY68GLkJfi2JFBZJ70USiRpFoUPuNIpT6tatm1p6skTIIPcbWZD+/PNPtXQU45QdWFoXRlxMpIhLoopIREeGCeFgbKL8n1bWtmKbItGR4bAsUEh8f3TnAhyKl4aVdWF5g1HQ1gFx0ZHy9E89r6KgrT3snLJ/lFGVhr3EsH0ZFDcUF6V8/rHRYUouNRnkVtM3MERcdJgI8ibS01IRHxcpgrFl+D66JOKWjE3Vh8fLtjdoPxo5wblb8bj5KFO8tK1nCksz5R6WfpOrTBXqlCqXMsQj32QkJknLZHRcOmLj02FmktnAlS9hiGf+yYhPzLkh1f9PkDvs7vPMTrZ5dUNYmCp3eOQ+iorTcA91gfKu+vDyT0VicqZrjmJ0KI4st3DPNwNegZnlp1ZpHZiqPKeYGQGxX1jGQiKAv46li2OSSHIpDFFeKQA+u7EqaCNihRTb06jIcEgkhjAxVW5PC1jbIioiXGldlEJ7amhkjFETFyA+LlZYc41NzDB2YHMUsrWTTwNQ1MFZvm8R+2LC4hQTHQFLqywCupivSu6Srl+BFy9e4MmTJ8KFRm412bJs2TLExsbC399fmGIfPXok30fxe/HixYW7zdraGsWKFRMLBXSvWLFCHtynbRycS4nK/NJbGixI+HndRTHXsmpPI84lKuLFc0+5sKNPv2f34Fyigvi9f8sS3LxwWJ4+MSEOoSH+KOyQ6Xen/yleKtN6lp2Q9cbatph8ocDrIN97Sucf5HMX9i6ZQZYydHR1UdS5PAJ97sjXBfndF9eqsENp+brgFw/g4FpF4//Hx7xDZFggHNwyBXN2EpeQgbfv0uSLT2AKSqgEaZdwMoBvkMqkOeJaSGORKpXMtGJYW+rCzEQXIcKyJMXVQT0YnNEe1JmTm1O2vAhOhYtKkDaNVnsZkqbxHvZvaYKyxTPveQFzHSGQZJal3AAJuIjYzOVVuHREmiIOBXXE+s/FSAL0a6wLYwkQlyi9Jq52OvDPDO3JVpyKl4Kevj58n2W2695P7qN4CfX21KVUORGkrdge+Tz1hGspqadh18bluPzvEZiYmgmBRMHfCXGxYqqAYi6lRBxp4IvMeMrgoBditLGZueYHtq9JRoaO1pbPISkpCVOmTBGhLeTdWb9+fZZpz58/j44dOwrvTvv27XH27Flom/97kURWJCsrK/Ts2RMlS5aUL23atIGbmxsOHz6MLl26YO7cufD09MT9+/fFd4JEEN1Ee3t7TJgwAc+ePRPTB1D8krGxsTCvZgfkUqvRsIOYINLf5xE8b57F2cOb0LCNNMaGLEcUi0RUqtUcCXEx2LtxIUKCfMUnTRpZuXZLsb1By144e2gDHt+9iJBAH2z+YxIKFXEUk0rKoPU0fcDXgAKrExOicXrnXIQG+4hPOn+ZtSklOVEEd8uo2rAPrp9ch2f3ziD45QOc2DYTler3EJNJyggN9kahosrB4IrbyBplVcgBX4NbTxJhYiSdINLORk98Gkp0cOO9tclAnyxL0mpLA2/O3U5At2ZmQkg5F9XHdz2scM8rSUxCKcPe1gDBKsHgTPZx3zsFxoY66NrISAzfp0+Jvg7uPkuR30Pz95a+9AypJap9PSMhrBxtdTGkrQke+qbidXjuEUmqkFWJgqrJJVbQQvpJ+ZJZm8jdqGpp+pAAoykCGlfQEaPgKhbXQQVnHdzwypn8GxoaoW7jtti8ep4QNXevn8eJg1vQrJ10kEpURJiIJSKq12kqBrBsX7dUjFajz6SkBNSo21xst7K2waGda8VxXvo8xdpl09C4VTchgkg0NWjRCdvWLoLvs4fw8XqA3Zv/QIPmHeUWrNxEBnS1tnwOixYtEoaITZs2YcaMGWK0OHl1VPHy8hKDprp27SoMG7169cK4cePEem2S++7MVxBJpEAlEvXRBb179xaC6MyZM8KyNGjQIJiZmaFv377iN1mYSAj99ddfInC7R48eMDExQatWrUTwdnbSZeAE7Fw7R8yITS62Nj1GoVJNaeD51OGN0XfUbBFTRBVzxKSVQlBR8LVdsZIYOflPGBpJTb/1W/YWlXzn33MQGx2B0hVqY8TEP5SeoCgmiUa9fQ0Mjc3QY7SHmHH73qVdsLUvhV5j10BiKD3/J7ePiVm0p655Jn6XrdEWUeGvcHzrdBGsXbpKCzTtOkHpmOSOM8oiP3HR4TA0tsg2K+DHILfZsm2RGNjeAo2qmSDwTQqWbomQxyTVLCedRXvgdOlovT1npHMeje5pBUMDHdx+moStx6KVjkmiKi4x93a4+Q3q9FcfiEevpsaoW0Ei4sX+2h8n5hEiqpQ0QP9WJhj9W5T4ffiKtAMe2s5EBH17+qRgzznp5IS5FcoLjVajGbcrueggNArYdSldDDIgyjjqoF0NXczf9WkxVQeuS481tCW5JYED19KFCy6n6DXkR2xePR+Lpo0UbWanXiPEyDPi+8GtMHTMDNRr2l5s+37qMmxaPR8XTu2HYzE3/DBtuXCzEc3a9kTY2xAs+3WssGzXadhGjF6T0Xvwj9i1aYWYkJKsUDQFQLf+OePazwvEx8eLKXbWrl2LsmXLisXb2xvbtm0T/aoiR44cQa1atTBggHQyYPLi/Pvvvzh+/LgYba4tdDIUA2wYjZBIql27NkxNpQG2Dx48QJ8+fcS0AKqjIr6UU565b5ivNnkdmfuGuGqbs2eVpxnIj/SY3xD5meML886rT74Uewcz5Hcals8bs8r/F+qUyZ6YVxnPfQO0dqySnzgpLI0y79evn/DYyAwXN27cwLBhw8Q6xYd3GjRFAfeKgmjo0KFCLNE0PNri/96S9CmQue/cuXMYPny4mBJg8eLFYi4lbQkkhmEYhslNaHPofnJyslgUIRGk6sGheQZpRLji+kKFCok4JRpFTrG/Mlxdlef/I4sTvTWD3G7a5P8+JulTWLJkiQjG7tSpEwYPHgwHBwd5XBLDMAzDMFnj4eEhRo8rLrROlYSEBDXhJPutKrIUeffuHcaMGYMqVaqgaVPpGyS0BVuSPgEK4KYgMoZhGIb5f0CblqQRI0YIA4MimuKAaV5CVTEk+21kpHkkQFhYmDg2RQ7RqHJtzzfFIolhGIZhmGwTSRINrjVNFC5cWEzCTPMP6r+fVZdccCSQLCzUB9u8efNGHrhNb7lQdMdpC3a3MQzDMAzz1SlTpowQRxSkLYNeD0ZvvFC1ENFIuG+++Uas37p1qxBY2QGLJIZhGIZhvvpkksbGxiL2d+bMmWIUOY0sp8kkZdYisirR2zEIimkKCAjAwoUL5dtoiYmRTouiLdjdxjAMwzCMEl/rxbSTJ08WImngwIFiXkIKyKaXzhM0efP8+fPFBM8nT54Ugql79+5K+3fu3BkLFizQ2vmwSGIYhmEYJleIJGNjY2EdklmIFKG3WsjQNAt3dsDuNoZhGIZhGA2wJYlhGIZhmFxhScptsEhiGIZhGEaJzwm4zs+wu41hGIZhGEYDbEliGIZhGEaJdHa3CVgkMQzDMAyjBMckSWF3G8MwDMMwjAbYksQwDMMwjBIcuC2FRRLDMAzDMEqwu00Ku9sYhmEYhmE0wJYkhmEYhmGUYHebFBZJDMMwDMMowe42KSySGIZhGIZRgi1JUlgk5RLCYgyQnzGSpCO/M6yXOfI7/1jdRH6m9c81kN9pdGE+8jt+BnWR/8n/7U1ugEUSwzAMwzBK5P/H2k+DRRLDMAzDMEqwu00KTwHAMAzDMAyjAbYkMQzDMAyjBI9uk8IiiWEYhmEYJdjdJoXdbQzDMAzDMBpgSxLDMAzDMEqwu00KiySGYRiGYZRIz/jaZ5A7YHcbwzAMwzCMBtiSxDAMwzCMEuxuk8IiiWEYhmEYJXh0mxQWSQzDMAzDKJHBMUkCjkliGIZhGIbRAFuSGIZhGIZRIp1jkgQskhiGYRiGUYJjkqSwu41hGIZhGEYDbEliGIZhGEYJDtyWwiKJYRiGYRgleJ4kKexuYxiGYRiG+X+xJKWkpGD16tU4cOAA3rx5g0KFCqFly5YYM2YMzMzM0KRJE4wePRpdunRB//79UaNGDbEtKzZv3ox//vkHAQEBsLS0RMOGDfHDDz/AxsYGX5sQ/yc4umUm3rx6Dls7N7TtPxN2zuWyTH/99CZcPbEOSQmxcK/eGm36/AIDQ2Ox7end09i1Svk6lKnaAj1GrZBv/3ffMkS9e40ijqXRus9UFC1WNptzCAT7P8HhTbPwJug5bO3d0H7gTNg7Z/2/V09uwuXj65GUGIty1Vuhbb9fIHmfx/A3/jiyZTYCvO/B2NQStZr1Rb02Q+X7Ht02D9dPb1E6Hu1P6bKDlOQkbF2zAHeunYXE0AgtO/YXiyb8/bywZfU8vPL3gZ2TC/qPnAJnV3exLSMjA4d2euDSmQNISkxA2Uq10HfYJJhbFhDb34QEYNuaBfDx8oSpmSWatumJVp0H4mvhYKOLXs2MYVdIDyHhafjnTAIC36ZrTCvRB7o2NkYlN33o6Ojg3vMU7L2QgOQU5XT6esDEvmbY/W8CvIPSkBfQlRig3o19eDRuNt5dvIm8QlJKChZsOYSztx/BSGKA/q3qi+VDBIdFoPsvv2P59wNRrbSL2vbZG/bBpoAFRnZqhq9NcnIS/v5zGa5fvQiJRIIOXXqJ5UM8ffwAfyydiz/X71Raf+3yeWzftAbvwsNQyr0cvh07ETa2RZDb4Xe35WNL0pIlS3Dq1CnMmTMHJ06cwPz583HlyhWMHz9ebN+zZw/atGnzSccigbRu3TqxLx3rjz/+wMuXL/HNN98gPV1zo55TJCfFY/vyEXAqURXDp+2Fg2tlbF8+UqzXxJPbJ3H+4Eq0GzALAyZswis/T5zes1i+PTTYFyUrNsZPv12SLx0GzRHb3r7yxr4141GvzXCMnHkARZxKi/9KSUrI9jxu+W0EipWsipEz98DRrTK2/pZ1Hh/fOoVzB1ah46CZGDxxIwJ9PXFq1xKxje7XlmUjYWJujVGz9qLDwBk4f3g1PK8dUbgGPmje7UdM/P2ifKlSv0u25W/Xpt/x0vcJJvzqgX7DJ+HQzjW4ffWMWjoSPsvnjEUJ98qYtmQrXEtVxPI548R64sKpvbh05iCGfT8XP89dh8h3odi46ld5vimtuUUBzFi6Q4irI3vW4frF4/gakOj5trMpfF+lYeG2WPgFp4nftF4TJJCcCuth5d44rNgTi2JF9NC1oVT0KgqkQW1MhOjKK+gaSlB5628wL1cSeY3fdx7Hk5dB8Jj4DSb174g1B8/izK2HH9xn3uYDSEhK1rht47EL2H/xFnILW9b9BV+fZ5g5bxmGjfoRu7dvFGInK/xf+mLpvOniYUURrycP8fuiWWjfpScWrfgbBgYSLFs4C3lldFuGlpa8TL4USfv378e4ceNQu3ZtODg4iM+ZM2fi3LlzePv2LaytrWFkZPTJxxo8eLCwPtGxqlSpgt9++w1eXl548OABviaPbx6HvoEhmveYCBs7V7TqPQWGRqZ4cuuExvQ3zmxBreYDhBCyL15eiKV7l/fJhU5YiC9s7UvAzNJGvhiZWIhtvo+vwMbODRXrdIK1rROadvkRsVGhCA3xydY8PrxxHAYSI7TsOQG2dq5o02cyJMYmeHTzpMb0105vRu0WA1CqUmM4uJRHx0GzcPfSPiQnJSAuOgxFnUqjw4AZKFjEGSUrNoSLey0EeN+R7x8a7Ac7Z3eYW9nIF5kVStuQwCHLT++hE1DMtQyq1GoirDv/HlN+EiVuXj4JA4khegz8HnaOLug9dDyMjE1w6+pp6XW6cwXV67VAqXJV4VDMTRzn6UOpZSI6MhxOxUui38gpKGznhApV66FM+erwfnofX4MqpQyQkgrsv5iIN+/Ssfd8IhKTM1C5pIHG9GlpGcI6RJamoLfpuPY4GS72mWKoiLUuxvc2QyGrvNOcmZVxRZ0ru2Di6oS8BgmdAxdvYUKf9ijjbI8mVctiYOsG2Hn2Wpb7HLt2D/GJSWrrYxMSMWHVNiGSilhbIjeQmJiAs6eOYPDwsXBxK4WadRqgY7feOH5kn8b0p44fxNTxo2BZQGq1VeTQvn/QoHELtGjdEfYOThgyYiwi3oUjOioyB3KSN0lKSsKUKVNQrVo11KtXD+vXr88y7ZMnT9C9e3dUrFgRXbt2xaNHj7R+PnmnVfkMyCR//fp1JUtP5cqVcfToURQoUEAInn37Mgv869ev0a9fP5QvXx49evQQAkjxWLdv30ZycuYTUJEiRXDs2DGULl1a/CaX3cqVK9G7d29xs/r06QNfX99sz2eQ331hRaJzlJ0rWVoC/dQ7v/T0NAS/fIhiJavL1zm4VERaagpeB3rJLUkFCztr/C8TMythZQnwvouM9HTcv7IPhsZmKGCTvY18kK8nnEpUUcpjMbcqCPTVnMdXLx7BuVQ1+ToH18w8mlvZoueoZTA0NhVPfP7ed+H/7DacS9cQaRMTYhEd8SbLa6BtAl8+R1pqKtxKVZSvK1GmEvy8H6lZKf2ePxTbFK+DW+lK8HsmFeqm5pZ4ePsSIsLfIjkpETcvnYRT8VJim5W1DUaOXwjj9/kmcfT8yT2ULlcVX4PiRfXhG5yqtI6sScXtNFuBdv2bKLYT1hY6qFbaAN6Bmfu7OejjeWAqlu6IRV7BukENhJ+/gSv1eiKv8TwgBKlp6ajolln3K5V0xiO/QI3W9cjYOCzfdQJTB3ZW2xYcGiFcd9tnjoG9jTVyAy/9fJCamoZSZTLDFkq7V4DPsyca83fv9g2M/mEK2nXsobbtycP7QmTJKFzEDn9t2AULSyvkdsgolqGl5XNYtGiREDubNm3CjBkzRN9KXhxV4uPjMXz4cCGmqD+nPn7EiBFivTbJlzFJAwYMwIoVK3DmzBkRP1SnTh2hSN3c3LK0FpFynTVrFv78808Rr3Ty5Eno6emJY/3888/iOI0aNUKtWrXEd1dXV6VjeHh44KeffhIuPrqpdPOOHz8u/NnZBVlybOxKKK0ztSiE0FfeamkT46ORmpIkhIIMXT19IX5IGFDnGf76BXwfX8alYx5CCLlXa4XGncZAT1+CstXb4Nn9f7FhQR/o6OpBR0cXfcatFnE92UlMZKiIQ1LKo2VBvA36tDzq6enDmPL47o1S2qXjmyIqPASlKjZC2Wot5CKRxMeFw6vh/fCS2K9uy0GoXK9TtuQtMiIMZhZW0DfItKBYWFqLOKW4mCh5PBERFREGO0flMmdhZY1XAVIx3qHncKyY+z3Gf9MKurp6sCxQCFMWbFT7z4kj2uJd6GtUrFYfVWs1xdfAwlRHxCEpEhOfgaIFP/zM1r+lMWqWlSAsKh3Hr2c2hJcfaHbh5GYCPHYgrxIWFQMrMxMY6Gd2H9YWZkhKSUVUbDwKWJgppf9tx1G0q1sFrvaF1Y5V0qkoVnw/CLmJyIhwWFhYwkChXlpZFRAPyjEx0bBUETg/T5snPs+dVnZfx8XGIDY2BmlpaZg97Sf4+/nCrVQZ4b4rWOjrx7Pmxhm34+PjsXv3bqxduxZly5YVi7e3N7Zt24ZWrVoppSVDhaGhISZOnCja7alTp+LixYtCUFG8sbbIl5ak7777DosXLxYWn127dmHs2LGoX78+9u7dqzF9s2bNhCWJhA8JpfDwcBHDRHTq1Alr1qxBmTJlcPjwYXFD6Fh//fWX0jEaNGiAQYMGiWPMnj0b7969kx8ju0hJSoSevrKLQt9AgtRU9U4jJTlRfJLgUYR+p6UmIyo8GCnJCeJ395G/Cxfew+uHcWqXNGYpITYCsVFhaN13Gr6ZuhMV63TEwQ1TEBcdnr15TE6EnoHyOevrZ5HHpET5NfhY+t6jV6Df938iJNALx7cvEOvCQl6QjQY2RV3Q/0cPVGvQDQc3TseTO1KXlrYhF6CiQCIoZkHkJUX5fJOSEtXTUr7epwt7GywCv8dO+R0T56xFgYK22LBSPfbhu4lLRJqAF8/xz4al+BpIDHSQqhJXnZqWAX29DzfKp28lYcmOWEREp2NUF1MeoPyVSEhOhoGB8vO15L1gSla5sTce++C+tz+GdWiCvOTuyapeyurbp7rtiPUey4XLbdKM+UhNScH8WZO+ejxrbrUkeXl5ITU1VViFZFStWhWenp5q14zW0TZF6zqFw9y/r90wgnxpSSI6dOggloiICFy+fBlbt24VSrNUKakLQpEKFSrIv9Pot+LFi8PPz08IH4IsR7TExsYKNx6NdPv999+FZap58+YiDd0c1WOQy61x48Zay9Olo6tx6ega+W+H4hWEK0kRqsQUw6MKxS4RJIgUod+U3qqQPSYuvw4jU0tR2Io4lRHWpP1/T0TLXpNwZs9S2DqURI0m0lFe7Qf8ilW/tBExTfXaDNNaHi8c9sDFIwp5dKmANJWGiQSPgcQ4yzymakyvfE3si0tN6a1TkrHHYwJa9pqASnU7olSlRsK6RhRxLIWw1y9x899/4F5Vep+1iYGBoWg0FZGJIxI8SmklGtKmJot0ZAVct3w6ug8ch4rVpWX22wkLMXF4W+GmcylZXr6Ps5u7/H/WLpuKHgN/UOsQtE2LGoZoWUN6b4iXr9NEoLUiJJBSUj/cmr5+J20k1x+Nx9zh5nBz0Mszo9jyE4YGBkihoDIFklOlv2mkm4zE5BTM3bRfBHYrrs/tkCD61Hr5IciiSzRt2Q4Nm7QU38dNmIahfTvhuddjlHbPrJf5neTkZKWQFYK8LKqeltDQUBESo7ieRqeTcI2MjBTxxIppVb1DBQsWFJYnbZLvRBIpURr6P2nSJPGbLnj79u3FFAAtWrQQIkcVcqspQoqVTK0hISFiKgESV3TTSPyQ1alp06bo1asXrl69KhdJ+gqmZ4JMrLq62jXUVWvYC2WrtZb/vnz8b+FyUyQuOhRmVuqmXBNTcusYivSFikqH36anpSI+NlIEaBPkXlKEgsHJfZUQF4Vg/8eo2SxzaLqOri4KO5YWFihtUr1xT5SrkWlWvXSU8himlIZ+U0C1KnT+0jyGwcZOmse0tFQkxEaK9LQ+wOc+3KtmDjGmYHASmkkJcTA1LyAXSJnXwAV+T9XLjDYga09sdKQ4R3ILyoKsJRIjmJiaK6e1tkFUpLLVLjoiHFYFCiEmKgLvwl7D0TlzlJR1oSIwM7dCeGgICtrawffZA1SpmSnYKfg7NTUFCQmxMDdQDzjVJuQOu/s8s9NpXt0QFqbKdcPCRAdRcepP13q6QHlXfXj5pyIxOdM1F5eYAVNjtiV9DWytLBAZG4/UNBK70rYzPCpGCCFzk0wR8dgvEEGh70RgtiJjftsg3G+aYpRyA9YFCyE6OkqpXkZGvIPE0BCmpsquxA9hYWkp+gUK2JZhbmEJcwsLhIe9RW5Hm6PSPDw8RBiKIhTWojr1TkJCgppwkv1WFVlZpVVN91/Jd+42EicbNmwQUe+qF49GtCkqURnPnz+Xf4+OjhZD/F1cXMQ+5B8lP6ciZGkhwaR4LMVg75iYGDGnkiar1X+BRIB14WLyxdG1EgJ978mHndJngM89EZCtCokaO+fyIvBaBgU/UyNAcx75PLqERWNrKg3pfx3wVPynqbm1iPOhwG1FKIbJysZBq3kkkVKwcDH54uhWSeRJKY/e9+Doqp5HEqVkIfJXGK0W6HNfxF5RHiNCg/DPyrEiBktG8MvHIn8kkM7uW4ENiwYrHfN1gJdwv2UHjsVLQk9fH37PModOU1A1WXtUBTZZg3y9PJWug4/XfbHe1NxCuBiDA/3k6WOiIxAbG4VCtvYIe/MKfy4cL4K6ZdC0AzQlAC3ZTXxiBsIi0+XLi+BUuKgEadNotZch6lYhym7/liYoWzzTElHAXEcIJJlliclZKI5IX08XD30D5evue7+Ee3EHpXJb1sURBxb8hB2zxsgXYtrgLvi2s/Yts9qiuEsJ6Ovr4blXZh/y9MlDuJUo/VkPvtS20ui4ly8yB/HQqLaY6CjY2BZFXpgnKV1LCwVU37lzR2mhdapQjJGqyJH9Vh2RnlXaTx25/n8rkijQiwKsR40aJWKIgoKChI+SouTpApI1SRVKR7FLPj4+IoC7WLFiIkCbTHdkMaJ1O3bsEMLn8ePHWL58OR4+fCiGHCoegyxY5GIjy5OdnR1q1qyZrXl1r9YSifExOLFjnhAw9Ekip2z11vJ4HkVLU/XGvXH15Dp43T2DVy8e4uiWWajSoLuYTNLRtTL0JUY4tOkXhL32g/fDizi9ezHqtvpG7Evp7l7cDc+rB/Hujb9wv0WGB6NSnewJapZRtro0j8e2z8fbVz7ik+ZIklmbKI8U3C2jRpPeYiLJJ3fOIMjvIQ5vnoVqDbuLYfz2LuVh51wW+9dNFcd67nkBJ3cuQcP20spK0wa8fHZb7P/ubQBu/rsD968cRN3WysJJWxgaGqNOo3bYvHoeXng/xt0b53Dy4BY0a9dbHqxNI9WIanWaIT4uBjvWLRFiiD4pTql63RaiMa7XpIOYc+nZ4zsI8vfB37//AteS5YXgKu5WVkwxsGHlTLHvgzuXsXvTcrTtljmJZk5y3zsFxoY66NrISAzfp0+Jvg7uPpNamyjcxdxE+hRLDSxZotrXMxLCytFWF0PamuChbypeh7NI+hoYG0qEJYjmPSJr0bm7j7HlxCX0blZHHthNrjayLDkVLqS0ELYFLEWgd27F0MgIDZu2wppVS+Hz/CluXruEw/v+QZsO3cR2GsJP7p9PoX3nnjh+aC+uXjqHoICXWPX7AjgXd0OJUmXw/4TkvSdGcdE0qKlw4cIiRIbikhTdaiR8LCws1NKGhSl7Gei3rW3mwB1tkO/cbQTFC5GbjMx7wcHBMDExEaPbKC6Jbo4qNISfJpikgGsKGKP9ZMFgJJDs7e2xfft2LFiwQJhPq1evLo5FQkgGufQoVonEGA1JpOh8VRectqEh+H3GrsaRLTNx9+IuFHYohT7jPCAxNBHbH988JoKrZ6yTWrnK1WyLyPBXOLJlhojbodm0m3efID9Wvx/+xsl/5mPtr90gMTJF1YY9UaeVtCMtV6ONECeXj3kgmmbcdiqDgRM2wtSiYLbm0Uic1184tGkmbp/fJeKEKKhalkeaR2n/uimYvfGp+F2hVltEhgWL9BTL5F6tOVr0GC+PEegzdhWObJ2NNXN6C+FUq3k/1GoudSPSvEq9vvsdZ/f/IaxKFKfVbeRiOLllBhFqm55DfsTW1fOxePpwGJuYoWOvEahaWzrq7MchLTB4zEwhgGjb2KnLscVjHi6e3geHYiUw7pcVMDSSxmb1GvIT9m3/U8QZJSclwb1STXwzbo4oxzp6ehg9eRm2r1mAeZMGibiKZm17ycVYTkNus9UH4tGrqTHqVpAgODQNf+2PQ/L7drFKSZrB2QSjf4sSvw9fkQrFoe1MRNC3p08K9pzL3klMmQ/zY6+2mL/5AIYv+htmxkYY0akZmlaTxvm1+H4eZg7thg71vs4UE9pg0DejsebPpZg5+XuYmJqiR9/BqFW3odg2rH9nfPf9ZDRunhn6kBW16zUSI9y2rP8LUVERKFu+En6ePk/ev+RmvsYLbsuUKSP6TTJsUD9KkNWJpudRteLRdDvUz5JVna4nfd69excjR47U6jnpZKhOEcp8Np/yapOPsf1y/r4N+nr5O3+EnWX+77j/OaEcsJvfaP2zdM6s/EyjC/OR3/GzrYv8Tnk39SkVtMm+m9qz1Hap8elOq+nTpwuxM2/ePDH5M03BQ2/NIC8QWZXMzc2FZYkGUlFMcNu2bYXHh4wUNPyf3rZBhhFtke/cbQzDMAzD5E0mT54swmYGDhwopuQh44MsTIY8QjQ/EkFeIQoIJ0sTzYtEUwLQdD3aFEj51t3GMAzDMEzee8GtsbExFi5cKBZVnj17pjZ9D00GnZ2wSNICW7YovzWeYRiGYfIyHIgjhd1tDMMwDMMwGmBLEsMwDMMwSrAlSQqLJIZhGIZhlEjX4ozbeRkWSQzDMAzDKMGWJCkck8QwDMMwDKMBtiQxDMMwDKMEW5KksEhiGIZhGCZXzJOU22B3G8MwDMMwjAbYksQwDMMwjBIZPLpNwCKJYRiGYRglOCZJCrvbGIZhGIZhNMCWJIZhGIZhlODAbSkskhiGYRiGUYLdbVLY3cYwDMMwDKMBtiQxDMMwDKMEW5KksEhiGIZhGEYJjkmSwiKJYRiGYRgl2JIkhWOSGIZhGIZhNMCWpFyCmVE68jPWJonI70QnSpDfsXcwQn6m0YX5yO+cbzgZ+Z1SXie+9inkedLzd5f0ybBIYhiGYRhGCXa3SWF3G8MwDMMwjAbYksQwDMMwjBJsSZLCIolhGIZhGCV4CgAp7G5jGIZhGIbRAFuSGIZhGIZRIkOr/jYd5FVYJDEMwzAMowTHJElhdxvDMAzDMIwG2JLEMAzDMIwSPJmkFBZJDMMwDMMowe42KSySGIZhGIZRgqcAkMIxSQzDMAzDMBpgSxLDMAzDMEqwu00KiySGYRiGYZTI0Kq/TQd5FXa3MQzDMAzDaIAtSQzDMAzDKMGB21LYksQwDMMwjFpMUoaWFu2dUwaWLFmCWrVqoUaNGli0aBHSPzCh0/3799GrVy9UrlwZLVu2xO7duz/7P9mSxDAMwzBMrmfDhg04cuQIVq5cidTUVEyYMAEFCxbE0KFD1dKGhoZi2LBh6N27NxYsWIDHjx9j8uTJsLGxQaNGjT75P1kkMQzDMAyjRHou9Ldt3rwZY8eORbVq1cTv8ePHY/ny5RpF0pkzZ1CoUCH8+OOP4rezszNu3LiBw4cPs0hq0qQJXr16Jf+tr68PR0dHYXYbNGiQVv/rjz/+wM2bN7FlyxbkBK9ePsHe9bPwOtAbhR3c0HXIDDgUL5tl+kvHN+P80fVISohFhZqt0GngVEgMjcW2lOQk7N84Gw9vnYaBxBAN2wxGw7aD1Y4R9tofSyd1wvyN9+Tr5o1rhoiwYLW0LbqORvMuo7SSVzq/rWsW4M61s5AYGqFlx/5i0YS/nxe2rJ6HV/4+sHNyQf+RU+Ds6i430R7a6YFLZw4gKTEBZSvVQt9hk2BuWUBsjwh/ix3rFuPpw1uQSAxRvW4LdO03WlyT7IbyuHfDHHjePCP+r3HbQWjcTnMZDXrxFLvX/YqQQG8UcXBF96HT4eiifu9P7/dA6OsA9Pl2rnxdfGwU9m2ch6f3L8FAYoRq9dujTc9x0NX9Oh73wlZAq6q6sLEEwqKBE3fS8Tri4/u1rqaDmATg8uPMBtzEEGhZRRfOhYGEJODK0ww8fPn1GviklBQs2HIIZ28/gpHEAP1b1RfLhwgOi0D3X37H8u8HolppF7Xtszfsg00BC4zs1Ax5DV2JAerd2IdH42bj3cWbyAskJyfjzz9X4uqVK5BIJOjStRu6dOmqMe3NmzewefMmhAQHo0iRIug/YCBq1aotb3v27NmN48eOIjo6BiVLlsTIb7+Fk1Mx5HZy2xQAb968QUhICKpXry5fV7VqVdHXv337Fra2tkrp69evjzJlyqgdJzY29rP+N9/GJE2ZMgWXL18WCynKESNGCP/lgQMHkFdJTozHusUjUbxUVYybsxvOJSqJ37ReEw9unsKpfavQdchMjJiyAQE+nji6Y6l8+9EdixH04rHY1nnQdJze9yce3DipdIzI8BCsXzIKqSlJSuvHzt6FaasuyBcSX0Ym5qhWv6PW8rtr0+946fsEE371QL/hk3Bo5xrcvnpGLR0Jn+VzxqKEe2VMW7IVrqUqYvmccWI9ceHUXlw6cxDDvp+Ln+euQ+S7UGxc9au8Eftz0QQkJyVi0tx1GPHTfHjevoj92/9ETnBo21IE+j3GqF/WoduQX3By31+4f+OUhjzGY82ib+FSugp+mrcTziUrYe2iUWK9InevHMOJPernvmf9HERFhGL0jM3o+90C3Lx4EBdPbMXXwEAP6FFfF4FhGdhwOh1BYRnoXk9XrP8QNUvpoJKLepPVta4uzE2A7efTceZ+OppW1EFJe3w1ft95HE9eBsFj4jeY1L8j1hw8izO3Hn5wn3mbDyAhKVnjto3HLmD/xVvIi+gaSlB5628wL1cSeYl16/6Gj7c35s1fgFHfjcb2bdtw+fIltXQvXvhh7pw5aNG8Bf5Y+Sdat26D+fPmws/PT2w/fuwY9u3di5EjR2H58hUoXLgwpk+bhsTERPw/kZycLMSJ4kLrPgdynxGKYogsRcTr16/V0js4OKBSpUry3+Hh4Th69Chq15YKWPy/iyRzc3Phe6SlaNGi6Ny5s7g4p06pd0B5hfvXj8PAwAjt+kxAYXtXdOg/GUZGpvBUETYyLp/Ygvqt+sO9SiM4upZH16EzcevCPiQnJQhhdePcXnTsPxkOxd1RvnozNGo3FFdOb5fv/+j2Gfz+S3foG0jUjm1mYQ0LKxuxkGXi9P6/0L7vRBSw0U7vRAKHLD+9h05AMdcyqFKrCVp1Hoh/j+1US3vz8klhhekx8HvYObqg99DxMDI2wa2rp8X2h3euoHq9FihVriocirmJ4zx9KH2iff3qJfyeP8TgMTNh7+SKku5V0Kn3t7hx6YRW8vHhPErvQeeBk+BY3B0VqjdDk3ZDcPlk5j2Qcf/aCXGdO/QdL+595wGTYGhM915antPSUoWV6R+Paf9r7z7Amyq7OID/W7pooS0texQoLS17T/lQpoAsAUUUZMqUIbKnIFtBQLYMQZCNsmTIkimCsimFUvaQ1V26oN/zf2PSpgkIWNrm5vyeJ0+bm7RNepPcc8973vPCM1cBk59nBumtdz5GngI+8C1RGRWqv4NLZ39HeijmZYP4J8CeU4l4GAHsOpmIuATAv4D5XioOdsC71WxRzd8GYVHGp7e5swH5s9tg0+9P8XcoEHQH+D0wEVX90uejjYHOz/uPYeCHTVCsUD7UrlAC7RvWxOrdR575M78cOYHoGOOTEIp8HIOBs1eoICm3hxssTZZiRVD90Bo4F/GCJWEAs3PHdnTt1h0+Pr6oXv0NtGrVCls2bzK57759+1CmTBk0bdYcefPmReMmTVG6dGkcOLBf3b5r169o0bIlKlepgnz586PXp70RERGOgPPnYU2F2/Pnz1dZn+QXbjP3v7927ZrZS3S07oSQmT09/ff/FnDx9/bu3VsFVa1bt36p/4NmgyRzOOxmb2+volgWcDFoKlmyJBo0aKCyTXp+fn5qnLNKlSro3r272rZ//34VaKk3RNOmOHIk6UMvPj4eY8aMQfny5VG9enVVXPY6XA86jUJ+5WFjozuY8GvBouVxLeikyX2fPn2CG8Fn4e2vG7slL58yeJIQj9vXA9Xl6ZMEFCyaFGkX9iuv/oZ+tkDAif14u1VvFUg9z29bl6hgqWLNd1Ptud64ehFPEhLg41fGsM23WFkEXzprMpuBQQ5vS/5/8fEvi+DA0+q6S1Y3nDl+QA2rMWP0x4Ed8Crsp25zy5Ydn42aBTd3T6Pf+Tj65VKyr4L7gMFNoaLlDNu8/cvhetAZk+d4Neg0vP3KGT3HwkXL4eqlU+o6g97b1y+i35c/opBv0v9MzzmrO/48uEUFyGGP7iHg1EHkK2Saik4LeT2gskfJ8Xo+411g4O4C2GWCyjqFRqW8zQZRMYlG2++FJiK3B2CbDv3rLl6/g4QnT1HGJykwKFu0EM4G3zA7Cyc0Mgoz1mzH8Pam753b90PU0N2PX/RGvhwesDQeNSvj4b6jOFTj5Q5K6Y1ZIBYFFyumG66n4iVKIDAw0GQf1qlTFx06djL5HdFRuhdk5y5dUKtWLcN2vm+ZvY6KTvFCzoCeJiam2oUjOX/++afRhdtSOnXqFOrXr2/2cvr0aZOASP995sy6EhJzoqKi1N+6evWqCsyed1+rqUlKiUHM3r17cejQIUyYMAHjx4/HlStXsHjxYvUPW7hwIYYPH46aNWsaIlPef+XKlepNcenSJfTo0QO9evVCo0aNVDaqZ8+ehqzUiRMn1NkDh/L27NmDiRMnqt9VpEiRVH0e4aH3kTu/j9G2rG6euHvzksl9H0dFqCEy12xJqclMmezgnMUdYY/uwsbGVh047eySovIsbp7qZ6IjQ1Wm6L1PdENSl88/u46AB91DO1eoIb3UrG8JDXmALK7usLO3N2xzdfNQNTxREWGGeiIKC3mAvAWM/9eu7h64df2y+r5p666YOb4fBnRpAFvbTCowGjbpe3Wbs0tWlCxX3fBz3N/MVhUrVRmvW3jIA7iofWBvtD/jk+2Df9v3rE+izC6u6Dvm2cNnrTqOwIo5QzGkYxUkJj5F0ZJV8XbLHkgPWTLb4EGYcZAUFQvkcGVUY1oIcS8MWHvQ/DTfqNhEONnbqCAq4Ylum6uzDTLZ2sDRHnj8chn9/+xBWATcszjD3i7po9XDNQti4xMQFhmNbK5ZjO4/beVWNH6jPIrky2Xyu4p65cHMfqlbQ5mWrs9fCUsU8ugRXN3c1Am1nrt7NnVAZhbIzc3dsN3LyzhLdu3aVTXtvGGjd9T1EiVKGt2+Y/t2PHnyFMWLP7uONKNIfPbM+pfG42ryDNCzMDHBYPRZNUlfffWVGnbjUFryITiOGJnDhEiXLl1w/fp1LF26VBVvvyzNZpJGjx6teiPwwgBm8ODBaN++vcoCsfBr7NixqqiL/7ROnTohNDRUjVnqMSXn7e0NHx8frFu3TmWJGBjx/l27dlW/Kzw8XN2X48zMTPENw8JwV1fXZ+7o/yI+7jEyJTugEofCEuLjzN5X3Z4sCNLd317dPz42xsxtuuvmft+znPp9GxwdnVGqcj2kJgZfyQMksv/n8cWneHyxfC4p72uX9H95cO+2KvzuM2w6Bo37Dtk8c2LJrDFm/+7aZTNUEfi7H/XC6xYXx+dovA8y2ZnfB7r9ZWbfJ8S/0N+6d+eKKvLuM+YHdOw/HXduBmHPpkVID6w9epLiA/jJEwbxL/+7bj8EImKA+uVs1O/NlgWoXFSXQsqUDp9uj+PiYG9vfO7p8E/AFKeP4v5x9FwQTl66hk+a1k7Txyiej58n9inea/qAiSfczxIWFoYJ48epAEhfuJ3chQsXsHDhArRs2RIeHpaXGUxvPM5ySJNZKD1+z20pi7b1J7yffvopbt68qSZW+fr6vtLf1WwmidMEmaIjR0dHFWlm+udTuHnz5mp4bc2aNSq1yv4J9ISf1P/Ily+ptoZZpxIljCP/fv36Gb5nVKsfBtHXQ8XGmtYYvKzdG+djz8YFhutePqXVcFlyPJjqZ6slZ2evm5mVkGB8sE2Ij1f353Cc6W266wwoXhSLw8tUbaCyVKnJ3t5RPdbk4p/x+FiPZHLfBP5fnFRqe9GMUXivfV+UqVRT3dZj4GQM6vqOGqbzLlrKKEDatflHdBswSdUuvW6652i8D578s0/sU+xTcwGR2vcO/76v7t+5hk3Lv8aoWbvgli2HIehau/hL1G7aOdX3XUrVitmgun/S++P2I9MAhm/NhISX/90Mtn4+/BTNq9mi/7s2iI7V1STVLWuD2Ff4ff+Vo7094uON/3DcP0+MM930YuLiMX7pT6qwO/l2kf7sHRwQn+K9pg+OHJ/x2RgSEoIRw4eqafPDhg83yaoHBJzH6FEjUbFiJbRt9zEsAT87Mxr2PGIzSc4ipKlTp6okh96jR4/U8d7FxUUlNzjlf+7cuSpxoc86MeB1d0/KBlptkMQGUwULmp9mOWjQIDVE1qxZM/VPZwCVspiL/+jktUzPow++UvsFVq1Oa5Sp0sBwfe/mRQgPfWB0n4iwB8jqrqvwT47DagyUIkIfIGde3ZRi1r9wGCerew5VTRcdEaq26Q+SvC+Lg52cXV/o8fEgfTngGGo16YLUxmxPZLjx4wsPfaiCAg6RGd3XIwfCQpOygOq+IQ/hni07IsJC8OjBXRQolDS7xiN7bmTJ6o6H9+8YgqQV303Gvu3r0KXfOFSsVgdpwc0jJ6IiTJ8j90FmZ+Pn6OaRS+0fo+cY+gCu/wQ9z3PzaoAa1tMHSMR6pNjHUao1AIftXqcTlxNx4UbS+6Gqvw1cUhxrsjgBkTGv9p65EwLM/eWp+p0MkrxzAdExiUgRq6SJnO6uCI2MRsKTJ7D753PhYViECoSyOic96XPBN3Dz/iNVmJ1c72lL1PCbuRolkTY8PbMjPCxMnTTrP9sZBOkPvik9ePAAw4YOVt9PmjzFaDiOTp8+hTFfjEa58uUxaPCQdGu78bKe08g63bAfEkd8mCHivmFBffK2PrzO2mEWae/YsUNlk1LWPrFT98u07NFskPQsHKNkx05mkTgMR7/99ttzAxsGWwEBAUbb2HOpXTvzPXtSCwMdXgyPw7cM9m5eqB6nvgDw6sW/UKeZaQEc34gFvEviSuBfKFJcV19z7dJJ2GayQ14vXdEyv2dbALYUoCsX/1I/86Jv4js3LuJpQjy8iuj+j6mpQOGiyGRnh+DAM2pqP10KOIlCPsVNHh8DnW0bvjf6vwRdOIl3WnWGS1ZXlYW5fSMYefIXVvePCA9BZGQYsufUZQs3rp6P33asV9P/K1ZPuz40+Qr6q+Do2qXTamo/cX95mdkHhXxKY/emRSn2/QnUbd71X/+OWzZdMBYR9tAQEN27HQxHJ2ejuqfXJSZOd9G79RBqplry+qP8njaqv9HLcnIAWr1hi/WHniLqn1nVRfLa4JrupDHNsY7ILpMtzly+gXJFdfUPJy9dRfHC+Y32aQnvAvh50udGP9t8yFSM7NgCVUu82rCASB0ss+CJ8YULAYaaovPnzsLXt6jJ+5KzpkaNHAEbW1tMnDjZZBiNxcJjx4xBhYoVMXjwULMn1OLF8f/H0hZezGFNsN6iRalTTmAZIW0qYvEYi7VZdM2xygMHDqj6pOdNI2S26fjx42rWGqciskKexdz6rp9ppXTlt/E4OgKbfpiIv28Gqa+s3dFnm+LjYlSBr171em3w29bFair/jctnsGHJWFSp1UoNt/HCnkZsTMnbeJ/9W5egRoMXD/zY0NIjZwGzLQL+K0fHzKj+VmMsmzcBVy6dw19H92LHxh9Qt3EbQ7E2Z6oRA5voqAisXPS1Cob4lXUFbArJIKRG7aaq51LguT9x81oQFk4fgSJFS6mAi/ffsmYhGrbooGbI8ffqL6+b2gc1m6qp+9cvn8GZY7uxd8v3qNnwI0OmKC5O9xzLVKmv9v1Pyybh7s3L6mts7GOUrfr2v/6dgr6lkSufN36cMwx3bgQh6Pwx1Z+pRv02RsPEaYVZJRZVc0jM01X3lWU8+mwTi7BTZpqehcEXWwTUKm2jZsGVKWyD0oVscPRC+pwGZ3Z0UJkg9j1itmjvX+fww/YDaFO3uqGwm0NtzCx55cpudKGc2dxUobdIP05OTmrW2uxZ3+LixUAcOXwYGzasR9NmzQxDOvpyitWrV+Hu3Tvo33+A4TZeOKOKZn07EzlyZMcnn3RTNUv621OjHON144lYYipdLJmdNQZJrJCfPHmySrmxnogz16ZPn66yReZmpLEgm521Of45bdo0VQA2b948VUiWlpycs6DTgDnYsHgMft+zFnm8iqLzwHlwcHJWt588sg1rFgzHVyt0PTjKVmuER/dvYf2iMar+qFSleninje7NTE3aDlaB07zxHVQjSHbL5n1eVGT4QzWr6nVp3ak/ls+biK9GdUVm5yxo9kE3VPhnKKx/p/qqtxEDIN7WZ/gM/DB/Avb/ugH5C/qi74iZcHTS1fV80OlzbPhxDr77ZjjiYmNRvGwVdOk7TgUIJ//4TdVnbVm7UF2SW/TTX3jdmrcbhLWLvsTsLzupIbYGrXqh9D9F8KN7vIU23ceh8pvN1b7vMnC2Cqh+371O7fuug+aqbNC/YaDYdfBcFVjNGvOxer2w43aDlqnTGf1lsScSZ6ux43ZZbxvcDwPWHHiqeidRsQI2aFzZFhPXGBc6P8vPv+t+V+e32UcJ+PnIUzUEl176f/AOJi77GV2nLESWzE7o1rwu6lTUZSTq95uALzq3QtMauuytyJi6fNIVc2bPwtAhg9UQ20dt2+GNN2qo29q1/RD9PuuPevXq4/Chgyrg6f9ZX6Ofr1O3Ljp06KRqkahDe+OTT/3PZ2QZcFWSdGGTaOlhnkZsOv5iBwRL5eGs/Q6z4TGpn1HLaE4FaTv53Cf/Rmjdvjef3/dMC/wuvP5msOnNp4iufOB1GfF96vXPGNfBcj8brS6TJIQQQojnS5RUkiJBkhBCCCGMyBiTjrZz50IIIYQQr0gySUIIIYQwwsaYQoIkIYQQQqQgc7p0JEgSQgghxGtb4NaSSU2SEEIIIYQZkkkSQgghhJGnMtymSJAkhBBCCCNSk6Qjw21CCCGEEGZIJkkIIYQQRqQFgI4ESUIIIYQwIqNtOjLcJoQQQghhhmSShBBCCGFEFrjVkSBJCCGEEEakBYCODLcJIYQQQpghmSQhhBBCGJHhNh0JkoQQQghhRIIkHQmShBBCCGFEYiQdqUkSQgghhDBDMkkZhK2ttsP2gvbXoHVtBt+H1k2ZUh5aFmz/BrTO78J2aF2gfwNonU984Gv9/TLcpiNBkhBCCCGMyAK3OjLcJoQQQghhhmSShBBCCGFEFrjVkSBJCCGEEEZkuE1HhtuEEEIIIcyQTJIQQgghjMjsNh0JkoQQQghhRIIkHRluE0IIIYQwQzJJQgghhDDyVAq3FQmShBBCCGFEhtt0JEgSQgghhBFpAaAjNUlCCCGEEGZIkCSEEEIIk47bT1PpkprZra+//hpVq1ZF5cqVMWXKFDx9+vRffy4iIgL/+9//sGHDhpf+mzLcJoQQQogMX5O0ZMkSbNmyBbNmzUJCQgIGDhwIT09PdO7c+bk/99VXX+HevXuv9DclkySEEEKIDG/ZsmXo06cPKlasqLJJAwYMwIoVK577M8ePH8fvv/+OHDlyvNLflCBJCCGEECZDW4mpdEkNf//9N+7cuYNKlSoZtlWoUAG3bt16ZpYoLi4OI0eOxKhRo+Dg4PBKf1eG24QQQghhJPEFan1eFIMVXpJj0PIygcv9+/fV15w5cxq2Zc+eXX29e/eu0Xa9efPmoXjx4qhRowZelQRJQgghhHht5s+fr+qIkvv000/Ru3dvo20xMTEqY2ROdHS0+po8sNJ/nzIAo6CgIKxatQqbNm36T49dgiQhhBBCGEnNWWndunVDx44djbaZyyKdOnUKH3/8sdnfwSJtfUDk6Oho+J4yZ85sdF8O8Y0YMULVL+mzTa9KgiQhhBBCvLZmkg4vOLRWpUoVBAYGmr2NGSbOUuOwW/78+Y2G4FIWZd++fRsnTpxQv2vy5Mlq2+PHjzF69Gj88ssvWLhw4Qs/dqsMkmrXrq2KvVIqX748Vq5cCUsQHxeLDUvG4fQfv8LewRFvvdMRbzXuYPa+N68EYP2iMbhz4xJy5y+Clp1Ho4B3CZP77fppPu7fvYY2PSYYtkWEPVR/5+Lpw+rvVKzZDA1b90WmTGn/0uFZw8y5C3Dg8BE4OjjivXeb4b0Wzcze9/djx7Fk2QrcunMXeXLnQsd2H6J6lcqG2/cfPIxFy5bj4cNHKFHcH/1790QuM2Pa6cHXOwsG9vSFdyEXXLkeha9nX0Lg5Uiz983qYodtq94w2hYaHo/GHx1W37/fLB/6dPExun3lTzcwe3Ew0up1+sP8yTh+ZA8cHB3RoFk7NGje1ux9rwVfwLK5E3HzWhDyehVB++5DUcinmOEDe/vG5djzy1pERYajQtVa+OiTgXDK7Gy4/edVC7Bv+3o8eZKAitXq4KNPBqjXbFqLi4vFwjnf4PfD+9WBoWmLD9TleQLOnca3U8djzuLVRtuPHNyHH5cuwKOHD+BXvCR69BmEHDlzI73xvThnziwcPnRIPccWLVuhRYuWZu/7xx9HsWzZUty5fRu5c+dGu4/bo2rVaob9tm7dWmz7ZSvCwyNQtGhRdO/RA15eBWFJbB3sUePoBpzt+yUe7f8jvR+OJuXKlQt58+bFn3/+aQiS+D23paxH4n137txptK1du3bq0rRp05f6u1Y7u23YsGE4ePCg0WXu3LmwFJtXfI0bwWfRY8RitOw0Ejs3zMGpoztM7hcbE42FU7qjsH8FfDZhDQoVLYdFU3qo7cn9dWgrdqybbfLzK2YNRkx0JHqPXYGP+32DE4d/wd7Ni5Ee5i9eiouXLuPr8WPRp0dX/LBytQp2Ugq+chVjxk9Gg3p1MH/mNDRuUB9jJ36Fy8FX1O3nAi5g/FfTVJA1d8ZU2NvZY9zkqcgInBxt8dXokjh1Pgyd+/2FswHhmDK6lNpuTiEvZxUUNW132HBp2/NY0u0FXLBh6y2j2xevvJZmz2f19zNw9XIABn05D+26DcHG1d/h2OFdJveLjXmMb77sC9/i5TB66nL4+JfGN+P6qe20b+cGbFy1AC3b9sTwSYsQ8vAe5k8bbvj5XzYsxZ5ta9Ht8/HoP2omAs4cw8+rv0N6+GHRXFwOCsQXE77BJz37Y+2P36tg51muXb2MqRNGmZy5Xzh/BtOnjEGTFq0xZeZC2Ns74JvJY5ARLFq0EEGXLmHCxEno2etT/LhiBQ4ePGByvytXgjF+3DjUr1cf386ag4YNG2HihPEIDtYF6dt++QUb1q9H9+49MWPGTHVwGzVypKpNsRS2jg4ot3waspYsCq31SUpMpUtqadOmjWomefToUXWZOnWq0fDco0ePEBUVBTs7OxQsWNDowm3sqcTX2Muw2iApa9asKkWX/OLu7g5LwADn6N71aN5+KPIXLo5SleqiVuNOOLTDNAt28sh22Ds4oclHA5ArXxE0+3gIHDO7GAIqnnWvWzQWq+ePhGeuAkY/mxAfh6xunmjZaQRy5/eBt38FlK5SH1cC/0JaexwTg207d6Fn187w9SmCGtWr4v2W7+LnLb+Y3Hf3b/tRtkwpvNu0MfLlzYNmjRuhTKmS+O3gIXX72g0/o26tN9G44dsokD8fenXrgkchIQgLC0d6q/O/nIiLe6oyPdduRmPGd5cR/TgBtWqY7/FRsIAzbtyKxqPQeMMlNCzecHuhAs4IuhJldPvjx0/S5LkwwNm/ayM+7Pw5ChXxV9mfhu+2w+6ta0zue/TgTjg4OKF1h77IW6Cw+hlmiY4d0gVUu7euxtvN2qJqzQbI51UEXfqOwanjB3Hn1lU8ffIEOzatQOsO/VC8dCV4Fy2J5m264VpQANJaTMxj7N65BR279oG3jx+qVK+JZq3aYNsW851+d27biOEDesItWzaT2zZtWIWateqjfsNmyJffC5269UHIo4cIDwtFemIAs3PHdnTt1h0+Pr6oXv0NtGrVCls2mxbI7tu3D2XKlEHTZs3VGX/jJk1RunRpHDiwX92+a9evaNGyJSpXqYJ8+fOj16e9ERERjoDz52EJshQrguqH1sC5iBe0JiMGSZ07d0ajRo1U0Xffvn3RrFkzdOiQNILC1+Hixal7Em+1QdKzREZGYujQoahWrRpKliyJBg0aYNeupDNfPz8/zJgxQ42ddu/e3dCsqkWLFurN36RJE+zYYZrRSU23rwfi6ZMEFCpa1rCtsH95XAs6bdKi/XrQKRT2Kw8bGxt1nV+ZTbp26ZS6HhcTjTvXA9H3y5Uo6FvG6Gft7B3w0aeTkT23LvV990YQzv25F0WKJfWpSCvMDrHDaolifoZtJYsXw4WLl0yec/06tdClfTuT3xEVpcuenTpzDjWqVTVs53DcisUL4ObmivRWwi8rTp8PM9p2JiAcJf3NPzZmim7c1mVbzCmYXxdEpYfrVy/iSUICfPyTXldFi5VF8KVzJvssOPAsfIuVMXqd+vqXQVDgaXX9/t1b8PZNGiJ298iOrK7ZcPnCGdy6EYzI8FCUr/KW4fZqbzbEgDGmmdHX7WpwEBISnsCvWEnDNv/ipREUeN7s8gknjh/Fp58NQ+Nm75vcdv7MSRVk6eXKnRdzl6yBq1v6nswxC8T3YrFixQ3bipcooeo/Uj7HOnXqokPHTia/IzoqSn3t3KULatWqZdjO/c6MWlS07vaMzqNmZTzcdxSHarSG1jxNfJpql9SSKVMmdXw+duyYahDJZpL6zwzas2ePyYy55LfxOP2yrLIm6XnGjx+PK1euqGiUFfMs8Bo+fDhq1qxpKDzbu3evql3iBwILx1i5/9lnn6m1YU6ePIkhQ4aotB67gr4OESH34ZLVHXZ2SYVwzPgkxMciOjIUWVw9DNvDQ+8jV37jmhTelwEPZXZxRe8xz+9YSrPHtEfwhePIX7gE3qjfBmnt4aMQuLm6wt7e3rAtWzY3VRsRHhEBdzc3w/aCBYwzYlevXceJU6fRpOHbiIyMQkRkJJ48eYLBI8eo4Mvfzxd9e3RD9uyeSG+eHo6qDim5kJA4FC7oYvb+zBTZZbLBgqnlkMPTEafOheHbhZfxMCQO2dzt4eZqj4Z1c2NYP3/Exj3B1l/vYuVPN9PkuYQ9eoAsru6wS7bPXN09VZ1SZEQYXN2SsiehIQ+Qz8vb6Odd3T1w6/plw8+FPtIVaeqzVFGRYYiICFUBlEsWVwRdOIX1y2cjIjwMFavVxnvte6shqrQUGvIQrq5uRq9Td/ds6nXKDIlbigBn8Ehd/d/eX7cZbY+KjEBkZIR6nX458nNcC74MH79iavjOM/urdQ5OLSGPHsHV7cWeo5eXcYbl2rWr6jOyYaN31PUSJZKCSdqxfTuePHmK4sVNayYzouvzLaOGVbw6q80kscq9XLlyRhf2YWA3z7Fjx6JYsWIoVKgQOnXqhNDQUDx8+NDws61bt4a3tzd8fHxUS/Tq1aujbdu2atyT6T/evnTp0tf22OPiYlSWJzl9wMQhMqP7xsYYBVPqvvYOSEgw7SvxPO92GIoeI5eon1v+rW4qZlqKjY01+lAm/fX4+KThpZQ4hDZm4hSULO6P6lUr4/E/NS6zFyxUQ25fjhqG+PgEDB87/oUWSnzdHB1tERdv/DjiEhLhYG/+reqVPzOcnTOpwGj0lPPI7uGAKaNKwtZWl0XSB1mDvzyL5Wtv4OP3C6pi7rTA117KfaYPmMy+TlO8phngJPyzbyvXqIet67/H7RtXVJC1cvE0tf1JfDxiYqLVz6/7YRZad/wMnXuPxMlj+7F6yXSkx+s0eVCofx7mnvO/DdvR4vkz1JDbkNET1f9i4pgh6f46jeV+tXuV92IYJowfpwIgfeF2chcuXMDChQvQsmVLeHgkneiJ9JERh9vSg9Vmktg/oX79+kbbmDlq3ry5Gl5bs2aNSiufO3dO3cYzOr18+ZIOMrwPM0sMsvT4QVG4cOHX9tjt7R1NPnD1QY+Do5PpfVMERPxZ1im9jLwF/dXXD7qNw/QRrfHo/i145Eibgy05ONibfADrr+t7ZqQUEhKKQSO/UAeVUUMHwdbWVqVrqWH9eqhXWzc8M3RAP7zXtiMCAi+iRDHd80wr7d7zUhe98xfDTQIiBzsbxMSaryNq1+s4WO/LOiYaMek8Ni6thuJFXXHybBgafXgI4REJ6rbga1Fwd7PHuw3zYs1G09mdqY0zy1LuM33QY/I6dWBAZPw6jY+PUzPiqOn7XVTGaESf99XMyrfeboEChf3g5Oyi9ilnlH3YZQD8S1ZQ9/+g42eYN2242sb9nlaSB3bJn4e55/w8tra612mdtxvjzdpvq+/7DhyJzh81x8UL5+BfvBTSC/dVfMKz3ovmn2NISAhGDB+qeu8MGz7cZJ8EBJzH6FEjUbFiJbRtZ75Pjkhblh7cpBarDZI4HMbMj7mGVeyvwIwQK+lZ0M3MUHLJD8ocm2cdkr4+SY+V9K+Lq0dOREWEqqJr/VT8iNAHKvBxcjauXXHzyKluS47XXbP9e8qes9oCTh1AmSpvGz7UcuUvor5GhYekaZCU3dMTYeHhKljVBzoMghwdHZDFxXQo6sGDhxgwfJT6furEcYbhOA7Zcd+wYFuP21xds+Le/QcooZtxnmZ+3nYbew4mrTv0UUsveGQzzqjwOofPzImNNc4qsGg7PCIeOTx1v0MfIOldvRGN7J5pMy3e3TOHqhVK/joNC30IBwdHOLtkNbpvNo+cCAtJytbq7+uWTdcIztEpM3oOmoToqEiwBCGzcxb0aV8P2XPmNbQByJO/kOFnc+crqDJOEeEhcHNPu2FUD8/sCA8PM3rOoSGPVLDn4pLlhX8Ph7P4OmXBtl5WVzdkdXXFwwevtpp5avHkcwwLS/FeDFGfiy5m34sPMGzoYPX9pMlTTIYcT58+hTFfjEa58uUxaPCQNA1qhfg38mpMUbS9ZcsWfPPNNyrTVK9ePZUifl5jLWaMrl27ZjTVcPfu3di8efNre5z5CvrDNpOdofiaOOOsgDeHWYx3qZdPGVy9eNLw+Pn1ysUTKOhjXKRtTlzcYyyfOQDXg3TFs3Tzynl1lpsjT9IBKS0UKVxYHTTOX0hqNHb2XAD8fH1MnjNnwg0ZPRa2NjaYNmkcsnsmpe75oe5bxFvVIiUfkmOPlty50r5PUkRkAm7diTFczl0IR6kURdqlirmp7Sk5Z86EbSuro1yppIMOh9tYh3Tt5mM0rp8bP86tZNKD6frNtCnk9irsh0x2drgceNaw7dL5kyjsW8Jkn3n7lVRF2slfp0EBp1DET5cxWfP9DBzcswXOLllUgMTi78dRkapVQEFvP9jZ2ePGlUuG33f75hU4ZXZBlqxJtWppobC3L+zsMuHihaTZWQHnz8DH1/+lDv4MsDg77uoVXU0WcVYb661y5MyD9MRSA74XL1xImj14/txZ+PoWNXmOnAk3auQI2NjaYtLkr9TJaXJXr17F2DFjUKFiRQwdOvy1nlwKy17gNr1IkJQMC7M55MYmVDdv3sSBAwdUfdKz1oahDz/8EGfPnlWBFd/wDI6mTZumpru+tsfpmBmVajbD+kVjcf3yGZw5thv7tnyP/zVsayjWjo/T9RkpU6U+YqIjsHHZJNy9GaS+xsU+RpmquhT+87i650CpynXx0/fjVUPK4At/Ys2CUajx9odwcn7xs+LU4OTkiPq138KM2fPUjLZDR45izU8b1TR/4hR+1oPQyjXrcOfuXQz6rI/hNl4i/5lRw/5IP23eqloCXLtxA19N/xZFCheCf1FfpLe9h+4ji4sd+n5SRBVl86uTky32HNQVLTs42MLDXVf/Ef34iSrU7tOlCPx9s6JokSwYM6g4jv71SA2tHTsRAk8PB3zayRv58jihzv9y4KOWBbBi/Y00eS4cenmj1jtYNm+CCmr++n0ftm/8AXUb6xorhoU8ULVEVKl6HURHReDHRVPVbDV+jY19jMpv1FO3u3vkwKbV36nfczUoAN99MxK1GrRSQRCDppr1m2PFd1NwOfAMgi6cxtpl36JmvWZp3vTU0ckJb9ZpgAWzpyLoYgD+OHIAmzesQqOmrdTtnMKvf53+mybvtsa2Tetx+MBe3Lx+FbOnT0Khwj7w9UvjdGcKTk5Oatba7Fnf4uLFQBw5fBgbNqxH02bNDL1q9M9x9epVuHv3Dvr3H2C4Td/LhmZ9OxM5cmTHJ590Uyek+ttf9H8kXh+WKTxNpYsls0m09DDvFTtus8+CuemArEdiG/N79+6prp5cb2b69Olqxlrjxo1VC4Bly5apFgB6hw8fVg2uLl68qBpV8WdYyP0ytvxlPCzybxjoMEhix20n56yo1bgjajbSjeV/3qYEWncfh8pvvquuMxPEXkh/3wpGXq+iquN2/sKmH7Qr5w5TX5N33H6sAqzJauo/VazZFO+0+cykGPzflMl6Ef9VTEwsZsyZhwOHf4eLszPeb9kcLZs1UbfVbfwuBvbrjbfr1kbH7p/ixk3Tmhu2BtAHTlu378SPa9YhNCwMZUqWxGe9eyDHf1zjp03/pNlX/0Ux36wY0MsXhfI74/LVKHw15xIuBes6bjeskwvD+/mjRpPfDB23P+3sjeqVPVUt04GjDzFjfhAionSvp9LFXdGzYxH4FHbBo9A4rFh3Axu333nlxzZlSvmXLvJdNm8i/jyyRwUzDZu3Q/2mH6rbOjaviM69R6NGHd0+DL54FkvnTcSdm1dRoKAPPu4xFAW9dTVi7IW0asl0HPntF5WVqP5mIzV7TR8EsQ5ozdKZOLxvqzpzZQuA1h37vfTstqz2/z3LFhsTgwVzpuLoof1wdnFR3bYbN9dN8W/1Tk306jcUteo1NPoZzm5b8+MSNcU/uV+3b8aG1T8gLCwEJUqVRffeA+GZ/b9lPDPb/PfnyAzRnNmzcOjQQTXExo7bzZvrPm/eadQA/T7rj3r16qNb1y7qhDOlOnXrokOHTmjXVvdaSEn/868q0L8B0to78YE4UqddmnXc5t97nZp0S70+Y5vnp29g/19YZZCUEb1skGRpUiNIyuhSK0jKyF42SLI0qREkZXSpESRldOkRJKW11x0kNf4k9Rp6bvkuqaeWpZEBYCGEEEIYSUzFJpCWTGqShBBCCCHMkEySEEIIIYxInyQdCZKEEEIIYUSCJB0JkoQQQghhJDUXprVkUpMkhBBCCGGGZJKEEEIIYUSG23QkSBJCCCGEkUQL75SdWmS4TQghhBDCDMkkCSGEEMKIDLfpSJAkhBBCCCPScVtHhtuEEEIIIcyQTJIQQgghjDyV4TZFgiQhhBBCGJHZbToy3CaEEEIIYYZkkoQQQghhRGa36UiQJIQQQggjMrtNR4IkIYQQQhiRTJKO1CQJIYQQQpghmSQhhBBCGJHZbTo2iYmJklMTQgghhEhBhtuEEEIIIcyQIEkIIYQQwgwJkoQQQgghzJAgSQghhBDCDAmShBBCCCHMkCBJCCGEEMIMCZKEEEIIIcyQIEkIIYQQwgwJkoQQQgghzJAgSQghhBDCDAmShBAZTlhYGJ4+fQpZNUkIkZ4kSBJCZAgMiObOnYsqVaqgWrVquHXrFgYOHIhRo0YhLi4uvR+eEMIKyQK3wmIdO3bshe9bqVKl1/pYxH83a9YsbN26FYMGDcJnn32GzZs34/r16ypIqlWrFkaMGAEtiI+Px+HDh3H58mXY2trCz89PBYb8XgiRsUiQJCyWv7+/0XUbGxuVjcicOTPs7e0RHh6OTJkywdXVFUeOHIGWcCiKB9V79+7hzz//VAdab29vWLI6depg0qRJKqAtV64cNm3ahAIFCuD48ePo27cvDh06BEsXHByMrl274tGjRyhUqJDaj9euXUP+/Pnx3XffIXfu3NAKPtfAwEDExsaa3Na8eXNYOr7vli5dqvbfvHnzVFCfL18+vPPOO+n90EQqskvNXyYyXhDBwOFFBAQEwNJcuHDB8P26devUZfz48ShSpIjadvPmTZV9qFGjBrSCH8z9+vXDV199pYKiFi1aqIPQ48eP1baGDRvCUj18+BA5c+Y02c4gNzo6GlrArFjp0qXx5ZdfwsXFRW2LiIjA8OHD1W0LFiyAFnz//fcq4OW+y5Ili9Ft/Eyy9CBp586dGDp0KN5//33s27cPCQkJsLOzw5AhQ1Q93YcffpjeD1GkFmaShDYdPXrUcFm4cGHiG2+8kbhixYrE06dPJwYEBCSuX78+sVatWolLlixJtHRVq1ZVzymlwMDAxEqVKiVqRcuWLRMnTJiQGB0dnfjtt98m1q5dOzE2NjZxw4YNiQ0aNEi0ZN26dUscOXKk+r5s2bKJ169fT4yIiFDbedGCMmXKJF6+fNlke1BQkHrOWlG9enVNfK48S5MmTRI3bdpk9Folbqtbt246PzqRmmQQXMMqV65suKxduxaTJ09WZzilSpVSWSZmISZOnIhly5bB0vHs9O+//zbZfvXqVTg6OkIrLl68iPbt26shxT179qB+/fpwcHBQ+/j27duwZF988QXOnz+PN954Q2XHevbsiTfffFMVcGulHql48eJmhw3PnDljMnxsyWJiYtTwqVZxiK1s2bIm25klNPc5JCyXDLdZCdaueHp6mmznwZa1O5aOwR8Lfjt27KgONqxN4oGHAWDv3r2hFdmzZ0dQUJAafmJAwfQ+sRA4T548sGSsx+GQKevHWM/CIYzChQur4VKtFDVXr14dX3/9Nf744w+UL19eDdFwqHvLli1o0qSJKl7X+/TTT2GpmjVrhh9//BGDBw+GFvn4+ODAgQMmw2o//fSTuk1ohxRuW4n+/furrArPyJMHEePGjVNFsqyRsHSrV69WGTPOGiJfX1+0bdsWTZs2hVYw6GPtEYOGokWLqufLolEeXJkV5IHWUj0rE8YsIQvxs2XLpgrxLVm7du1e6H58zpaW4eVz09dAcgbfiRMnVODLovSUQa6lPbeUOJmge/fuKuhlRpdBIbNLZ8+eVW0s2MJCaIMESVYiMjISo0ePxvbt29WMGuJZLN/cI0eO1NSQlNYxg8SAghkWJycnnDx5Un219OGaEiVKGF6b5vD1WrduXRXQpywGFukveRbs31hylkzvwYMHWLFihTope/Lkicp6MrOUN2/e9H5oIhVJkGRlOJPmypUr6oyPb2otHWysZUouh0cZ1PLCGX4HDx5UAYaln71yqG3hwoUq26mv92C2c8KECSpDVrVqVZVFK1iwoMqAWprTp0+rQJY1ZLRr1y41tMgM2XvvvYdcuXJBSzhbka9Vfs7QL7/8oto75MiRA5aO9XKff/65YSat0C5tDPSLF5Y1a1ZVXMjibS0FSJySy/4zDIoYBCafksvaCK3ggbVmzZoqIGQw+NFHH6k6CH5oL1++HJbs22+/VQERM2R8bfLCwI+ZI56x83XLadf8H1haxoFT3lu3bq3aUhCDeNbKnTt3TmUCOSTMWjOtYPBXr149daKSfIitUaNG6rVr6f766y/1+SK0T/ayhmm9T1LKVD9nRzHjsGrVKrWtU6dO6qx15syZmulbMn36dPTp08dQAMxibRb97t27VwUTrMGyVFFRUWYPPKxnYQaUGDix3sWSfPPNN6onEvvpMFvEPjpz5szB//73P0NfpBkzZmDq1KmqnkULOJOWNTs8cdHj+3L+/PkqEF6/fj0sGT9P2BX+gw8+UMNrKcsVpMO/dkiQpGGWXhz5MqxlSi6X6dA3jNy9ezcaNGhgKFJnF2dL9vbbb2PYsGGqqWLJkiXV5AJmWji0xlokNsxkUMF9akl+++03FcTrh9N4nYEeM0t6zLpYeiYwOU4S0b82k+NrlwGipdM/B75WU+KJqaWfdIokEiRpGHvn6LHWo3Hjxppa9sAap+TyrPXo0aPqgMthxdq1a6vtHNbgMheWjAccZsM6d+6shkuJs9rYz6tbt26qvxCDJmbQLAkzR8k7iXMoihmz5DVkHAbXP2ctYDf4bdu2qf2WHGeCeXl5wdIl7/YvtE2CJCvBGgieqWsVa1WY3v/999/VWTqfb/IpuVrBoTb2g+JsmrfeekvVlnFog0MZLzO7KCPikAWzRswm6fskMSPBAJCZJAZI/GppGBRcunRJBbh8bTKTxBMYZ2dnw30YAHKdOq3g0jmsk+Pz4qQC4jpunDrP2jMt4OuTxel8LxIzn3FxcSqLxNoroQ0yu81KsF6Hb2DWCPDDWj/DRkvu37+virS1PiWXw2ocQixWrJi6zoCCa2Sx0aQWsLD3559/Vu0q2LqCM4jatGmjitQtddh7yZIlqtEpgwROMuBitqxJYtC0f/9+FRgyyOd9tIKF6JyxyIwnM2eclcj9qIVgkJMH2DolNDTU5DbWQXKfCm2QIMlKcFiGvXWeVcht6WPo+mwDO4hrnRbPYLn0CAOjjRs34saNGyro4/RxFjNb6nNKGSjx+fH9x6aL+gVeefLChqAsAOYit1rpLK71KfKsrWJxdocOHVTgx1o5BkwcLuZz5xCx0AYJkqwEl0F40folS8Q1vlj7wenxrL3idS02yNTaGSxnOTF4YIaFdTsM5rkeHQ9AZcqUUUGTlmrKUrp7967K6np4eEBL2NOKHfCZPdIiTixg3ycOpbKGjoEST9JYFzllyhSj1gfCsklNkpXQB0EcvuAMKR54mH3QSq8k1nlwGQQOZbBGh/2ReMBlFoLDGiwA1gJmVjgT6llnsJaG2RMeSLnPtLR8zPPwvJS1OhwW5uuS2ZYqVapAS7Q+RZ6ZTs62JA7rs5CbQRIL1vW9sIQ2SCbJSjAgGjt2LDZs2KCu79ixQx2Y+EafNm0a3NzcoCUs8uVzZBNC1kNwRpgWaO0Mlq/HrVu3qoJ7HnhYjM7nw4aSXABWa5kkFi9zSQ4Ol3I2Ij9+WZzOAy37eXGdMy143hI5Wpgiz4kinBjCz1TWXPG9xx5m/MzRX4Q2aGMAXPwrvolZSMkp8fqzOnb8DQkJscglHp4lOjpaBREsjGURN6fKv+iiopZ6BkuWegbL2o1FixapII/BA7Oc/MrhGq7jxuDW0ppHPg/XT+QwIp8vA0S+H5kFZbaFw6hawdflsy6WHiAlz4By9iyDeu7TVq1aqZOywYMHp/fDE6lIMklWgrU6s2fPVlPGy5Urh02bNqlZJsy4sDO1pWdaeLDhUNvhw4fVLC8Os3HNNktf9NUaz2BZp8Mu4gx2uZivu7u7WoiZz93SsREms2P69cz0OPTGgPHUqVPQCq1NMODsSrYT4YmKXkxMjFpcWl/KwBNQrQztCx2pSbISXPLB3Mwvnq3rP8QsGZd+YIdfziLiWZ1W8Qx2/Pjx6gyWgQODIp7BsucOF3/VAjY87dKli7pwKEofMGkhSOJrk80kUwZJXAtM39LBGiYYWGKQxNYUKbOaXB6IQS9POLVS3ymMSSbJSgwcOFANRbEOiYXMzCTRgAEDVB0EC4ItGV/GL7pOndbIGWzGlrzJJ7NknM3HIRpmlTjl/+LFiyoQ5Lp7bBSqBVqcIs+sNAvuPT09DduSZ+WFNkkmyUpwyQc2rOMsN2aPWrZsqRYNZYGspdZCfPzxx+oAxPR3+/btNbuOHQ+qL0rff0dkHCmHsnlg5TAUFyVOnmFidlAr2OuKi9lyggEnG7DRKwNDBoUcIrbEIElYJwmSrATXhuJyAPzwYv0D6wWY8rfkZm8M+PTZE0vv8/Q8nPX0IphJkyAp4/nhhx9gbWSKvNAKCZKsxLFjxwzfu7i4GJa34Ow2BhqsE7C05Ts4Cyp5bQCnjGsRFwV9ntjYWE02ztQKZgJZg8Omkf+WFdRKkMtmrmPGjFETDNgDitmjWrVqqRq65Iv9Whou2pu89ohZ+V9//dWkGahW9qOQmiSrwS7GPIPjm5o9kbjbuewDsw+88DprJJhtssQPMT521gqweJsdt/WLamoNu4pz6NTX19cQJPKAxACRByRmDEXGwqam7CyeLVs29f2z8H24e/duaKVOjhMMGCBxggFrItkPSz/B4Hn/h4zqRR+zlvajkCDJanDq6r59+zBp0iTDzBoOvbFOiW9+fpBxHamXGd7JaLP3WOPBNgDsQcPMGItHeQbv5+cHrejbt6+qZ+FZun6olFOqJ0yYoDKBLMwXGdODBw9UOwM2NyW232ATTQb3PIlhAGHpONOLmRVmp+vUqaNOWPRkgoGwRBIkWQk25/v+++9N+gaxVoAzUPhhzVolLiOQfGjOErF3Cdcw4zRkntHlyZNHzR7SgooVK6o1sVLWkl26dEn1cfm3NfpE2uOs0v79+6umkXwdct+xrxfbObDdAQMH9g/SNz+1VEuXLlXDatWqVVOBIGeCdezYUT13ISyVdNy2Iqw/MrcteZ8kLUyj55RqNuXjmTpn07CBplawcR2nkafE+jJ9hkJkLMzM3rp1C8uXL1eFywya2OWeQ8Ss0WGdC2eZWnqfq1WrVqkhtoULF2LevHmqrQg7UMt5uLBk8qlqJdhwkO3yuegkp+Tyg4tBxIwZM/Duu++qYIkf0pY6S4wZFA61MXvEuh0WifK5stM4C2a1glOnOUTK56avu2I2kPuRQ6Yi4+HrksOhFSpUUNcPHjyohoe5XI5+6In7tVu3brBkHL5nFkmPw/ic4Xbv3j2LzpAJ6yZBkpX4/PPP1aw2dqbmhxaxQJsN7LhQKpfzYCaCRcGWiN2ZGRCxGR8DJHPdxbVSk8QAl7Vl+m7GLAjmAbdr167p/fCEGewRxH5BenyvZcqUSWWP9LiUjn7KvKViW5Hk2Ux+rx9KFMJSSZBkJTiM1qNHD3Vh1ogfYMlnQrELNy+Wiuu08bklPxhpScqCWNaQcYiN12VGW8bGLAqzLCysZ4DL2iQ2j+QsU70TJ06o2jkhRMYiQZIV4cKo7OprblV1S+/rwQLtXr16QYuSF8TybJ1rmLHuSgpiLQOHQVmrwywgg9s7d+6ozK4eh0unTZuGpk2bwtJJHyGhNTK7zUqwmPLrr79WZ6/6ZpJa6usxZ84cdTbOmXo8Y0/ZXNHSGmUmx1YGrFfRH1hY48JA6fjx45ootNc6Bras92MjSe4vLqfD9cuILRuWLFmCt956S9WVWXJTUOkjJLRIgiQrwY7UrD3iRYtStjbQBw/6hW/ZS8hSsdCeBxV98SsPupwZxb5QUhBr2QIDA9Xs0uLFi6f3QxFCmCHDbVaCS1ewYZ1WafnMVApitUtLjU6F0CIJkqxEkyZNVLM6zv7S4hBNvnz50vshCCGE0BgJkqwElwRYt26d6vibP39+k6UBli1bBksfbnte8GfJw20kBbFCCJH2pCbJSsyaNeu5t+sXS7VUKZfjYJ3H9evXVVFsv3791MK3lkoKYoUQIn1IkCQ07ejRo5g4caKaWSSEEEK8DFm7zYps2rRJLX/ARVLZ3I69WxYsWAAt43BUcHBwej8MIYQQFkhqkqwEi7bZS6h79+6GhTQ5tZxrSnGWlKUPt5nLFHF9LNZhlS1bNl0ekxBCCMsmw21Wgg0JucAtm9aVK1dOZZUKFCiglkjgem38aslS1u2wPofF6aVKlVI1STL7TQghxMuSTJKVuH37NooUKWKynYGSfqFUS/XgwQPVhVrfS+jcuXNq+QdPT0/VG8rZ2Tm9H6IQQggLJDVJVoILaqYckmIScfHixap7syXicBqHD7kwL9elo59++gnvvfceVqxYgfnz56v+UHfv3k3vhyqEEMICyXCbleCCqF27dlXZFS6oycVSr169isePH6t13YoVKwZLM2nSJBw6dAhffPEFypcvr54LAyZfX1/88MMParht9OjRKpjiunVCCCHEy5DhNitRtGhR7NixA5s3b8bly5dVH6E6deqolcdTLnhrKTjExsLzChUqqOsHDx5UAVG7du0MzTI5m4+LwwohhBAvS4IkK8L1vhg02Nra4t69e/jzzz/V18KFC8MS3b9/H15eXobrhw8fRqZMmVCjRg3DtuzZs6sMkxBCCPGypCbJSjAg4lAUO1MzMGKwxFltrNnhkheWKFeuXKrfE3HUmDP0WHvl5uZmuM+JEyeQJ0+edHyUQgghLJUESVaCXacbNWqkgog1a9aorBLreb788kvMnDkTlqhZs2aqISaX4uCw2507d/Dhhx8abmft1bRp0yx6SRIhhBDpR4IkKyrcbt++PTJnzow9e/aoqfEODg6oXLmyag9giXr06KEK0IcNG6Zqrfr06YPGjRur2yZPnqwWe2UtFu8nhBBCvCypSbISrM0JCgpCdHQ0zp8/jyFDhhjqeCx1OIp9kYYOHaouKTFA4lBi8eLF0+WxCSGEsHwSJFmJDh06oFevXqpom12omUGaN28eZs2apYbitMbPzy+9H4IQQggLJ32SrAgzSBxa4+wvJycnnDx5Un319/dP74cmhBBCZDiSSbIi+fPnV0uTsGibRc3Hjx9HiRIl0vthCSGEEBmSFG5biV27dqFmzZqqFQCX8Pjoo4/UEh49e/bE8uXL0/vhCSGEEBmOBElWYvr06Wr2V/Xq1bF27VpVrL1161Y1RZ7rtwkhhBDCmARJVuL69eto2LCh+p59herVq6e+5zpnjx49SudHJ4QQQmQ8UpNkJfLmzYujR4+qLtVXrlxB7dq11Xb2FypUqFB6PzwhhBAiw5EgyUpwqG3QoEFqYdu33npLtQFgw8VVq1apNgBCCCGEMCYtAKwIh9X+/vtvFCtWTF0PDg6Gq6urajQphBBCCGMSJFmRhIQEPHz4UGWTiLs+Li4OAQEBal03IYQQQiSRIMmKWgCMHDkSoaGhJrflyJED+/fvT5fHJYQQQmRUMrvNSkydOlXNaOO0fw6xsRaJy5Lky5cP/fr1S++HJ4QQQmQ4UrhtJW7cuIH58+fDy8sLJUuWxP3791G3bl21ltuUKVPQokWL9H6IQgghRIYimSQrwezR48eP1feFCxdWy5KQt7c3bt68mc6PTgghhMh4JEiyEm+++SbGjBmDoKAgVKlSBRs3bsS5c+ewevVq5MyZM70fnhBCCJHhSJBkJYYPH46CBQvi7NmzapitTJkyaNWqFVasWIHBgwen98MTQgghMhyZ3aZxzBj9+uuvsLe3R506ddC4cWPDbZGRkXB0dFS3CSGEEMKYZJI0bOnSpRg2bBhiYmJUPdLQoUPVgrZ6WbJkkQBJCCGEeAbJJGkYF7Tt1q0bmjdvrq7v3LlTBUrHjx+HjY1Nej88IYQQIkOTTJLGp/1Xq1bNcJ2L2jKjdO/evXR9XEIIIYQlkCBJ48uQ2NkltcLi96xB4lIkQgghhHg+CZKEEEIIIcyQjtsat23bNlWgrff06VM1283Dw8Pofvq6JSGEEELoSOG2hrEG6UWwiHv37t2v/fEIIYQQlkSCJCGEEEIIM6QmSQghhBDCDAmShBBCCCHMkCBJCCGEEMIMCZKEEEIIIcyQIEkIIYQQwgwJkoQQQgghzJAgSQghhBACpv4PI/w+9WXYXwQAAAAASUVORK5CYII="
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data",
|
||
"jetTransient": {
|
||
"display_id": null
|
||
}
|
||
}
|
||
],
|
||
"execution_count": 12
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "5f9853b3",
|
||
"metadata": {},
|
||
"source": [
|
||
"**Observations :**\n",
|
||
"* Corrélation entre SibSp et Parch, ce qui est logique étant donné qu'ils sont tous les deux relatifs à la notion de famille\n",
|
||
"* Corrélation entre la survie et la classe des passagers : très intéressant pour l'apprentissage\n",
|
||
"* Corrélation entre l'âge et la classe des passagers"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "ae39b80c",
|
||
"metadata": {},
|
||
"source": [
|
||
"9. En vous basant sur vos observations de la visualisation précédente, confirmez vos impressions en proposant deux visualisations. Par exemple, si vous avez observé une corrélation entre un attribut A et un attribut B, mettez en valeur le fait que les mêmes valeurs de A sont souvent trouvées avec les mêmes valeurs de B. "
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"id": "987a4b5f",
|
||
"metadata": {
|
||
"ExecuteTime": {
|
||
"end_time": "2025-09-16T10:06:45.859230Z",
|
||
"start_time": "2025-09-16T10:06:45.727921Z"
|
||
}
|
||
},
|
||
"source": [
|
||
"# Proposition 1 : survie et classe des passagers\n",
|
||
"sns.countplot(data=titanic, x='Pclass',hue='Survived')"
|
||
],
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"text/plain": [
|
||
"<Axes: xlabel='Pclass', ylabel='count'>"
|
||
]
|
||
},
|
||
"execution_count": 13,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
},
|
||
{
|
||
"data": {
|
||
"text/plain": [
|
||
"<Figure size 640x480 with 1 Axes>"
|
||
],
|
||
"image/png": "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"
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data",
|
||
"jetTransient": {
|
||
"display_id": null
|
||
}
|
||
}
|
||
],
|
||
"execution_count": 13
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "ae68615d",
|
||
"metadata": {},
|
||
"source": [
|
||
"**Observation :** les passagers de classe plus élevée ont eu plus de chance de survie."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"id": "9d66606a",
|
||
"metadata": {
|
||
"ExecuteTime": {
|
||
"end_time": "2025-09-16T10:06:46.041959Z",
|
||
"start_time": "2025-09-16T10:06:45.865680Z"
|
||
}
|
||
},
|
||
"source": [
|
||
"# Proposition 2 : âge et classe des passagers\n",
|
||
"sns.histplot(\n",
|
||
" data=titanic, x='Age', hue='Pclass', multiple='dodge',\n",
|
||
" bins=range(1, 110, 10))"
|
||
],
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"text/plain": [
|
||
"<Axes: xlabel='Age', ylabel='Count'>"
|
||
]
|
||
},
|
||
"execution_count": 14,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
},
|
||
{
|
||
"data": {
|
||
"text/plain": [
|
||
"<Figure size 640x480 with 1 Axes>"
|
||
],
|
||
"image/png": "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"
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data",
|
||
"jetTransient": {
|
||
"display_id": null
|
||
}
|
||
}
|
||
],
|
||
"execution_count": 14
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "57cd6075",
|
||
"metadata": {},
|
||
"source": [
|
||
"**Observation :** les passagers plus âgées sont souvent dans des classes plus élevées."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "67adc89f",
|
||
"metadata": {},
|
||
"source": [
|
||
"10. Mettez-vous à présent dans le rôle d'un data analyst face à un nouveau jeu de données : quelles autres visualisations vous paraissent nécessaires ? Proposez au moins trois visualisations qui vous semblent pertinentes. Gardez en tête que l'objectif sur ce jeu de données sera de réussir à prédire si un passager à survécu ou non."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "02d7f820",
|
||
"metadata": {},
|
||
"source": [
|
||
"**Suggestions de visualisations :**\n",
|
||
"* Survie par classe d'âge\n",
|
||
"* Survie et nombre de membres de la famille\n",
|
||
"* Age et prix du ticket"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "081aafe5",
|
||
"metadata": {},
|
||
"source": [
|
||
"## Nettoyage des données\n",
|
||
"1. Pour chaque colonne, comptez le nombre de valeurs nulles."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"id": "ba42c62a",
|
||
"metadata": {
|
||
"ExecuteTime": {
|
||
"end_time": "2025-09-16T10:06:46.062200Z",
|
||
"start_time": "2025-09-16T10:06:46.057233Z"
|
||
}
|
||
},
|
||
"source": [
|
||
"titanic.isna().sum()"
|
||
],
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"text/plain": [
|
||
"PassengerId 0\n",
|
||
"Survived 0\n",
|
||
"Pclass 0\n",
|
||
"Name 0\n",
|
||
"Sex 0\n",
|
||
"Age 177\n",
|
||
"SibSp 0\n",
|
||
"Parch 0\n",
|
||
"Ticket 0\n",
|
||
"Fare 0\n",
|
||
"Cabin 687\n",
|
||
"Embarked 2\n",
|
||
"dtype: int64"
|
||
]
|
||
},
|
||
"execution_count": 15,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"execution_count": 15
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "e5bc166d",
|
||
"metadata": {},
|
||
"source": [
|
||
"### Traitement du port d'embarquement\n",
|
||
"\n",
|
||
"2. Les valeurs manquantes du port d'embarquement sont très peu nombreuses. De plus, s'agissant d'un attribut discret, nous pouvons considérer l'information de valeur nulle comme une valeur possible supplémentaire. Commencer par affichez les lignes pour lesquelles le port d'embarquement n'est pas renseigné."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"id": "5219c39a",
|
||
"metadata": {
|
||
"ExecuteTime": {
|
||
"end_time": "2025-09-16T10:06:46.092543Z",
|
||
"start_time": "2025-09-16T10:06:46.083739Z"
|
||
}
|
||
},
|
||
"source": [
|
||
"titanic[titanic['Embarked'].isna()]"
|
||
],
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"text/plain": [
|
||
" PassengerId Survived Pclass Name \\\n",
|
||
"61 62 1 1 Icard, Miss. Amelie \n",
|
||
"829 830 1 1 Stone, Mrs. George Nelson (Martha Evelyn) \n",
|
||
"\n",
|
||
" Sex Age SibSp Parch Ticket Fare Cabin Embarked \n",
|
||
"61 female 38.0 0 0 113572 80.0 B28 NaN \n",
|
||
"829 female 62.0 0 0 113572 80.0 B28 NaN "
|
||
],
|
||
"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>PassengerId</th>\n",
|
||
" <th>Survived</th>\n",
|
||
" <th>Pclass</th>\n",
|
||
" <th>Name</th>\n",
|
||
" <th>Sex</th>\n",
|
||
" <th>Age</th>\n",
|
||
" <th>SibSp</th>\n",
|
||
" <th>Parch</th>\n",
|
||
" <th>Ticket</th>\n",
|
||
" <th>Fare</th>\n",
|
||
" <th>Cabin</th>\n",
|
||
" <th>Embarked</th>\n",
|
||
" </tr>\n",
|
||
" </thead>\n",
|
||
" <tbody>\n",
|
||
" <tr>\n",
|
||
" <th>61</th>\n",
|
||
" <td>62</td>\n",
|
||
" <td>1</td>\n",
|
||
" <td>1</td>\n",
|
||
" <td>Icard, Miss. Amelie</td>\n",
|
||
" <td>female</td>\n",
|
||
" <td>38.0</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>113572</td>\n",
|
||
" <td>80.0</td>\n",
|
||
" <td>B28</td>\n",
|
||
" <td>NaN</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>829</th>\n",
|
||
" <td>830</td>\n",
|
||
" <td>1</td>\n",
|
||
" <td>1</td>\n",
|
||
" <td>Stone, Mrs. George Nelson (Martha Evelyn)</td>\n",
|
||
" <td>female</td>\n",
|
||
" <td>62.0</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>113572</td>\n",
|
||
" <td>80.0</td>\n",
|
||
" <td>B28</td>\n",
|
||
" <td>NaN</td>\n",
|
||
" </tr>\n",
|
||
" </tbody>\n",
|
||
"</table>\n",
|
||
"</div>"
|
||
]
|
||
},
|
||
"execution_count": 16,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"execution_count": 16
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "363787e8",
|
||
"metadata": {},
|
||
"source": [
|
||
"3. Remplacez ces valeurs nulles par la valeur 'U' (pour unknown). Vérifiez vos résultats en réaffichant les lignes obtenues ci-dessus :"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"id": "2be8a958",
|
||
"metadata": {
|
||
"ExecuteTime": {
|
||
"end_time": "2025-09-16T10:06:46.297992Z",
|
||
"start_time": "2025-09-16T10:06:46.287441Z"
|
||
}
|
||
},
|
||
"source": [
|
||
"titanic['Embarked'] = titanic['Embarked'].replace(np.nan, 'U')\n",
|
||
"titanic[(titanic['PassengerId']==62) | (titanic['PassengerId']==830)]"
|
||
],
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"text/plain": [
|
||
" PassengerId Survived Pclass Name \\\n",
|
||
"61 62 1 1 Icard, Miss. Amelie \n",
|
||
"829 830 1 1 Stone, Mrs. George Nelson (Martha Evelyn) \n",
|
||
"\n",
|
||
" Sex Age SibSp Parch Ticket Fare Cabin Embarked \n",
|
||
"61 female 38.0 0 0 113572 80.0 B28 U \n",
|
||
"829 female 62.0 0 0 113572 80.0 B28 U "
|
||
],
|
||
"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>PassengerId</th>\n",
|
||
" <th>Survived</th>\n",
|
||
" <th>Pclass</th>\n",
|
||
" <th>Name</th>\n",
|
||
" <th>Sex</th>\n",
|
||
" <th>Age</th>\n",
|
||
" <th>SibSp</th>\n",
|
||
" <th>Parch</th>\n",
|
||
" <th>Ticket</th>\n",
|
||
" <th>Fare</th>\n",
|
||
" <th>Cabin</th>\n",
|
||
" <th>Embarked</th>\n",
|
||
" </tr>\n",
|
||
" </thead>\n",
|
||
" <tbody>\n",
|
||
" <tr>\n",
|
||
" <th>61</th>\n",
|
||
" <td>62</td>\n",
|
||
" <td>1</td>\n",
|
||
" <td>1</td>\n",
|
||
" <td>Icard, Miss. Amelie</td>\n",
|
||
" <td>female</td>\n",
|
||
" <td>38.0</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>113572</td>\n",
|
||
" <td>80.0</td>\n",
|
||
" <td>B28</td>\n",
|
||
" <td>U</td>\n",
|
||
" </tr>\n",
|
||
" <tr>\n",
|
||
" <th>829</th>\n",
|
||
" <td>830</td>\n",
|
||
" <td>1</td>\n",
|
||
" <td>1</td>\n",
|
||
" <td>Stone, Mrs. George Nelson (Martha Evelyn)</td>\n",
|
||
" <td>female</td>\n",
|
||
" <td>62.0</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>0</td>\n",
|
||
" <td>113572</td>\n",
|
||
" <td>80.0</td>\n",
|
||
" <td>B28</td>\n",
|
||
" <td>U</td>\n",
|
||
" </tr>\n",
|
||
" </tbody>\n",
|
||
"</table>\n",
|
||
"</div>"
|
||
]
|
||
},
|
||
"execution_count": 17,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
}
|
||
],
|
||
"execution_count": 17
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "a5e22413",
|
||
"metadata": {},
|
||
"source": [
|
||
"### Traitement du numéro de cabine\n",
|
||
"\n",
|
||
"4. Le numéro de cabine est l'attribut qui contient le plus de valeurs nulles. Cet attribut n'a pas de lien évident avec la survie des passagers. Supprimer cette colonne dans votre dataframe."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"id": "aef1705c",
|
||
"metadata": {
|
||
"ExecuteTime": {
|
||
"end_time": "2025-09-16T10:06:46.473560Z",
|
||
"start_time": "2025-09-16T10:06:46.465828Z"
|
||
}
|
||
},
|
||
"source": [
|
||
"titanic = titanic.drop('Cabin', axis=1)\n",
|
||
"titanic.info()"
|
||
],
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"<class 'pandas.core.frame.DataFrame'>\n",
|
||
"RangeIndex: 891 entries, 0 to 890\n",
|
||
"Data columns (total 11 columns):\n",
|
||
" # Column Non-Null Count Dtype \n",
|
||
"--- ------ -------------- ----- \n",
|
||
" 0 PassengerId 891 non-null int64 \n",
|
||
" 1 Survived 891 non-null int64 \n",
|
||
" 2 Pclass 891 non-null int64 \n",
|
||
" 3 Name 891 non-null object \n",
|
||
" 4 Sex 891 non-null object \n",
|
||
" 5 Age 714 non-null float64\n",
|
||
" 6 SibSp 891 non-null int64 \n",
|
||
" 7 Parch 891 non-null int64 \n",
|
||
" 8 Ticket 891 non-null object \n",
|
||
" 9 Fare 891 non-null float64\n",
|
||
" 10 Embarked 891 non-null object \n",
|
||
"dtypes: float64(2), int64(5), object(4)\n",
|
||
"memory usage: 76.7+ KB\n"
|
||
]
|
||
}
|
||
],
|
||
"execution_count": 18
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "112f1631",
|
||
"metadata": {},
|
||
"source": [
|
||
"### Traitement de l'âge\n",
|
||
"5. L'âge est un attribut plus délicat à traiter : il contient un nombre conséquent de valeur nulle, mais il est très pertinent à prendre en compte pour la prédiction de la survie d'un passager, ces deux informations étant assez fortement corrélées. Il existe plein de stratégies pour remplacer ces valeurs manquantes :\n",
|
||
"* Mettre une valeur aléatoire '(tirée entre le min et le max)\n",
|
||
"* Remplacer par la moyenne\n",
|
||
"* Remplacer par une valeur déterminée en fonction des autres paramètres (classe, age, etc)\n",
|
||
"\n",
|
||
"Commencer par calculer pour chaque genre et pour chaque classe, l'âge moyen (6 valeurs à obtenir au total)."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"id": "2b2a06ff",
|
||
"metadata": {
|
||
"ExecuteTime": {
|
||
"end_time": "2025-09-16T10:06:46.635521Z",
|
||
"start_time": "2025-09-16T10:06:46.627947Z"
|
||
}
|
||
},
|
||
"source": [
|
||
"age_avg_pclass_sex = titanic.groupby(['Pclass', 'Sex'], as_index=False)['Age'].mean()\n",
|
||
"print(age_avg_pclass_sex)"
|
||
],
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
" Pclass Sex Age\n",
|
||
"0 1 female 34.611765\n",
|
||
"1 1 male 41.281386\n",
|
||
"2 2 female 28.722973\n",
|
||
"3 2 male 30.740707\n",
|
||
"4 3 female 21.750000\n",
|
||
"5 3 male 26.507589\n"
|
||
]
|
||
}
|
||
],
|
||
"execution_count": 19
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "e42013d7",
|
||
"metadata": {},
|
||
"source": [
|
||
"6. Pour chaque ligne du jeu de données, si l'âge est manquant, remplacez la valeur nulle par une des valeurs calculées ci-dessus. basez-vous sur le genre et la classe du passager pour choisir la bonne valeur. "
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"id": "3daa4cc6",
|
||
"metadata": {
|
||
"ExecuteTime": {
|
||
"end_time": "2025-09-16T10:06:46.716203Z",
|
||
"start_time": "2025-09-16T10:06:46.710056Z"
|
||
}
|
||
},
|
||
"source": [
|
||
"#La méthode transform() est conçue pour conserver l'index d'origine, ce qui permet d’assigner directement le résultat à une colonne du DataFrame.\n",
|
||
"#apply() modifie l’index → pas idéal pour affecter une colonne.\n",
|
||
"#transform() conserve l’index → parfait pour des remplacements ligne à ligne.\n",
|
||
"\n",
|
||
"titanic['Age'] = titanic['Age'].groupby([titanic['Pclass'], titanic['Sex']]).transform(lambda x: x.fillna(x.mean()))\n",
|
||
"\n",
|
||
"#groupby(['Pclass', 'Sex'])['Age'] : groupe les âges selon la classe et le sexe.\n",
|
||
"#.transform('mean') : remplace chaque ligne du groupe par la moyenne du groupe, en gardant le même index.\n",
|
||
"#.fillna(...) : remplit uniquement les NaN avec la moyenne du groupe correspondant.\n",
|
||
"\n",
|
||
"# Vérifier que les valeurs manquantes ont bien été remplacées\n",
|
||
"print(titanic['Age'].isnull().sum()) # Nombre de NaN restants (devrait être 0 ou très réduit)"
|
||
],
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"0\n"
|
||
]
|
||
}
|
||
],
|
||
"execution_count": 20
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "cf24aa8b",
|
||
"metadata": {},
|
||
"source": [
|
||
"## Préparer les données\n",
|
||
"\n",
|
||
"Nous entrons dans la dernière phase de traitement des données : nous allons terminer des les mettre en forme, pour qu'elles soient prêtes à être manipulées dans un processus d'apprentissage."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "f2ab2605",
|
||
"metadata": {},
|
||
"source": [
|
||
"# Encodage des données catégorielles\n",
|
||
"\n",
|
||
"1. Réaffichez les infos sur le jeu de données. Vous devez avoir 11 colonnes, toutes remplies avec 891 valeurs."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"id": "3ac1186f",
|
||
"metadata": {
|
||
"ExecuteTime": {
|
||
"end_time": "2025-09-16T10:06:46.803405Z",
|
||
"start_time": "2025-09-16T10:06:46.796857Z"
|
||
}
|
||
},
|
||
"source": [
|
||
"titanic.info()"
|
||
],
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"<class 'pandas.core.frame.DataFrame'>\n",
|
||
"RangeIndex: 891 entries, 0 to 890\n",
|
||
"Data columns (total 11 columns):\n",
|
||
" # Column Non-Null Count Dtype \n",
|
||
"--- ------ -------------- ----- \n",
|
||
" 0 PassengerId 891 non-null int64 \n",
|
||
" 1 Survived 891 non-null int64 \n",
|
||
" 2 Pclass 891 non-null int64 \n",
|
||
" 3 Name 891 non-null object \n",
|
||
" 4 Sex 891 non-null object \n",
|
||
" 5 Age 891 non-null float64\n",
|
||
" 6 SibSp 891 non-null int64 \n",
|
||
" 7 Parch 891 non-null int64 \n",
|
||
" 8 Ticket 891 non-null object \n",
|
||
" 9 Fare 891 non-null float64\n",
|
||
" 10 Embarked 891 non-null object \n",
|
||
"dtypes: float64(2), int64(5), object(4)\n",
|
||
"memory usage: 76.7+ KB\n"
|
||
]
|
||
}
|
||
],
|
||
"execution_count": 21
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "a62f568e",
|
||
"metadata": {},
|
||
"source": [
|
||
"2. Trois colonnes sont liées à l'identification unique d'un passager, et ne sont pas pertinentes pour la prédiction de la survie. Supprimez ces trois colonnes de votre jeu de données."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"id": "a4b7d99c",
|
||
"metadata": {
|
||
"ExecuteTime": {
|
||
"end_time": "2025-09-16T10:06:46.865523Z",
|
||
"start_time": "2025-09-16T10:06:46.858478Z"
|
||
}
|
||
},
|
||
"source": [
|
||
"titanic = titanic.drop(['PassengerId','Name', 'Ticket'], axis=1)\n",
|
||
"titanic.info()"
|
||
],
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"<class 'pandas.core.frame.DataFrame'>\n",
|
||
"RangeIndex: 891 entries, 0 to 890\n",
|
||
"Data columns (total 8 columns):\n",
|
||
" # Column Non-Null Count Dtype \n",
|
||
"--- ------ -------------- ----- \n",
|
||
" 0 Survived 891 non-null int64 \n",
|
||
" 1 Pclass 891 non-null int64 \n",
|
||
" 2 Sex 891 non-null object \n",
|
||
" 3 Age 891 non-null float64\n",
|
||
" 4 SibSp 891 non-null int64 \n",
|
||
" 5 Parch 891 non-null int64 \n",
|
||
" 6 Fare 891 non-null float64\n",
|
||
" 7 Embarked 891 non-null object \n",
|
||
"dtypes: float64(2), int64(4), object(2)\n",
|
||
"memory usage: 55.8+ KB\n"
|
||
]
|
||
}
|
||
],
|
||
"execution_count": 22
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "9126e8f9",
|
||
"metadata": {},
|
||
"source": [
|
||
"3. Reproduisez une manipulation vue au module 3 : combinez deux colonnes relatives aux familles de passagers pour n'en faire plus qu'une. Pensez à supprimer les deux anciennes colonnes."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"id": "ad70a5d5",
|
||
"metadata": {
|
||
"scrolled": true,
|
||
"ExecuteTime": {
|
||
"end_time": "2025-09-16T10:06:46.923527Z",
|
||
"start_time": "2025-09-16T10:06:46.915421Z"
|
||
}
|
||
},
|
||
"source": [
|
||
"titanic['Famille'] = titanic['SibSp'] + titanic['Parch']\n",
|
||
"titanic = titanic.drop(['SibSp','Parch'], axis=1)\n",
|
||
"titanic.info()"
|
||
],
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"<class 'pandas.core.frame.DataFrame'>\n",
|
||
"RangeIndex: 891 entries, 0 to 890\n",
|
||
"Data columns (total 7 columns):\n",
|
||
" # Column Non-Null Count Dtype \n",
|
||
"--- ------ -------------- ----- \n",
|
||
" 0 Survived 891 non-null int64 \n",
|
||
" 1 Pclass 891 non-null int64 \n",
|
||
" 2 Sex 891 non-null object \n",
|
||
" 3 Age 891 non-null float64\n",
|
||
" 4 Fare 891 non-null float64\n",
|
||
" 5 Embarked 891 non-null object \n",
|
||
" 6 Famille 891 non-null int64 \n",
|
||
"dtypes: float64(2), int64(3), object(2)\n",
|
||
"memory usage: 48.9+ KB\n"
|
||
]
|
||
}
|
||
],
|
||
"execution_count": 23
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "a96edd21",
|
||
"metadata": {},
|
||
"source": [
|
||
"4. Parmi les colonnes restantes, sur lesquelles un encodage one-hot vous semble pertinent ? En vous appuyant sur [la documentation de scikit-learn](https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.OneHotEncoder.html), appliquez cet encodage aux colonnes identifiées. Enfin, pensez à supprimer les anciennes colonnes.\n",
|
||
"\n",
|
||
"Note : pour le genre, il n'y a que deux possibilités dans ce jeu de données. Avec l'option `if_binary` de l'encodeur, vous pouvez ne générer qu'une seule colonne (l'autre s'obtenant immédiatement par déduction)."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"id": "1316c770",
|
||
"metadata": {
|
||
"ExecuteTime": {
|
||
"end_time": "2025-09-16T10:06:46.969766Z",
|
||
"start_time": "2025-09-16T10:06:46.953407Z"
|
||
}
|
||
},
|
||
"source": [
|
||
"enc = OneHotEncoder(drop='if_binary')\n",
|
||
"\n",
|
||
"one_hot = np.array(enc.fit_transform(titanic[['Sex', 'Embarked']]).toarray())\n",
|
||
"one_hot_label = enc.get_feature_names_out(['Sex', 'Embarked'])\n",
|
||
"\n",
|
||
"df = pd.DataFrame(data=one_hot, columns=one_hot_label)\n",
|
||
"titanic = titanic.join(df)\n",
|
||
"titanic = titanic.drop(['Sex', 'Embarked'], axis=1)\n",
|
||
"\n",
|
||
"\n",
|
||
"titanic.info()"
|
||
],
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"<class 'pandas.core.frame.DataFrame'>\n",
|
||
"RangeIndex: 891 entries, 0 to 890\n",
|
||
"Data columns (total 10 columns):\n",
|
||
" # Column Non-Null Count Dtype \n",
|
||
"--- ------ -------------- ----- \n",
|
||
" 0 Survived 891 non-null int64 \n",
|
||
" 1 Pclass 891 non-null int64 \n",
|
||
" 2 Age 891 non-null float64\n",
|
||
" 3 Fare 891 non-null float64\n",
|
||
" 4 Famille 891 non-null int64 \n",
|
||
" 5 Sex_male 891 non-null float64\n",
|
||
" 6 Embarked_C 891 non-null float64\n",
|
||
" 7 Embarked_Q 891 non-null float64\n",
|
||
" 8 Embarked_S 891 non-null float64\n",
|
||
" 9 Embarked_U 891 non-null float64\n",
|
||
"dtypes: float64(7), int64(3)\n",
|
||
"memory usage: 69.7 KB\n"
|
||
]
|
||
}
|
||
],
|
||
"execution_count": 24
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "439cb51a",
|
||
"metadata": {},
|
||
"source": [
|
||
"5. Vous devez à présent avoir un jeu de données sur 10 attributs, tous numériques (`int64` ou `float64`), sans aucune valeur nulle. Enregistrer ce jeu de données au format csv, afin de pouvoir le réutiliser par la suite. Il est inutile de sauvegarder l'index présent dans le dataframe."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"id": "363a15e3",
|
||
"metadata": {
|
||
"ExecuteTime": {
|
||
"end_time": "2025-09-16T10:06:46.988251Z",
|
||
"start_time": "2025-09-16T10:06:46.978468Z"
|
||
}
|
||
},
|
||
"source": [
|
||
"titanic.to_csv('Titanic.csv', index=False)"
|
||
],
|
||
"outputs": [],
|
||
"execution_count": 25
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "abd641a1-6329-4e31-9625-4428b2d4f6d7",
|
||
"metadata": {},
|
||
"source": [
|
||
"### Le fichier est enregistré dans le même dossier que le notebook."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"id": "bfacd97a-51a4-4faf-887d-561b1cbfacf7",
|
||
"metadata": {},
|
||
"source": [
|
||
"# Fin du TP !"
|
||
]
|
||
}
|
||
],
|
||
"metadata": {
|
||
"kernelspec": {
|
||
"display_name": "Python 3 (ipykernel)",
|
||
"language": "python",
|
||
"name": "python3"
|
||
},
|
||
"language_info": {
|
||
"codemirror_mode": {
|
||
"name": "ipython",
|
||
"version": 3
|
||
},
|
||
"file_extension": ".py",
|
||
"mimetype": "text/x-python",
|
||
"name": "python",
|
||
"nbconvert_exporter": "python",
|
||
"pygments_lexer": "ipython3",
|
||
"version": "3.12.7"
|
||
}
|
||
},
|
||
"nbformat": 4,
|
||
"nbformat_minor": 5
|
||
}
|