{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "# Construire un classificateur des k plus proches voisins\n",
        "\n",
        "CSI 4506 — Introduction à l’intelligence artificielle\n",
        "\n",
        "Marcel Turcotte  \n",
        "2026-09-13\n",
        "\n",
        "# Introduction\n",
        "\n",
        "Ce cahier construit un petit classificateur des $k$ plus proches voisins\n",
        "(KNN) à partir des principes fondamentaux. KNN illustre une approche de\n",
        "l’apprentissage très différente de la construction d’un arbre de\n",
        "décision ou de l’estimation des coefficients d’un modèle linéaire. Son\n",
        "idée centrale tient en trois mots :\n",
        "\n",
        "> **Mémoriser, chercher, voter.**\n",
        "\n",
        "L’apprentissage consiste à mémoriser les exemples d’entraînement.\n",
        "Lorsqu’une prédiction est demandée, le classificateur trouve les\n",
        "exemples voisins et les fait voter. L’implémentation accepte plusieurs\n",
        "classes ainsi que le vote uniforme et le vote pondéré par la distance.\n",
        "\n",
        "# Préparation\n",
        "\n",
        "L’installation ne s’exécute que si `palmerpenguins` est absent, ce qui\n",
        "permet d’utiliser le cahier dans une nouvelle session Google Colab."
      ],
      "id": "06f26264-bbbe-4b48-a976-7b40c7dca173"
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {},
      "outputs": [],
      "source": [
        "import subprocess\n",
        "import sys\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "\n",
        "try:\n",
        "    from palmerpenguins import load_penguins\n",
        "except ImportError:\n",
        "    subprocess.check_call([\n",
        "        sys.executable, \"-m\", \"pip\", \"install\", \"-q\", \"palmerpenguins\"\n",
        "    ])\n",
        "    from palmerpenguins import load_penguins\n",
        "\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.preprocessing import StandardScaler"
      ],
      "id": "imports"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "# Manchots de Palmer\n",
        "\n",
        "Nous conservons la longueur et la profondeur du bec et classifions les\n",
        "trois espèces de manchots. Ces deux attributs produisent un exemple\n",
        "instructif : les classes sont discernables, sans être parfaitement\n",
        "séparables. Modifier le voisinage change donc la frontière de décision."
      ],
      "id": "13ea51ed-7e70-41c9-93a2-906c2af025d5"
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {},
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Exemples d'entraînement : 273\n",
            "Exemples de test : 69\n",
            "Classes : ['Adelie' 'Chinstrap' 'Gentoo']"
          ]
        }
      ],
      "source": [
        "feature_names = [\"bill_length_mm\", \"bill_depth_mm\"]\n",
        "\n",
        "penguins = load_penguins()\n",
        "penguins = penguins[feature_names + [\"species\"]].dropna().copy()\n",
        "\n",
        "X = penguins[feature_names].to_numpy()\n",
        "y = penguins[\"species\"].to_numpy()\n",
        "\n",
        "X_train, X_test, y_train, y_test = train_test_split(\n",
        "    X, y, test_size=0.2, random_state=42, stratify=y\n",
        ")\n",
        "\n",
        "print(f\"Exemples d'entraînement : {len(X_train)}\")\n",
        "print(f\"Exemples de test : {len(X_test)}\")\n",
        "print(f\"Classes : {np.unique(y_train)}\")"
      ],
      "id": "penguins-data"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "# Mise à l’échelle des attributs\n",
        "\n",
        "La distance euclidienne dépend de l’échelle numérique. Nous\n",
        "standardisons donc chaque attribut à l’aide de la moyenne et de\n",
        "l’écart-type du jeu d’entraînement :\n",
        "\n",
        "$$\n",
        "z=\\frac{x-\\mu_{\\mathrm{train}}}{\\sigma_{\\mathrm{train}}}.\n",
        "$$\n",
        "\n",
        "Le transformateur est ajusté uniquement sur les données d’entraînement.\n",
        "Les données de test ne doivent pas influencer les choix de prétraitement\n",
        "effectués pendant l’apprentissage."
      ],
      "id": "59896230-99c3-4623-94c4-da430512a93e"
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {},
      "outputs": [],
      "source": [
        "scaler = StandardScaler()\n",
        "X_train_scaled = scaler.fit_transform(X_train)\n",
        "X_test_scaled = scaler.transform(X_test)"
      ],
      "id": "scaling"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "# Recherche des voisins\n",
        "\n",
        "Pour une requête $x$ et un exemple d’entraînement $x_i$, la distance\n",
        "euclidienne est\n",
        "\n",
        "$$\n",
        "d(x,x_i)=\\sqrt{\\sum_{j=1}^{D}\\left(x^{(j)}-x_i^{(j)}\\right)^2}.\n",
        "$$\n",
        "\n",
        "L’implémentation la plus directe calcule toutes les distances, les trie\n",
        "et conserve les $k$ premiers indices."
      ],
      "id": "e9da1149-1a43-41ea-a7e8-7ebd06d32984"
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {},
      "outputs": [],
      "source": [
        "def nearest_neighbors(X_train, x, n_neighbors):\n",
        "    \"\"\"Retourner les indices des voisins et leur distance à x.\"\"\"\n",
        "    distances = np.sqrt(np.sum((X_train - x) ** 2, axis=1))\n",
        "    indices = np.argsort(distances, kind=\"stable\")[:n_neighbors]\n",
        "    return indices, distances[indices]"
      ],
      "id": "nearest-neighbors"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Le tri stable rend le résultat déterministe lorsque deux exemples\n",
        "d’entraînement se trouvent à la même distance. Les implémentations de\n",
        "production peuvent éviter un tri complet ou utiliser des structures de\n",
        "recherche spécialisées.\n",
        "\n",
        "# Vote\n",
        "\n",
        "Avec le vote uniforme, tous les voisins retenus reçoivent le poids 1.\n",
        "Avec le vote pondéré par la distance, le voisin $i$ reçoit le poids\n",
        "\n",
        "$$\n",
        "w_i=\\frac{1}{d_i}.\n",
        "$$\n",
        "\n",
        "Une correspondance exacte demande un traitement particulier, car $1/0$\n",
        "n’est pas défini. S’il en existe, notre implémentation laisse seulement\n",
        "ces observations voter."
      ],
      "id": "cf4c45e8-e84e-4a86-8289-c2d59763e514"
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {},
      "outputs": [],
      "source": [
        "def voting_weights(distances, mode):\n",
        "    \"\"\"Retourner des poids uniformes ou inverses à la distance.\"\"\"\n",
        "    if mode == \"uniform\":\n",
        "        return np.ones(len(distances))\n",
        "\n",
        "    exact_matches = distances == 0\n",
        "    if np.any(exact_matches):\n",
        "        return exact_matches.astype(float)\n",
        "\n",
        "    return 1 / distances\n",
        "\n",
        "\n",
        "def vote_probabilities(labels, weights, classes):\n",
        "    \"\"\"Convertir les votes pondérés des voisins en probabilités.\"\"\"\n",
        "    scores = np.array([\n",
        "        np.sum(weights[labels == label])\n",
        "        for label in classes\n",
        "    ])\n",
        "    return scores / np.sum(scores)"
      ],
      "id": "voting"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "# Classificateur\n",
        "\n",
        "La méthode `fit` mémorise le jeu d’entraînement. La majeure partie du\n",
        "travail se fait dans `_predict_proba_one` : trouver les voisins, leur\n",
        "attribuer des poids et regrouper leurs votes par classe.\n",
        "\n",
        "La validation permet de produire des erreurs claires pour les entrées\n",
        "non prises en charge, mais elle ne fait pas partie de l’idée centrale de\n",
        "KNN."
      ],
      "id": "d7b097ea-ee50-4ec8-a7aa-ffe6b9407016"
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {},
      "outputs": [],
      "source": [
        "class SimpleKNeighborsClassifier:\n",
        "    \"\"\"Un classificateur didactique avec une petite interface de type scikit-learn.\"\"\"\n",
        "\n",
        "    def __init__(self, n_neighbors=5, weights=\"uniform\"):\n",
        "        self.n_neighbors = n_neighbors\n",
        "        self.weights = weights\n",
        "\n",
        "    def fit(self, X, y):\n",
        "        X = np.asarray(X, dtype=float)\n",
        "        y = np.asarray(y)\n",
        "        self._validate_training_data(X, y)\n",
        "\n",
        "        self.X_train_ = X.copy()\n",
        "        self.y_train_ = y.copy()\n",
        "        self.classes_ = np.unique(y)\n",
        "        self.n_features_in_ = X.shape[1]\n",
        "        return self\n",
        "\n",
        "    def _predict_proba_one(self, x):\n",
        "        indices, distances = nearest_neighbors(\n",
        "            self.X_train_, x, self.n_neighbors\n",
        "        )\n",
        "        labels = self.y_train_[indices]\n",
        "        weights = voting_weights(distances, self.weights)\n",
        "        return vote_probabilities(labels, weights, self.classes_)\n",
        "\n",
        "    def predict_proba(self, X):\n",
        "        X = self._validate_prediction_data(X)\n",
        "        return np.vstack([self._predict_proba_one(x) for x in X])\n",
        "\n",
        "    def predict(self, X):\n",
        "        probabilities = self.predict_proba(X)\n",
        "        return self.classes_[np.argmax(probabilities, axis=1)]\n",
        "\n",
        "    def score(self, X, y):\n",
        "        return float(np.mean(self.predict(X) == np.asarray(y)))\n",
        "\n",
        "    def _validate_training_data(self, X, y):\n",
        "        if X.ndim != 2 or y.ndim != 1 or len(X) != len(y):\n",
        "            raise ValueError(\n",
        "                \"X doit être 2D et y doit être 1D, avec le même nombre de lignes\"\n",
        "            )\n",
        "        if len(y) == 0:\n",
        "            raise ValueError(\"le jeu d'entraînement ne peut pas être vide\")\n",
        "        if not np.isfinite(X).all():\n",
        "            raise ValueError(\n",
        "                \"les valeurs manquantes ou non finies ne sont pas prises en charge\"\n",
        "            )\n",
        "        if not isinstance(self.n_neighbors, (int, np.integer)):\n",
        "            raise ValueError(\"n_neighbors doit être un entier\")\n",
        "        if not 1 <= self.n_neighbors <= len(y):\n",
        "            raise ValueError(\"n_neighbors doit être compris entre 1 et len(y)\")\n",
        "        if self.weights not in {\"uniform\", \"distance\"}:\n",
        "            raise ValueError(\"weights doit être 'uniform' ou 'distance'\")\n",
        "\n",
        "    def _validate_prediction_data(self, X):\n",
        "        if not hasattr(self, \"X_train_\"):\n",
        "            raise ValueError(\"appelez fit avant d'effectuer des prédictions\")\n",
        "        X = np.asarray(X, dtype=float)\n",
        "        if X.ndim == 1:\n",
        "            X = X.reshape(1, -1)\n",
        "        if X.ndim != 2 or X.shape[1] != self.n_features_in_:\n",
        "            raise ValueError(\"X n'a pas le bon nombre d'attributs\")\n",
        "        if not np.isfinite(X).all():\n",
        "            raise ValueError(\n",
        "                \"les valeurs manquantes ou non finies ne sont pas prises en charge\"\n",
        "            )\n",
        "        return X"
      ],
      "id": "classifier"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "# Entraînement et évaluation\n",
        "\n",
        "Les deux modèles ci-dessous utilisent les cinq mêmes voisins. Seule leur\n",
        "règle de vote change."
      ],
      "id": "d712747c-6ea9-476e-b1ad-b237771770b6"
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {},
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "k=5, weights='uniform' : exactitude=0.957\n",
            "k=5, weights='distance' : exactitude=0.957"
          ]
        }
      ],
      "source": [
        "for mode in (\"uniform\", \"distance\"):\n",
        "    knn = SimpleKNeighborsClassifier(n_neighbors=5, weights=mode)\n",
        "    knn.fit(X_train_scaled, y_train)\n",
        "    accuracy = knn.score(X_test_scaled, y_test)\n",
        "    print(f\"k=5, weights='{mode}' : exactitude={accuracy:.3f}\")"
      ],
      "id": "evaluation"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "## Examiner une prédiction\n",
        "\n",
        "Puisque KNN fonde directement une prédiction sur les exemples mémorisés,\n",
        "nous pouvons examiner le voisinage responsable d’un résultat\n",
        "particulier."
      ],
      "id": "7dee4be6-19a8-4262-a72d-ef17bde9c599"
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {},
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Classe réelle : Adelie\n",
            "Classe prédite : Adelie\n",
            "\n",
            "Voisins :\n",
            "1 : Adelie    distance=0.018\n",
            "2 : Adelie    distance=0.053\n",
            "3 : Adelie    distance=0.088\n",
            "4 : Adelie    distance=0.100\n",
            "5 : Adelie    distance=0.128"
          ]
        }
      ],
      "source": [
        "query = X_test_scaled[0]\n",
        "indices, distances = nearest_neighbors(X_train_scaled, query, n_neighbors=5)\n",
        "\n",
        "print(f\"Classe réelle : {y_test[0]}\")\n",
        "print(f\"Classe prédite : {knn.predict(query)[0]}\")\n",
        "print(\"\\nVoisins :\")\n",
        "for rank, (index, distance) in enumerate(zip(indices, distances), start=1):\n",
        "    print(f\"{rank} : {y_train[index]:9s} distance={distance:.3f}\")"
      ],
      "id": "inspect-prediction"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "# Frontières de décision\n",
        "\n",
        "Notre classificateur suffit pour effectuer des prédictions, mais lui\n",
        "faire calculer quatre grilles denses détournerait l’attention de sa\n",
        "logique essentielle. La visualisation suivante utilise donc les versions\n",
        "optimisées de `KNeighborsClassifier` et de `DecisionBoundaryDisplay` de\n",
        "scikit-learn.\n",
        "\n",
        "Un pipeline standardise les attributs à l’interne, ce qui permet de\n",
        "conserver leurs unités originales sur les axes."
      ],
      "id": "93192752-001a-4076-9e06-79ad6d6b2fe7"
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "fig-height": 7,
        "fig-width": 11
      },
      "outputs": [
        {
          "output_type": "display_data",
          "metadata": {},
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABBYAAAKeCAYAAAAREZglAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90\nbGliIHZlcnNpb24zLjExLjEsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvctoD+AAAAAlwSFlzAAAO\nxAAADsQBlSsOGwABAABJREFUeJzsnQWcFOX/x793XNMpIKGUgHTYhSiCioK0KJ12/sFGFONnF90g\nIi1dYisg0qXSBiqdd3DH3f/1eXDWvb2N2dmZ3Zndz/v1Oo6NmXnmmb39PPPNuJycnBwhhBBCCCGE\nEEIIMUC8kY0IIYQQQgghhBBCaFgghBBCCCGEEEJISDBigRBCCCGEEEIIIYahYYEQQgghhBBCCCGG\noWGBEEIIIYQQQgghhkkwvikhhBCrQeOec+fOqR9CCIlV8uXLp37i4uIiPRRCCCFeoGGBEEJsSmZm\npuzfv19OnToV6aEQQkjEyZ8/v5QpU0YSExMjPRRCCCEexOXAHUYIIcRW4Kv5119/VR66UqVKqYU0\nPXWEkFj9PoSh9Z9//lHRW1WrVuX3ISGE2AxGLBBCiA05e/asWkCXL19eUlNTIz0cQgiJKCkpKZKQ\nkCB79uxRRoakpCReEUIIsREs3kgIITYmPp5f04QQ4v59yGBbQgixH1yxEkIIiRjHjh2TP/74Q9d7\nEcGxY8cOyc7OtnxcsQo8wZhj3rgRQgghJBhoWCCEEBIxJk2aJJ07d9b1XhSyRG718ePH1WMaGsxn\n7969ao7PnDnj9XXOOSGEEEK8QcMCIYQQU8GNP7zeVrfIPHDggDRv3pxdMwyAiAQYao4cORKxOc/K\nymJ0RASu+19//aWKIJoNrychhMQ2NCwQQggxlQceeEB5vWfNmmXpzJYuXVrdmBYsWNDS40QTiEQY\nMGCAlChRQqpVq6auEwqEvv/++7rSH8yc899//10dn+1Uw3Pdn3rqKdVhplKlSlKnTh0pUqSI9OvX\nT3cqUiB4PQkhJLahYYEQQqKc06dPy86dO9Vvq4EHfM6cOfLcc8/JiBEjvL4HkQzwfPvj0KFDAcfr\nLyxfz/axeI3atGkjn332mcyfP19OnDghBw8elFWrVsmuXbskPT09z/z+/ffffufcvSaDt/e7fy48\noyP27dunfuPY2MfRo0fz7A+pGfg/nsMPbl49DSDu28BrjnNyCuG87rNnz1Z/mzDkIGoBc9mwYUP5\n4YcfdP39BLrW3q6nO3g/ro8vAr1OCCHE5uQQQgixHRkZGTlbt25Vv42SmZmZM2DAgJy0AvlxJ6Z+\n4zGet4p333035+677845ffp0TsmSJXN+/fXXXK+PGjUqJ3/+/DmlSpXKufjii3PuvffenOuvv971\n+tKlS3MqV66cU6JEiZyCBQvmtG3bNufYsWPqtd9++02dx5EjR7w+DrS9HdGuUYH8aepc8Nuqa/T5\n55/nxMXF5axbt87ne3C9MI6nn35aXaOiRYuq+fz555+9znmg9//55585l112mboW+Dxce+21OXv2\n7FGvVapUSW2L39hm9OjRrv393//9n7qGeB6fJfzGT9myZXOKFCmSM378+Dxjfvzxx3OKFSuWU6BA\ngZx69erl+ezZ8brnz19AjR2/rbruy5cvV9d9/fr1Ad/r7+8n0LX2dj3Bxx9/rPaH65+amprzwAMP\n5Jw9e9Z1zECvm/29SAghxBpoWCCEEBtixgIaNyqJyUk5l/fvlNNq+GD1OyEpST1vFTVr1sxZvXq1\n+v+zzz6rbhA1Nm3alJOYmKhucMH27dtzihcv7jIs7Nq1S93MzJw5Uz3GDeWtt96a89BDD+kyLATa\n3o7gWqQkJea8dnvjnO8fbql+JycmWHKNnnnmmZyLLrrI73u0m0fMWVZWlvq56667lAHIn2HB1/uf\nf/75nJtvvtl1o/jdd9/lzJs3T/1/9+7datsTJ07kOX7Pnj193mR///33OYULF87Zu3dvrm3at2+v\njoPt7rnnnpyrrroqx67g+iYlJefc3v6hnIefH69+JyZa87eJv0PP675//341b/jB//X8/QS61t6u\nJ/7mExIScubOnase45pVrFgx5/XXX9f1uic0LBBCiH1hKgQhhEQhCGP+4KMPpUGPNlK30+1SqmZl\n9btBj7vkw48+siT0+ptvvlF95osWLapCoW+55RaZOHGinD17Vr2OUOwbb7xR/YBLLrlEunbt6tp+\n+vTpUqNGDZX/je2R+92hQweZO3euruOHun24wTX46MMPZFCzevJEk9pyWYWS6jceD/3oQ9OvEVp7\nIsdeD08//bTky5dP/dx2222yfft2Q+9H/QaEuG/btk09vuqqq+T2228PeHyk0iQkJOR6LiMjQ4Xb\nlyxZUqpXry7fffddrtdffPFFSUxMVNsNHjxYvv/+exXybzdwXT/88CNp1qqvNLn1XqlQ6VL1u1nr\nvvLRR0PDct0xVyjCiVQI1EQJ5u8nmM8GUi+uvPJKadmypXpcoUIFue+++2TGjBm6XieEEOIccqs2\nIYSQqAAV/0+fPCVl6tbI9XzZejVl9fCp6oYLRdzMZOTIkSqvGjcsGjAq4Oahffv2qq4Civ+5U6ZM\nGdf/cSODG1D37UFKSoqu44e6fSSu0clTp+W6yrnn5PoqZeSpBWtMv0blypWTTz/9VNd7Cxcu7Pp/\nUlKSyzgU7Pt79uypCj2+8sor6gb00ksvVf+vWLGi3/1dcMEFrv+jFkSXLl1k8eLFUrx4cXU9MXf4\nccf9s6V9rrx95iINxn3q1EmpXL1BruerVG8oC6Z9YPp1v/DCC2X37t25nhs2bJj6jeKNWk0KvX8/\nwXw2MP9ly5bN9RyujVafIdDrhBBCnAMjFgghJArB4jytQH7Zv+G8p1jjz/VbJX+BAqbfbB0+fFgV\nBdyyZYur0B5+Ro8e7SriiJulTZs25dpu48aNrv8jggGeaPft8eO5jS9C3T4S16hA/jT5emdur/pX\nO/ZLwQL5Tb9GrVq1UkX54Jn2BMX49HSFCBbsF17vqVOnyrp161S0ATzSAB5v4K34pjujRo1SN724\nAUbBQVxTXGfP7dyvMz5XiJ4JZMCI1HXPn7+A7Ny+NtfzO7b/JAUKFDT9ut95550+r7uZfz/erqe3\nv/kNGzZI5cqVdb1OCCHEOdCwQAghUUhaWpo8eP8D8tOYmbJ+yjz5Z+tO9Xvt2FnywP33q9fNZMKE\nCSqsulChQrmev+mmm+Tbb7+VX3/9Ve655x5VAR9t79avXy8fffRRrpsdeKVx89i3b19ZuXKlrF69\nWl5//XXp3bu3rjGEun24wTW4/4EH5YUl6+SNFRtl9b4D6vegpevlvvsfMP0a4cYR0QLdunVTqQJI\nJVizZo2MGzdOGjRokKcrhBk88cQTKuwe1wPH2rx5s8vjjRtoeMORIuOti4BGamqq8mxje3xusE/8\n9uTBBx+UL7/8Ur7++mv1GcDnwfPzaAdwXR944H5ZMnu4rFg4Ufbt2qJ+L509Qu6//z7TrzuMBdp1\nf+GFF9T8wACIziC4LsnJyab8/Xi7nvibRwTGww8/rAxLY8aMUdESjz76qNom0OuEEEIcRKSLPBBC\nCLG2K0T+Av9Wni9gXeX5li1b5nz00UdeX+vQoUPO22+/7SrWdscdd+TUr18/p2/fvjnvv/++6iKh\nceDAgZxHH30058orr8y54oorcp566qmco0ePqtdQZA7V5o8fP+71caDt7Yh2jQr+27mjYBg6d6B4\nJq5J7dq1VceG3r17q8+aVjwPc3rmzBnX++fMmaOur7c5D/T+U6dO5QwePFhdi0aNGuU8/PDDOYcP\nH3a9d8qUKepaValSRXUR8LY/FGR88sknc+rWrZvTsGFDtb8ePXq4ug5oRQUXLlyY06JFi5wGDRqo\noqF27hzg6gZSoOD5biAFClp+3b/44gv1t4Z5RNeM1q1bqy4t7nPt7+8n0LX2dj2169O5c2d13Ftu\nuSXns88+yzWuQK+7w+KNhBBiX+LwT6SNG4QQQnJz5swZ1Q8eocKaR9EoKAYHryA8imZ7Q4k58BoZ\nB97xqlWrqogLu9bT8AWve+S+FwkhhJgLizcSQkiUA2OC2YUaibnwGsUmvO6EEEKiBdZYIIQQQohj\nQWcCFPtDsUZCCCGERAZGLBBCCCHEsVSoUEGlQxBCCCEkctC8TwghhBBCCCGEEMPQsEAIIYQQQggh\nhBDD0LBACCE2Jjs7O9JDIIQQW30fxsXFRXoohBBCPGCNBUIIsWlBunz58smff/4ppUqVksTERC6m\nCSExCTqjZ2Zmyj///KO+F/F9SAghxF7E5eDbmhBCiO3AQnr//v1y6tSpSA+FEEIiTv78+aVMmTI0\nLBBCiA2hYYEQQmwMbL8I/83Kyor0UAghJGIkJCSolqJMgyCEEHtCwwIhhBBCCCGEEEIMw+KNhBBC\nCCGEEEIIMQwNC4QQQgghhBBCCDEMDQuEEEIIIYQQQggxDA0LhBBCCCGEEEIIMQwNC4QQQgghhBBC\nCDEMDQuEEEIIIYQQQggxDA0LhBBCCCGEEEIIMQwNC4QQQgghhBBCCDEMDQuEEEIIIYQQQggxDA0L\nhBBCCCGEEEIIMQwNC4QQQgghhBBCCDEMDQuEEEIIIYQQQggxDA0LhBBCCCGEEEIIMQwNC4QQQggh\nhBBCCDEMDQuEEEIIIYQQQggxDA0LhBBCCCGEEEIIMQwNC4QQQgghhBBCCDEMDQuEEEIIIYQQQggx\nDA0LhBBCCCGEEEIIMQwNC4QQQgghhBBCCDEMDQuEEEIIIYQQQggxDA0LhBBCCCGEEEIIMQwNC4QQ\nQgghhBBCCDEMDQuEEEIIIYQQQggxDA0LhBBCCCGEEEIIMQwNC4QQQgghhBBCCDEMDQuEEEIIIYQQ\nQggxDA0LhBBCCCGEEEIIMQwNC4QQQgghhBBCCDEMDQuEEEIIIYQQQggxDA0LhOjg+PHj0qtXL1m5\ncqUl83XgwAH54IMP1DGmTJlim2uSlZWlxvTVV1/5fd+WLVvU+/766y9Lx4Nj/PTTT2E715ycHJk9\ne7Y88cQT0rt3b/n222/FSWjX5cyZM5EeCiGEOI5x48bJCy+8YPp+f/zxR/Xd7Pnz3HPPiR349NNP\nZeDAgQHfN3jwYDVHVjJr1ix5+eWXw3qu+/btkyFDhki/fv3k6aefFqeB6zJnzpxID4PEIDQsEKKD\n06dPy5gxY2THjh2mz9f9998v9erVk19++UUJ9Pfff2+ba4KbbZz3tm3b/L7vt99+U+87evSopePB\nMXbv3h22c3377bela9euUrx4cbn88sulZMmS4iS065KZmRnpoRBCiOOAoXnmzJmm73fnzp3qu7lm\nzZpyxRVXuH7q1q0rduCHH36QqVOn6rrpD+R4CJXVq1dbepPsea5HjhyRRo0aqePWr19f/TgNXJc1\na9ZEehgkBkmI9AAIiXV69uwp7733niQkJMiwYcPETiQmJsqoUaPkmmuukWjH27kuWLBAbrnlFnnq\nqaciOjZCCCHRx9133y2lS5cWu9GxY0e58sorJRbwPFdEkyCKFI6FypUrR3RshDgNGhYICYGhQ4fK\npk2b5Nlnn5ULL7zQ0D4aNGhgasrGY489pkIqy5QpI2PHjlVRBE2bNpU77rgjz/vXrl2rPAGHDh2S\nSpUqSefOnXMtcpAKgPSPqlWrSvXq1V3PL1u2TBYuXChpaWnKo++Nf/75Rz7++GMVYXDBBReoG3R4\nAcz0xr/44ovy5JNPyiWXXOJ6fuPGjfL+++/Lq6++qiIM3OekYsWKMnr0aDW2G2+8UVq3bu31XAsV\nKiTPP/+8uraFCxdW24KRI0dKfHy8rrlzP2758uVVNArCK/GZgScE1+bdd9+VDRs2KO8CjoOwy1Kl\nSqkIg8mTJ6tj4Nz69OkjSUlJeebgu+++U9cB1xjj7tatmxQpUsS0OSaEEJKbXbt2yWuvvSZXX321\nT/0LN9CXPXv2qLSNTz75RFatWiVly5aVHj16KE1x59SpU0pfNm/eLKmpqXLTTTdJs2bNcr1n7969\nsm7dOunQoYPrOegmdOuPP/6Qyy67TGmeJ9BRaBK0CSl4tWvXVjfuKSkppp0rdBM88sgjuZ7HuUOL\ntWuizQnWCdOnT5dvvvlGzQXSGrEm8XauSEn97LPP1PNIS8EaB+uE2267TffcaccdNGiQzJgxQx0X\n78Ma7PHHH1eRKVgPYS2CNcHNN9/s2j/2i3UT1gBdunSROnXq5Dl/RFQgZRbRlfnz55eWLVvGhPOH\nOAOmQhBiAIgnRA259xAEo0YFq1I2Fi9eLPfcc48SPtwIt2vXLo8Iv/HGG9K4cWP5/fff5eKLL1a1\nBGA8cK8j4S09AGLZokULtWjQbrohyu78+uuv6ob4888/lypVqqj9PPjgg/LRRx+FdH6IKNCME7ih\nx9j279+f6z0Qajx/7NixXHOyYsUKNSeIDMGctG3bVt58802v54p5g/hDtGGc0MJU4+LidM+ddtwl\nS5ZIp06dJF++fGoxhn0g7QWvTZo0SXlF8PnBYuyqq65S44YXa+vWrco4hMUNFoee9O3bVxmMTpw4\nobwqMHLUqlUrV6oIHmPOzFzUEUJIrAJjL76n//77b2nfvn3I+3v99dflvvvuUze/odQP0lI2Hnjg\nAXUjC2M2bqZxY4qbXHeDPG72ESUJwwN0Dze83bt395segH0gTWPatGlq33i9f//+ecaBm3ZEYUJn\nYciHgQHOk+zsbMPn1qZNm1y1J7C+wY8nOH+cu+ec/N///Z9ai1SoUEGdE3QY+uztXLFugZEeIAUC\nuq+t7/TOnXbchx56SJYvXy6XXnqplCtXTr0Go8GXX36pjBgnT55UDoM777xTGWzgsIHhoUSJEqo+\nEqIosFZwB06JatWqqf1gfrEWhWEDNRU8jSzujhNCwgUjFggJkrNnzypLMkQAooFFhgbEz/NG1xsQ\nAQiTVeCGFYsUzXsNYYcVHx5t1HOAaKFYERYziLYADz/8sFx//fVy7733ys8//+zyzHt6alDQ6J13\n3lGGAoCFDDw37kyYMEGKFSsm8+bNc92MIwIAnoFQ5kqLHDACxoQQR0QjAIwLXiecN9Ig3ClatKg6\n1vDhw9VNu/txg5077VrACOPJ+vXrXfm7MB7AkHD77berhRC8EADP4fgoXnXRRRep5yZOnKiiJ/AZ\nhCdEG8O1116rrsv8+fPVc1jMhDJnhBBCzgO9v+uuu5RBYcSIEcpYDFAXCTeGgcCNIm66NXATm5yc\nrL6ncTMKoz2++2FsMAKMyvi+15wIqN8EYwBS+RDFoOk1nAJwBmiaBP2GoR0OA1/GEhQwLFCggCpg\nDC8+gHcfN7hYU2hRepgH3KS77we6r60DjMwVjPhGgUEEugnjDUD0BB5jPeDNMIKbdJwHtB8OAc0g\nEOzc4Vog0hDGBU9wLVD/ANEV2poS64nrrrtOORjwuYKew3AAI4bmkEEUA47RsGFD9T5tnYG5gvGl\nVatWrggHGhVIpKBhgZAggHf41ltvVVZkWMZReMkdiDjEIBDwiFsJFj/uIfHwgD/66KMqfA6LAITn\nwdLtLnq4uYZwIrwRogervifwVkDM4JXQwCID1nfcJLuDiALcZGspFFhYaDfGkZgrCK9mVNAWEBBt\nLepAL8HOHbwR3owKAIYIDVwvzAkWJJpRATRp0kR5e5Dioc0fFmaI3NCMCgCLERgnkE6BcE1EWxBC\nCAkd3CzDOI8oRRjX3dGi2gLhnq9/ww03qLB3ePYBPNXwMsOQju/85s2bBz1GeNARyeauzbg5hwEc\nxmjcwKJu0IABA3JpEtYLmjffm2EB+0X0A25+NaMCwDoAhhBPkIYBvdUML+46b2SuzOj2pAFDAdZt\nnuuVQEBTg5k7GAF8GfXhjNKMCtpnAWsRpEdoc4bPBRwF7tGgiLqAkQbRju7OCxgR4MhBSqW31AlC\nwgkNC4QEAcQfQoUwdYT4ewLLtR3wHBvECjfPqEStRR4gx9D9RhsgxE573ZthAc8jLNAztF4LHdSA\nJwALEYQAYj8I2cdCw726crjnynOhgnBDgBaZwRgWgp07b58TDc/jYnHgGT2B5wBCbzVgsIHhwHPh\nglDNc+fOqRxYbTyEEEKMA92EcR5GW0+jgqZ/nhoYCG8FGxEJh8g03CAaMSwgss/TEI9xQRPgucfa\nBf/31AYY/fE+aJc3kF6IbT21DOsAd48+NBFRF/hBHQLUDsDNMrReG5eRuQoFpG141ieC9gfbGhtz\nEMzcIdLQ3QgTSPd9PY8oS3fdB6jNoEWgaMD5gDESEmloWCAkCBDaBss/vNPIafe8wbZLKgQ8E54g\nhA9hl5o13Nt7tOc0L4on2B778bZvd2B8gDcGdQ0Qro+F0iuvvCIvvfSSPPPMM6bMlbZYwILHM6rE\n19jd0Sz+WCwEQ7BzV7BgQZ/78jYmPePEczBueHp+8Bj1NJzWFpMQQuwKIsUQ6Td+/HhlIIeh3B2j\nqRCe4OYbrY0PHjxoqu4Dd13xpV/+dN99X972r4GUR9Qygncf+o+oBhgaUH8IumTGXEH73Wsk+NN+\nTz3V9NOI7gczd8Hqvq/nPXVfSw3xdGxA+/05MQgJFzQsEBIECD9HfQVY4hHijurB7sYFu6RCIGze\nM4wPFnUsigAiCSDuSANw9zho4YF43RvICYWlHwsfzeMP0D3BE3je4a3AD4okwuPzv//9z2VYCHWu\nNGPDn3/+met5b2MxE6NzZyZY5CIqgfUTCCHEWhDxBw88UiGQpjZ37txcaWhmhfejxSF+9OiiN6DN\n2N7dsAw9hOccWoWbVPwfnYg8b4xRtBhpnr60FvUVPLUV6Y7enAMI80etH/zgWFr6JSIZzZgrjOfr\nr7/O9dzhw4eVJlppXDIyd2ai1bJAKoeRiBZCwgENC4QECQohItcNCwttkaHd/FoZ3o9iQsi3Q9Go\nQEB0ka6h1YBAkUIsKrQbURQlQj4nbvLhhUE4H1pJoUMB8vpq1Kjhdb+oIYB8TeRaonATQIilZ4Vm\nFA+8/PLLcy1wYGhwN0aEOldaTQKEBKJWAc4BxhNv1aLNxOjcmQnyfOE1Q9st924fWh6oGdXKCSGE\n/GdcQLQivMboBACnAhwMRsP7caON9YN7NAAiIeH5dm9hiZx6RPrhO92zraEnCJ1HZCCKK2vbImwe\nhQsxbvyggLMWdanVP0KXIxgJEJHpDWgcjAJoo4h9abV+cCwUOnZPxdu+fbtrXtwjCzXtNyMVAjUJ\nsA5Chw6tXbfnWMwG6xcjc2cmaCmJSA7UecB5u7cRRWcLnL97W3BCIgENC4QYtBy7GxfQ/cBoFMKn\nn36q0gW0PLkvvvjCZQBAvqWWi4n34KZRj2EBwgehg5jjphddCYYOHepqm4T8PxwXIYso9gMxQjFK\nHAtVnn0B4UI3AggshAyeFeT1oZChe0tEeA9gWICXBO/ZsWOHakHp3r7KDCDqWOShpgFyKXFc3Ghb\n2Vvc6NyZCYp7oao15h3RE2grieuMoqLhWOAQQkisgRtzGJPdjQuBbvZ9ge5CKHyInH0UA0SqANoP\nagWWNfC9jmJ9+I4PdCykx8HTj8hK3PxjLYEoQ617EUDHCTgDoJk33nijinDA+gDFAz27O7kDQ7rW\nOhL6A92HBrqnKcJI8tZbb6luFFoRQayTYIxHkUOzQLrfsGHDlHEdazAUPEZ0BM7fSozOnVnAwIPW\n1oj+xOcGhgZEUcCJhM8QjB6ERJq4HJQ3J4T4JT09Xd00QlDdQ/RgnUf7JeS8wXtuBCwoUI/Al4Bq\nFYixQEAIIW4o/YVC4sYXLQ7RqQHGiKNHjypPOm68veUkos0VbshROAji6F5tGFEO8FJAwNwt4TgO\nejFD1LDYQWoEIgXcxwsPDHouYwECwce+rUgBQfgjzgHHxVgwtiVLligPD/IQfV07jBl1MmAYwvi8\nnSsWjtgvPCTBzp2v4wIYWbAtIi3c8yqXLl2qxoHWVRqoIYHFrOc10CIU8PnD4hNGHCz6fHWgIIQQ\nEhyI/oNWuN8YwwGgFc/DTbO31sx6gDajOB/SCWD0R7cAT42EJqHTAzTDX4oEjP3oSIS1BIr84f9Y\nC0C7vI0PqXt4L44HbfG8Kce6BBEPWEe4nzd0C+mHaHkIPcKNLupCoFWiBiIHEV2JG+HatWtbUqwR\nOgnDCYoaw4uP9AvUcoIXH+fj69qBRYsWqQgUzVDjea4wHqC1KG7gvRVgDDR3vo4LYDhCLQT3As84\nBzin0C7SPaoTzhusn9yvgbthCimviAhBZGo4IiUJ0QMNC4Q4ANws48YRlml/1f7dDQvwJhBCCCHE\nmcBggDD8UaNGBXyfZlgghJBIwVQIQhwAPAUzZ85kC0FCCCEkRmjdurWKiCSEECfAiAVCogh/4feE\nEEIIiT78hd8TQki4oGGBEEIIIYQQQgghhjFWcYYQQgghhBBCCCGEhgVCCCGEEEIIIYSEQkK0FrpD\nG5+UlBTV7oYQQgghoYMO1RkZGVKkSBHDbfasgtpPCCGERE77o9KwAKMC+uoSQgghxHwOHTokxYoV\ns9XUUvsJIYSQyGl/VBoWEKkAVu7cJimp5/9PCCFO4MTZHVKuQNFID4MQr2SkZ0iN8te4dNZOUPsJ\nIRrUUkLCr/1RaVjQ0h9gVEhJTY30cAghRBcnzv4iKanJkkqDKLE5dkwzpPYTQjQy81FLCQm39kel\nYYEQQpxmUNAoX8Be4eWEEEIIIYQEgoYFQgiJEDQoEEIIIYSQaICGBUIIiSCMUCCEEEKsMdoTQsIH\nDQuEEEIIIYSQqDAo0GBPSIwaFo4fPy4JCQmSlpZm6HVCCCGEEEJIbEKDAiH2ID5SB160aJE0btxY\nKlSoICVKlJArr7xStmzZovt1QgghhBBCCGGUAiExbFhYvny5jBgxQo4cOSKHDh2SihUrSocOHXS/\nTgghhBBCCCGEkBhOhXjrrbdc/09NTZV77rlH7rzzTsnKylKpD4FeJ4QQp8LCUoQQQgghJJqwzR36\nZ599JldccYVPo4G/1zMzM5XBQSM9Pd3SsRJCiBHYXpIQ86D2E0IIIfYhYqkQ7iDlYcaMGTJq1ChD\nrw8ZMkQVd9R+ihcvbvGICSHEeHEp5oISEjrUfkIIIcQ+RNywMHToUHn22WdVTYWaNWsG/Tp45pln\n5PTp064f1GQghBC7QYMCIeZB7SeEEELsQ0RTIV5//XV59913ZcWKFVK7du2gX9dITExUP4QQQgiJ\nDaj9hBBCiH2ImGHhhRdekPfee0+mTZsmBQsWlD179qjny5cvL/ny5Qv4OiGEEEIIISQ2YSFkQuxF\nxAwLCxculCJFikifPn1yPb9y5UopXbp0wNcJIYQQQgghsQULIRNiTyJmWPjxxx9Dep0QQpwCvSqE\nEEKIebBmESH2wzbtJgkhJJo7QRBCCCGEEBKt0LBACAkb6adPy4G//5GSF5SS1LS0qJ15GhQIIYSQ\n2NF9QogN2k0SQqKfrKwseevFl+WqarWkWYMr1G88xvPRCqMUCCGExCqxqPuExDqMWCCEWM57Q16T\nccNGSsMebaRM3Rqyf8M2GTd0hHrt8Ree5RUghBBCogjqPiGxBw0LhBDLwyAnjxqrjAp1O92unitV\ns7Lk5OTIx6PHyX1PPhY14ZEs0kgIISTWiSXdJ4T8B1MhCCGWgtzK9FOnVaSCO2Xr1ZTTJ0/JwX8O\nRIVBwb2uAtMgCCGExCqxoPuEkLzQsEAIsRRVsCl/mkp/cOfP9VslrUB+KVGqpKOvAA0KhBBCSOzo\nPiHEO0yFIIRYCsId7+ndQ9VUQBgkPBZYXKwdO1O639c3KsIhGaFACCGExI7uE0LyQsMCiThsRRT9\nPPzMQPUbuZWrh09VHgssLrTnCSGExBbU/uiGuk9I7BGXA1NilJGeni5paWmy/s/dkpKaGunhEB+g\n5RCqBqPAD3LxEDYHCzfEKCGBNq9oXUgitxJhkNHisUAqBCMWSKyQnp4hF5doJKdPn5ZUm+krtd8Z\nUPtjCyt0n7pLiD21n3dvESTWrfXBtiKy03zZaSxOAnNV/qKKkR4GIYREBGpHcNpvp/my01icBHWf\nkNiBxRsjZK1/68WX5apqtaRZgyvUbzzG87HaightiPC7QY82Klwer9txvuw0FhJ52F6SEKIHakdw\n2m+n+bLTWGId9w5MhBD7wYgFB3jqY60VEXLwETanebbtNF92GguJHO4LG6ZBEEICQe0ITvvtNF92\nGkusQs0lxBnQsBBhaz2AxR6lLmCtv+/Jx2IixM69FRHO31crIjvNl53GQuzRXpIQQgJB7QhO++00\nX3YaS6xCzSXEOTAVwkbW+tMnTylrfSy1IvppzAxZP2We/LN1p/qNVkSde3V3CbWd5stOYyGRhUYF\nQoheqB3Bab+d5stOY4llqLmEOANGLNjUUx8L6GlFZKf5stNYCCGEOANqR3Dab6f5stNYCCHE7tCw\nECFrPfLzEEoHqzcECtZ6CGsshdShpSTyExFK6KsVkZ3my05jIeGHBaMIIUagdgSn/XaaLzuNhRBC\n7E5cDr4powy797LWejjDWo9QOli9EQIIaz0El9h3vuw0FhJ+4wHDMUmso7eXdSSws/ZTO5w7X3Ya\nS6zqMrWXEGdoPw0LEQRFgXx56om958tOYyGhw4rThOiDhoXQoHY4d77sNJZYgYUbCXGW9tPUGkEg\nTFpLReKs+bLTWIhxaFAghIQTaodz58tOY4l2aFAgxJnQsECiAngSUL1ZFVpyoCfB6eN3GjQoiJw+\nnS7//HVASpUuKWlp9gobJ4SQaNdNp4/fCmhQsBbqPrEatpskjga5j2+9+LJcVa2WNGtwhfqNx3je\nCTh9/E4GOZuxmLeJz9aQ59+R2hWvkytq36p+4zE/c4QQJ+B03XT6+K0mFnXZaqj7JFwwYoE42so/\n9I23ZdywkdKwRxvVZxotoVC9GaDqtN1BQSgnj584j9cHfyCjPpggg5rVl+sql5avd/4lL7w/Qb32\nzOBHIz08QgjJA3WfEONQ90m4CLp446lTp+SHH36QTZs2yZEjR6RgwYJSrVo1ueaaa6R48eJiB+xc\nGZqEXpl58qixkn7qtKSkpcq5rHPSoGcbqdvpdtf71k+ZJ5snz5Pvft5k6/BCLJTgqajT5U5Hjt+p\nxHKoJcIgEaHw3I215IkmtV3Pv7Fiowz5cots3PM10yJIWIo3nj17VlavXi3r1q2TgwcPqn1VqlRJ\nrSXKli1raJ/U/uiDuh9bsAOE+VD3STi1X3cqxC+//CI9evSQUqVKSYcOHWTatGlqUTBnzhy5//77\npXTp0tKqVSv57rvvTDkBQnx593Ej3mr4YLmkVVPJPHtWefrdQZ9ptIRC9WY7g6gLGEicOn4nLlhi\n2agAUFPh1Ol0FangzvVVysjJU+ly4O+DERsbiQ32798vjz32mJQpU0ZatGghEydOlJUrV8rChQtl\n4MCBUq5cObnxxhtl7ty5kR4qsQHUfUJCg7pPwokuwwIE/qabbpIKFSrIjz/+KIcOHVJRC4sXL1aG\nhN9//1127dolN9xwg/Tp00deffVV60dOYgp49xGpgJQBePdL1awsDbvdJfmSElX6gDt/rt+q+kyj\nJZSdUQWb8qc5dvxOwt2gEKtGBYBCjfnTUlX6gztf7dgvBfKnSskLSkRsbCT6Wb9+vTRq1Ej9H+uH\nY8eOqTXFkiVL5Ouvv1briL///lvuvvtuee2116Rnz56RHjKJINR9QkKHuk9sV2OhQYMGKmIhJSXF\n53vKly8vjzzyiDz88MNqcUCI1d79hJRkqdTkCvlx1DRBRg88/bgpXzt2pnS/r6/t0wgwvnt691A1\nFZw4fqcRywYFDXR/6NH/blVTAZ85RCrAqDBo6Xrp81BXpkEQS7nwwgtl69atUrhwYZ/vKVmypPTq\n1Uv97Nixg1ckhqHuExI61H1iO8MCQhP1EhcXJ5UrVw5lTIT49e4jWkGjSMUyEh8XL5smzZPVw6cq\nTz9uyh9+ZqAjZlEb58ejxzly/MR5DHj+QfV7yPAp8tSCNSpSAUYF7XlCrAJGg2CoUqWKZWMh9oe6\nT4g5UPeJbYs3ehZKQr5kdna26zl4IoJdPJgNCzhFJ2jHBO9+gx5t8nj373vyMVWTAOkDTvT0I+TT\nyeO3OywI5b2gE2oqIP0BHg1Cwlm80bOQ459//pmr3R4KMAdbxJHaH31Q92MLarW1UPeJ1dpvyLCA\nTfr16yejR4/OZVQAffv2leHDh0sk4eIiuqtDw7uP4obw7nfu1V159xMSEiLW+oqGAPvPTawvVrCY\nQAEn5FrSiEDsZFh4+eWX1c+ZM2dyPd+0aVNZvnx5kGNjR6hog7ofW7of61ptJtR94hjDAgotIf9x\n3rx5cskll6j0Bw3c4IX7Js8TLi6im0h69z1bXyE9A3USImHcsBt2nZtY7gSBa4L+1WOHTVHdIPL/\nW2MBYZGx/nklkTcs7Ny5U+rWrSvz58+Xxo0bS758+Vyv4f+JiYlBjo2GhWiFuh/9uh/LWm0m1H0S\nSe03tLJE1eYmTZpIvXr1QhkjIYaAMaH8RRUj2voK3SlQSBI1H5CeAR5/4VnHePDtMjdWwkWKKKPC\nqA8myKBm9VWLSXSDQOFG8MzgR8N+TQjxXEvUqVNHdZQiJJp0P1a03wzd17Qa0KgQOtR9EkkMRSzA\ny9CqVStZu3Zt0B6FcECvBbHkc3X6tFxVrZbU6XKnanmpsX7KPNk8eZ589/OmXIsHu3rw7TA3VkKD\nwn9hkLUrXifP3VhLnmhS2zU/b6zYKEO+3CIb93zNtAgS0YiFU6dOSe3atVXLyeLFi4d8Naj9xA7a\nFivaH6ru06BgPtR94siIBXR96NixozRs2FBuuukmSUpKcr12xRVXKKMDIXYkFA+Ct9ZXAIUk0dEB\n6RnuHhW7efCtJNi5sRp6PUTVVED6AyIV3EGLSXSDQOHGiheXD9s1IcST/Pnzy6uvviqNGjWSli1b\nqoKNGlWrVpWePXty0oijdD+WtF/P3BQpm7t2iifUanOh7pNIY8iwgOrNb7zxhmoFhVBG9xoLR44c\nMXN8hJiCGR4EX62v0J0ChSRR88F9IYNjYWGhWfKxDQKEUHwSXSyiKTQymLkh4QGFGlFTAekPl1X4\nb/6/2rFftZhENwhCIgkKNg4cOFBSUlLkwIEDuWosFClSJKJjI84n3Lofa9rvb25SC6RJSpFjIpJC\n40EYoe4TRxoWPv/8c6lfv7588cUXIR1879698s0336jFxDXXXCPly5fPE9a4YMECOXjwoIqEYE2H\n6MeqnMRQPAjuY8KiBNthkeDZ8tJ9vHbz4FsNzl3v3JDwVHBO+7dQI2oq4JogUgFGhUFL10ufh7oa\nOgarTBMzQTolHBObNm2KqhBxYg/tD7fux5r2+9P9zn3bS0oajQrRoPtWjJNEL4aUvGTJklKqVKmQ\nDvz444+rrhIwGJw4cUKFPA4dOlS6deumXocxAcYGhErWqFFDeTUGDBggTz31VEjHJfbEypxEox4E\nb2O6u2c36dqvj3wydrxaJMBjgcUFxhnrHnxtDjCn/uaGhK+CM/YBhgyfotIfEKmAxYX2vF5YZZpY\nAdYSRYsWpVEhhrFK+yOh+7Go/f50H9fvt5Ps9OBU3bd6nCQ6MfSpuPbaa9WN/tdffy3XXXedoQOj\nNsP//vc/V+jjW2+9JY899pjLsDBkyBApWLCgfP/996pA5LJly+TWW2+Vzp07S4UKFQwdk9gXK3MS\njXoQvI1pwvBRSjRRlMhfy8tY9OBDZHCtsGCLVDtQ92JQTsHKCs64JtjHowP7qZoKSH8w4m1glWli\nBajXVKhQIZk6daq0b99e4uPjOdExhlXaHwndj0XtD6T7BZOqKV3+7eRh9Zg1FZyj+1aPk0QnhrpC\njBs3Tu677z7JyMhQlSHdizf26NFD3n777aAHMm3aNOnevbscP35cGRuqVasm/fv3l0cf/e+De+GF\nF8qgQYOkd+/eubbNzMxUVjX3FApUmF7/525JCbFqNXF+RwEj+zdjTJrnA5b80ydPKUt+517do64y\ntB2IdHVpo2GCZlZwtipUMZQxMnwy+jCzK8RXX30lt912m+oOkZycrGotaKAF5Zw5c/xuT+13NlZq\nf6R0H1D77anTVmBE48zu3GA37afux7b2G3IPtGjRQkUQoD7C0qVLZf78+a6f+++/31ABJxgjOnXq\n5IpgQP0Fz8gE1GDA854gugHVpLUfM9pWkfDhz7OAG3JYwUNB8yD8NGaGWiD8s3Wn+q3yAHt197pQ\nMGNMmiUfi5Fl61ap33hMo4K5CxX39pLhXqxgATnk+XeU+F5R+1b1G4/dDZ1GKzifPJWuvA1Wj8GK\nMVo9JhIdoG7S4sWL1Vpi+fLludYSL7/8csDtqf3Oxkrtj5TuA2q/dxC9gB+ACAYtisGJhKJxZuh+\nqGPQQ7DjpO4TYMhtWrp0afVjBvA4wKCAEMj33nvP9TwCKdwrRAM89hZg8cwzz6j6C54RC8QZhZfC\nkZMYbP6/mWPCfERLsSY7EknPR6hhgmZUcLY6VFF5QVJTZPbGPVKrdFFJS0oIOEa7h0/So2IPChcu\nrGopGYXaH3nsrP2R1H1A7feOZlxwYuqiGRpnVucGu2k/dZ+AkOKxcQO/f/9+yc7OzrVQQEEmPSCV\nok2bNioMEpEPKNSoUbZsWfnjjz/ytLnE856gBgN+iDMLL4UjJzHY/P9Yy5Mkxm5OUdAIoq6FCWKR\ngM8LiiYhvzFQWGKoFZzNGEOgv+93Xhsu585lyRtfbJIPvtkq7epdLJeULCwvLd/gdYxWjykUWIjK\nnpw9e1bpu7unDdGH3vTeHWp/5HCC9lP3iRWEqnFmdG6wm/ZT90lIhgV8cPv16yejR4/OZVQAffv2\nleHDhwfcx8mTJ+WOO+5Q9RkWLVqUJ18DxR1nzZrlSq346aefZM+ePdK0aVMjQyY2L7wUro4CwXgQ\norHLgVXtPCNBpL0d/sIEUYkZYYIVL87dQtdIBWd/3nWzxuALzQPxoptH5LmFP4nki5d+PqpMWz2m\nULC7RyUWQcoDfpAS6Q60HukRxJ44SftjXfftrP1Ih3BavQUzNE5P5wYnaT91n4RUvHHJkiXSq1cv\n1S7ykksuUX2o3S3EeqzVWDTAWPD000/nKv6I/RYoUEAZES677DK58sorpU6dOqpg5J133ikfffSR\nrkgKeDtYvNF5hZewz0h1FHDSmOzS0ivSN/eRXJBYUYDJvYKzHu+62WMw4/ysHFMo2HVcsVy8cefO\nnVK3bl1VU6Fx48a50h/x/2AjEan94SEWtN9u47FjK2+zcK+T5ATMLrrs2bnBidpvV32167iiWfsN\nfav8/fff0qRJE1V4yShXX3211K5dW4U/unPu3Dn1+6KLLpKNGzfKpEmT5NChQ/Lhhx9Kq1atDB+P\nmI/Rdk5Oy0m045gi3dIrGqs7B4sZ4Yye+3P3MOjxrps9BjM8EFaOKRTs7FGJVbCWgOMAHSCIc4gF\n7bfbeOzYytss/ivo6AwDg5ka56n7TtV+6j4JybAAo8Abb7yhCi8arW0wePDggO9Bgcgnn3zS0P6J\n9YSj6KITQvnsDuYN3gosLDTvEq4XxAjhnqg7oXc+aVDIHZ6oJ5zRCIcOHpExQz/WlT9p1RhCKTBl\n1ZhCwayCWcQ8EK3w119/KecBCy47h3BqP3XfHtofLgNDpKMgI6n92P++3b/J2GHO1H7qPjFsWKhc\nubJ07NhRGjZsqGohuKcyXHHFFYwsiBHCVeDQCaF80e5dokHBf3gixN4znDGUY4z+aLKkZ5zR5THA\n3wC8GGaNwQwPhFVjCgW7elRiGRRsfvXVV6VRo0bSsmVLlcKoUbVqVenZs2dEx0cip/3UfXtGlsQi\nVmu/5/6BE7Wfuk/U58DINCB9ARELVapUUaGM7jUWjhw5wpm1KVZY/sNR6MgJoXx29sCY5V2ye3ii\n1WjhiU83rSPVSxWR7f8clSFu4YlmhNG7H+O1zzcG5THwFlIZaqvFUD0QescULuzoUYllULBx4MCB\nkpKSIgcOHMhVY6FIkSIRHVu0YJXeWK391H3nRZVGK1Zrv3vqw2UVSsitI5c6Wvup+7GNoeKNqHsw\nduxY+eKLL8SOsIBT+C3/VhU6sqJIVKSIpAfmrRdfVsaYBj3a5PEu6THOIGIhlg0LEORaFa6Vy8oW\nkTW/H5STZ7KkQHKCNCxXQtbsPyqb9n4TsqfAs8jQ0wvWyHtfb5FBt9TP4zEw0sEglFaL3gpMOZlo\nOx+nFm/84YcfpHPnzvLLL7+Y8h1I7Q+/3lih/dR9+2h/uLHbWsNq7fdWXBDa/+5Xm+XF5g2o/SZC\n3bdx8caSJUtKqVKlQhkfCaOnwQzLf6CxWFXoyKmhfN7mK5IemFC8S8HmPAZjFXcKOJ/T6Rmycu8B\nGdS8wX8FlRb/JGeysk0p/OdZMGlw8waSeS5bBi9dJ+kL1kj+tJSQvOuhtFq0mwciVKLtfJwK1hJF\nixZlSpsF2h8O3bdK+6n75hGu9plmfGbtWF/Bau33Vihx4I115Jd/jskzC9dIdo6EHFlH7T8PdT88\nGDIsXHvttSp88euvv5brrrvO/FHFOGZ6GkIt3hPpPEenhfL5mq8+jz4c0SJKuFZYTOI4er1LwdZV\nCMUjbncKFioo+eLj1MLCs6DSc4t+kgIFC4R8DPeCSQ0uLC7PL14ro1Zul/TMc+rYnbq2MTyXMPbg\nuugpCEVIuEC9pkKFCsnUqVOlffv2Eh8fH9OTb5beUvfDi11136j2B3PTj3Mf/toYmT5mppw+nSFp\naSnSrmcb6TewpyGtslO0Qji035vuD/1uq4qMwHHv7tJaBr32pBQqVNDQ/qn9JNwYWu1PmzZNfv75\nZ7n++utVOIR78cYePXrI22+/beYYYw4zPdt6Lf++rM2RznMMV4FIs/A1X8ePHrNF5IUe75LRQo2h\nWMXtzonjJ+Rcdo7XgkpZ2Tly8sRJKV6iqGkFkxZt+01WeXpIRn8qqWkphuaSrRaJHYFz4scff5Qv\nv/xSunXrpmotaKAF5Zw5cySWMEtvqfvhxe66H0pkibYe8LUWGPL8O/LpiE9z6/7wT6VQkjGtijXt\nD6j7U+aq/RudS2o/cYRhoUWLFrJs2TKvr5UpUybUMcU0ZrcHCuTxL1KsqMrB8+YhyTx7NuLW9nCG\n8ll57eZOmumoyItgvQbRbhUPV6tCRCRkZmbJqA8nypBbG5k2l2y1SOxIvXr1ZPHixV5fi7XijWZq\nP3U/fEST7gdb7yDadT9c2mmV7odr/ISEbFgoXbq0+iHmY3ZuYSCP/8h33vfpIWnX5R5bWNvNCuWL\n9LXr0L2LzBzziSMiL4Il2q3i4WpViM96t94dZPj7E0ydS7ZaJHakcOHCcs0110R6GFGn/dT98EHd\nj17dD5d2WqX74Ro/IUEbFtasWaNSHi699FJdrSg3b94szZo107NrEoaaAr48/sj/u65mXZ8eku4P\n9LeVtd2qApHhunZPDHpWChUuZOvIC6PFk2LBKh6uVoVG5zJQ0Uy2WiSRZseOHapF9dVXXx3wvceO\nHVOdp1q1ahWWsUWb9lP3w0M06L7RtUAs6H64tDOUuaT2E8cZFnCjeccdd6hCSx07dlSLgipVqqie\n03jtt99+k1WrVsmMGTPkm2++kXfffdf6kUcpVtQU8OXx37d7j18PyakTJx1V38Du165goUK2jbww\nWlchlqzi+DtCniNCEq1sVRjsXOotmhmu8RPiC3wG77//fuWouPvuu1Xx5xo1arjqNP3111/KkYHa\nCvPnz5cBAwbEzGSarf3U/fDgZN0PdS0QC7ofLu00MpfUfmJH4nLwCdbBWeTbT54sY8eOlZUrV0p2\ndrakpaWpvtH4f61ataRLly7Sq1cv1T4qkji9l7VWYRgW7tMnTykLd+de3S3pP31VtVpSp8udrogF\nsH7KPNk8eZ589/MmSUxKCstY7Nim08g+/V071Kwwe1yRNih4E7lxw6fIyVPpysrevV90dIUIN8HM\nJYpn5SmauWSdWoxEQ/Es4sxe1r7AemH27NkyatQoVbTxzJkzkj9/fsnIyJBz585JpUqVlNGhX79+\ncuGFFwY5Nmp/wDmi7puu/U7TfaOFGr1B3TePYOeS2k/sqP26DQvunDx5UrZt2yZHjhyRAgUKSLVq\n1aRECfuEPDl9ceEuaFZbuFG4EZb2Bj3a5LG0u1ehDsdYQsWK1phG9+k+X5pxJlItO0MpzmQEhOXR\nIx6eucTrtSteJ8/dWMtV8Am8sWKjDPlyi2zc83XUeI1IdBgW3IFRYcuWLXLw4EG1LxgVgjUm5B4b\ntV8P1H1rtN8pum/FeoC6bx565pLaT+yq/Ya+3WBMaNy4scQqVnjEI1VTQG/HhXCMJdR5DaVVl9nt\nNt3nSy3iTGzZGa7PnxF+O3lY/Y4vmV8OZafLoZPptutL7SSwqNAKNnnLowylaGagvExCrCY5OVka\nNGjgiIkO5/eu1XpL3bdG+63UfX9jtZtWEfPm0pdOG9V+6j6xmnjLjxBFwIINsUD6QLMGV6jfeIzn\nnYqWh4m0h2XrVqnfeBxOi7oZ8+rZ8glFlPAbkRgwmuD1YI/tb5+T/ezTjHFZNU++jAGaQcCMfRRM\nqub6IaGD64uQR0QmXFH7VvUbj/G8e8End5CbmRAfJ+NGTs3z+fC3P0JI3r8/6n506b7V2m+m7ps1\nV4EiFYwWbibWEEin/Wl/SlKiFCueu10vdZ+EC3vGY9mUUDzidieSHRfMmFejrbr8HduMdptmthCz\n4vOn3fxjURGqcYGGBGtAzmWeGgrvT1CvoYaC14JPS9bKNRdfIOOGfSyJiecLT+ndHyHkP6j70af7\nVmu/2W3DrfoMmlljiZhLIJ3Wij0+/974XNr/wuK1ci4nR95/czR1n0QEGhZ04mmB9mzNiIq/Voam\n2TEEzowxmTWvgVo+5S9YQHXBcB+rv2NPGjVWWtzVSuLi473uMy5fvNpnuFqIWf35s8ookHE6Q/b8\ns88VxscwvODAfKHbAxYXWg0FtKLCdUfrK1SpRmGnzMwsee7DiZKVnSMFkxPl4etqyeDmDeSdrza7\n3qfNf6D9MS2CEHvovh213+m6Hw7tN7N1qFWfQSOFGoPBXesB0+6Cmzs9Ov3gE71k+HsTZPDSdZK+\nYI3S/keuryWFkhPlNeo+iRBMhdCJPws0qgDDAh0rYZhmjsmMecVxh77xtqq+/OOoaaqrxT9bd6rf\nP42ZIVWrXyJN6zbONVZUIH/tmRd8Hjvj1Gm56/qbJSc7O88+14yZITnnslU7Tr2tqDAO932gQCaq\nRutdEITj86eFQ+r50fsZubnWnSqMr2bF6+TWJncz/D5I/OVRomo08iiRttStdwdlVJje7Ub5Y1An\neeW2RpKQLz7X+/TujxASWd23o/bbTfcBCiRWq1Fdt+5jrPixWvvN0n0z58obVhgVML8vP/+OXHrR\n9Ur7q5W9UmpceCXT7oJAr04fPnhEMrOyZEHvZvLL021d2n9jtbLUfeKsiAVYzdq0aSOPPfaYXHPN\nNeq533//XTp37iwLFixQxR2jDTMt0HYMwwzGC2HmmMyYV4xn7NARcmmbW+TUwSPy09iZsvrsVElM\nTpJadevIlk2b84x19bffy+aNmyRfUqLXYyekJEu9e+6Qn8bPktTiRdQCAwuKxLQUuaBWVTn6y17d\n11xvoSyr58kszwVSJrRtfEU6eH5GVo34RDb9tFleatGQ4fdB4J5HCY+FxrKff5e01GQp8K/nTHvf\nzoMnJK32f1/rCI1EyypUl/a3P7wvf1qKa3+EhItHHnlEFYPG+kHrOnXbbbep9tbly5ePSd0Pl/Y7\nWffBW4Nekk0bNkqpS6vI2vGzZXXGVBVRULJkSa+6rzFjylTLtd8M3TdzrsLFa4M/kBEfTlL1JA7v\n+k32fv6DvHhLA+p+EATS6dOnM1RUg/a+1fsOynWVyxjSfbwv/XS62h+jFUnEDAvoPX38+HGXUQGU\nK1dOGjVqJGPHjpWHHnpIog3NAg1xgmHFszWjFSGK4QjDDLalktljCnVeTxw/LuM+Gi4SFycbpy6Q\nxNQUubT1zZITJ/Lr3C/kl23bvY4Vi4WG3dtIVnrGeS+E27HXjJ4htds1lwZdWkl8Qj5ZN3GOWmhs\n/GSh1GrXXDZ9siCoa64VyMTcGG3ZadXnz0h7Ke39vgwMnp+RrIwzcnj7LmVUYPh9cGh5lFoNhasr\nXSDPLfxJvt39t5zLzpHLajRTryMdwmuthaXrpc9DXV0LBs/9ae97fvFayTyXnWt/dm2JRqKHffv2\nyZw5c+Tdd991PQfHRMeOHeWNN96Q999/P+Z0Pxza73TdV9ETg16W8cNHqZv+Q7/slZp3NpXqd9wo\nvy79VtZPniuX9emQZ6yTR42TnJxsadSzrZw5ftJS7TdD982YK39odZXMilzAzemY4Z8oowLWYR/f\n2V+l5FH3g8ObTq/45U95YclaSUjIJ00ua60MBXhPt74d5YUPJxnS/ReWrJMciZMbLrvLtT9qPwkV\nQyvHv//+W4oXL57neTz3xx9/SLRilgVaL2YXADLDC2HFmPzNKxY0v+3ZKyJxUv6iCnlE9M0XXpLs\nnBxp3Kuda/xYLFRqcoVknE5X7/E2VixGila8UCpefb7VGRYQODbyKnMkB3E5uc6rRJWL5NzZs7Jt\n+hLD1zzUApnh/vwFY2A4cHSjHPznkFxU7vI8n5HTh47KmfQzhloixgKB6k5A6AFyK08uWCNJ+eJl\nyK2N8niA3N+HeYUnokOX1ioP09f+8L588XFy7cUXyOBbG8r3u/9hIUcSNv755x/bryUi8b1rtfY7\nQfdxzAKFCsrJ4ye8toJELQQYD9x1H8aAktUrK333NVaA10pUuygs2m9GYWwrPoPeCjcHa2Dw1C78\nX/ucUPeDnz9/Oo0izAnx8fKiRzHHXg/cq4wIweo+Okecy86Wwc0bSpOqZVjEmZhGXA7MV0Gyfft2\nufrqq2Xjxo1y4YUXqueQs3755ZfLwIEDpVOnThJJ0tMR0pMm6//cLSmp5vdnh+iFYoEO5jjIDazT\n5U6X5R0gV2/z5HmqNWQoxzeyfyvH5D6vyJ1856VXZeLwUap2AkQ/X7546dq/jzz63NPKG+BvLEiH\nSExIkLi4OK+vaxEL8EwAeNXXjJ8lW2cvk7p3t5SNn8yXe+cMlc2zlqqFR51Ot8uWqQtlxYYfpZiX\nhXA4MfPzZyRiwdNzhOrFKDSEnMC0tBRp3eVOmTlxrtTt2soVsQDPxYtN67g8F+CNFRtlyJdbZOOe\nr2MyBM9z7gJ5DA4dPCKNq98szzet7XceTxw/IS89+7bMmDJPTqdn+Nwv9tfokpvkqSaXytM31fO5\nP0LcSU/PkItLNJLTp09Laoj6irSHsmXLqijIBg3OG3mzs7PlzjvvlMsuu0yee+4522h/uHQ/HDpr\nZ913j6RAWgOMBClpqXJvn57qRhrrAV9jgVYjssAzYkF7fdOkeSpiQdOmWNb+UFIifWkXbmbrVmmq\nro0WsUDdD037YXz4bc/vcuv1neS5G31r/7msLF26j/3tw/6u6xRwLUGIEe03FLFQvXp1adeundSp\nU0ctAJKTk2XJkiVSpkwZad++vUQ74WrNaHUYZiAvxO9790nVGtWDHpNn3qbePE73eUWhpfHDRkqj\nXm1dHokfR0+XsR8Ol/j4eOVV8Tv+s1OlbeeOUqhwYa9jrVO/nqyfOEd5OLTnt0xfLJfe1UzKN66j\nii5hsbF5+mKVV4n3tu3cKdcfU6SqdUeyNaiulkhjZkrVOtXUHGrzXqx6JRXC7ytcL5q7Rfg6t2Db\nPsJggAVDoMgPtJmaNml2wP1if+kZZ+Smahf63R8hVoG0BxgPkFbZunVrKVq0qHz99dfK4PDxxx/b\nauLD+b1rpfZbpfuemqgdK1jd94ykgO4XveQi3a0gN06Zr/TdXX/cxwo8z4Pafz6CQU9h5kDa1bNf\nJxn+wUQ1vxWuayzPLfreb5h+tGq/v/MKRvuxbXJKspw67V/7J4+boUv3sb+UlGRdawlCwhaxoDF1\n6lRZvny5nDlzRkUr9OnTR5KSkiTSWB2xEE60XEiEwKECMELgUFXYVy5kMPjzQvw4cpr6MtO8BO7H\n8jWm+wc8IR+9/qbL2wAvwyU1a6g6B3ryOPWMC5EISYmJ8v0vm9VzPj0X42bJ979skdS0VJ9jfe/l\n12T88JHKI6J5RlC4qWSNSrJ//XZkX4hk50hCIiIf4pWnBOdwd89uKjVjyphxQZ2XHQklYgHCWbvi\ndfLcjbW8Wr0792wvk8ZOl9MnMUepUuPSqvLLll9VtWKE63Xvd7c8/nR/eeuVYbq99tHilTh7NtPv\n3HnzGASab2wD9O5Xz/6iaaFH7BexoLF06VJVa+HIkSPKYfHAAw9IwYIFDYyN2h9u3dfC8d0jDTw1\nM1Td1yIHtn26SJavX626PfiKROzWr7c8/OxTai3ia6zn6zPEtvYbXQ8E0o11O1bIe2+OlrEjPlHa\njznMFxcnZ85munRfC8sPJmIvWqIRjOhuoG1WbV2q6iPp3Se1n9guYkEDBZbwQ6zDrAJA3vDlhUCu\nYpWbr5Zilcp5zbv0NSZPb4PqBLB+gzTu3T6oKtKBIhHSz2a6cjr9eVEKFjq/MPU1fwNeHiRrvl8p\nmzduVGkR5RrVPu8hGTVNatS6VMbM/lTeG/K6zJwyVRUj0s5BRU3ky5crmsKKTh1ObokEq3ePPh1l\nwHMPKOs3qhNrkQnuj4c8/05QXnsn4c8r0blbG79z581j4KsAk7sHaM+ufbr3q2d/hISDZs2aqR9i\nvfabrfvAXfsP7/pddq34QRr2DK57RKBIimIXlVNGArR69KX73fr1UbquHcvX3FH7rdP9w4eOyrOD\nH5XHBvZzaf356/uf7oNo1f5A0QiB5s+I9iP6MJh9UvuJlRg2LPz1118yaNAg+eGHH+TWW2+VXr16\nyYwZM2TAgAHmjpBYGoapWfBRLRnijRZMte5qJpf17ajSBAJVfdYCXrx1Aji4fbcyKgRbRdpfeyWM\nDxELWoslvUWN3OfPvTDUjl9+8Vo9etPkeerx3GkzclXCxqILCzBUlTZaHTvUFIpIpWB4EqiFkbaI\n8BQ07TGMDLDqQ4CjrWp0oHPr91C3gHPnbZ8d7mklmZlZMmTMNFehJiwsNA+QnmvijreCj+77I8Rq\n0GFqyJAh8vnnn0vVqlVl+PDh6vHrr7+uauTEOlZovzfdv6TFdXJ5/06SlD9Nt+57aj/y6ifdeZ80\n6tXOVN1Hq8fDe353tVc0ovuR1n4zdNsO2q9XYzy13/3/0ar9es4rWI3WUioe+rcYozetRgRksOsJ\naj+xingjGyHcsEmTJnL27FmVAoHwxcqVK8u8efNk9erV5o+SWAa8EBDjW1vfoVo2njubKdvmrpA1\nY6ZLdtY55Q2AlwBW/1ytnl58WYUtNmtwhfr92jMv5OkEkJme4dX74Lk/Xx6VNaOnq/DGf7buVL8h\n6hgTXtNEVfOioHjUsnWr1G889hZK5znuJrUb+vSQpJ88Jb9u3Z7ndZwX5sjIeXmbNzzG83oIdXuz\ncVm9l6xTIXer9x1Qv2FFR7hjoIWBP8s90iVgZXcqgc7t5ImTuucO1xfeHYRCXlO/pUwaPU3u6dle\nvls/X4U4wguifd6DvSbYDttjP6s2L8qzP0KsBAvuNm3ayJYtW+SWW26RAwcOSOHChZXjAqmWxBo0\n3Zyy6LPzT8TFKd3/+K4HZfWIqVK69iW6dB+P//rjz1ydAELVfdRG8NT9EpdcLBsmfabSGfC+YHQ/\n0tpvhm6HU/u1DhFW6X40a7+e89I7f+66f0XtW6Ve5RvV82t3rMij1UauCbWfWIWh1eNXX30lJUqU\nkLFjxyrvwvr169XzCGWcPXu2quZM7Gtx9hwPjAJzps2Qy/t1zNW6CSQVzO/yEvhrVTVzzFSVS6d5\nG9KKF5HE1BSv3gfP/XkDxo7s7ByZOGKUrD5zvhVUvnz5pOsD/by2WMJcYp/+5tZz3L+v2aRqNngb\nY0r+NCl/ccU8HhScF7w7Xr0qyUl+zyvYFl9mb28FoVi9g7XcOwk95+Zv7twLP73z2vC8oZXDPlbt\np7yFjBq5Jp7eJULCwc6dO+Xnn3+WXbt2qc4Qq1atUs83b95cZs2aFfEOU9Gs+xjPrMmfqMjERm6a\nAu3/e+sOXboP/cnKzHLpJCIWQtV9MPnfSASt/sHRX/b6jUjA+ezbvceW2m+GbodL+7UCjoHaT4bq\n7Y5W7dd7Xr7mD1EJSGf0qft+UkWMXhNqP7GFYeHw4cNSsWLe0LzExEQVxUDyohU+0oobRbroj+d4\nIJhnjp9UvZ218D4Ir+TkSI/7++Wq+uye8gC096NgIqIMtLzHEtUvVvUKjFS1xpw8+eJz8sCAx1WV\nahRMKlexvNft9Mytr3H/8dPmPGPEOSAy4rYrrpNqNarnqS59LjMr7zbooa0qPnnH37zpDaMMZXt/\nYBFhtICjZvVGiJ9nDmUgojnPT++5ec5dUlKiR+GnFDl7NksGNaunO2Q0lGtCSDjBWgLtJj01EGsJ\nGJKdjt11HzfupetcIrXbtVAGBk1TtCKIenT/0wmTpGO3e2XSqLHq8UXXNVadHIzqvntthPwFC6ia\nCr7qS9hd+83QbSu135dxAWgGBm9rg1A1Jlq1X+95ec5f0eJF5IM3R6uoBOo+cTqGlK1evXry+OOP\nqxQI9y94RCs89dRTZo4varCbt9nbeLQoBdRX0Aolduh+by4vgd8CS2emStt7Osn8iXOUtwEiX7te\nXZWz6C8P0h8QTM/WV0bm1te4G/dqL3Pvf1E2Tprr8pBgoYXn/978i1po1KpbR/XqxuvwZsDYUunG\nK1SlajyH/M8KV9STPd+scRWVDLYwlft23rxbwWxvhYciEEat3lbl+dmhhZXec3OfO28FrdCmc/s/\nR4NuC0VPBLE7NWrUkO3bt8uOHTtyPT99+nRp3LixOB0n6D6MAEh9hO5rmoIogfbduri2C6Q/d3Rs\nL6dPnZa5k2b+2xUiUTkasCYwqvuanhUrXtyx2h+q7ge7DyP4azEZaD0QisZYof1O0n33+aPuE4l1\nw0LNmjXlzjvvlNq1a0uFChVUz2mkPyA3slWrVuaP0uGE2+IcynggmA26tlaW+NT8+WXgyy/m8qz4\nLbCUnCTzZ82RjNPpqtVkx25d5PFBz6pWTWZ3tAh2bn2N+69NP6uFz/wfvpIWl10jtTu3lAZdzn+G\nS9eupvaDhQVaXMFzAg8KWl0Vr1xBrnmkm6wa/on8uvQ7tbDAwmTK6HHqnD29Uf7mTQsR9ed90bO9\nWR6KUAwMwWK2dz1Qq6dwEuy5+Sv8NHjpOvmozVWSlpQQFSGjhAC0lHzmmWfU+qFBgwby66+/yg03\n3CD79++XcePGOXqSnKj7aLmoaUrpsmV0636nW25Xuo/3dOjeRZ74VwOt0n2naH+ouo99WaX97gaF\ncOm9VdpP3SfEPhheaaO2ArwKy5cvVz0tUcSxb9++rOLsBastzmaPZ834WbJ1xhKvoYu+WlUhZDA7\nOztXW8ZJo8aougvwHFh1fnrn1te4tRBN5ImeSc9QLSe97QcLC+0ctP3sW7le/tmyQ1WJ9nbOeubN\nPUTUs12np/cl0PahEkkDg1ne9UCtniKB3nPzV/gpfcEaeWnpOmld56KoCBklROPJJ59URgVEPJYs\nWVLq1KkjDz74oOTPn9/Rk+Q03f9t1UY59sdfXjUlGN2fOeYTKVS4kKW67xTtN0P39ezDKJEwKFih\n/dR9QuxDSC68du3ayW233abqKhQpUsS8UUUZVnqbzR4PiiT++tkKv6GLnq2eENmADMPGBlpMhXNu\n/bWoQlSFv/3AW6EVh8L7szIzZfzwUV7bVfk6Z3/H1+N9CdRiy19Io6fxwK4GhlCIRAsrf6GXwYZl\n+iv8lJKUKENX/SpvfLHJZ2ilHcJACTFC06ZN5ZprrlHRj8X9hL47CUfpfr54Wfbcu37TFuyk+07S\n/lB1H/vS214zFolU60qztD9U3Q/2eITY1rDw7bffykMPPSTr1q1Tjy+66CLVd7p9+/Zmji8qsNLi\nbPZ42ne9Rwa+PMjvmDwLLKWfTpc7rr5Byta/NOyemWDm1nPc7iGaeM2XRwZ5lgiBdA9TbNf1Xhn3\n0YigvFH+jv/nb7/r8r54214ZFP6tmerPAAAjAd6rx7igt4iTnfDn8Q9Uj8DM0EtgJB0jUOEnXyGj\ndgoDJSRYtm3bJvfff7/qNgXvNzpOPfvss/Lwww87ejKdpPvd+vWRu3t195u2YCfdd5L2m6H7/vYR\n64RT963QfqO6H2gs1H4SKQytOtFjumXLltK/f3+ZNGmSJCUlybJly6R79+5Svnx5ufLKK80fqQNx\nL8bjaXFGDYJ7/82jiwT+LODevpC8FRZyb/UUyHNgZbutYK357oWhAu0HC4stmzZ7abGVadgb5e34\nwXhftO1xw3/iX4NC8fhUJbCn41N9WqxhGAjUo9pfgUe7E84WVv5CL4HRdAx/hZ/wd+ltgWQ0DJRe\nDhJpEO2IqMfrr79e1qxZo+o0rVy5UhkVYGDo3LmzRJPu4zu+zd0dpc+jD8eE7vvaRzjOx27aH6ru\n+9qH3VqZhptwt660QvuN6H6gsfg6HnU/djh+JN3U/WWkZ+h6X1wOTGRBMm3aNBkzZowsWbIkT64k\njAxDhgyRSJKeni5paWmy/s/dkpIa/rAgX8V47h/whLz38mvyybgJrkJHkWw9pYmSPwu43nZZb7zw\nkowfNkIa9myby+LfpW9vSUjIF5Z2W4HOJdj9aMWa6nS50xWmCNZPmaeKOnXoihZbY1R+qae3xEjF\nb5VrOXREwP253+SXSSmk22KtGRX0Riy4g2PaPWIBoLryyPchtPXyWP7NqrEAYa5d8Tp57sZartBL\n8MaKjTLkyy2SnZ0jzzet7fW1jXu+1hWqiGPoLfjobyzejkcvBwmF9PQMubhEI1VbKTVEfV27dq2K\nckTRxri4/1r2jRgxQlasWCGffvqpY7Tfn1ZCU94c9LLMnTbDFm0nrdb9bv37yqPPPRW2VptO1n69\nuu/rGk0ZOcal/Xf36al7fp2i6XbR/XBov17d1zMWz+NR92PToFA0+SLT9gt9RcOGQNpv6NsdXgVv\nhZUKFCige6GBwo/vvfee6/GMGTOkdOncoUyffPKJKuh07NgxqV69ujz66KMq5cLu+GqBtPrb771a\nwENtPWXEYu2+jb9wRf3tsnJUGOtPY2eqNpX5khIl+1y2/PTDStm6eUtY2m358kYEO1/afpBX6S9M\nsUO3e1WxJrO8UcHUUNAWBN7aFHlarN2jFBKyysm+P/ZErXfDqvaVwYReglDDMs0o+OjreHYsdEVi\nE6wlYAhwNypoawm85iT8aSWYOWWq7bTf1zah6j6eD2erzUhrfygRqEbrJ2B+Jw0dnvt7/KPhPufX\n89zhYPjt5Pk1hdMNDOHQ/XBofzCFLIPVfup+bEUkFDXRoBAshiIWYK249tpr5ZFHHpGOHTtKvnz5\n5Msvv1SpEUuXLpWKFQN/wf/555+ya9cu2bdvnwp33L17dy6jAbpODBgwQN588021P0RIfPfddyof\nEy2q7Oq1wJf3VdVq5bF0/zRhtqwdPytXwR93C/h3P28K+kZPr1fB6Da+zsVzzO7vu7T1zXL68DFJ\nK1ZYNs9aKj+OmiYNu7dxtXEK9ZxDIdj50nv+J44fN9Ub5el98dUWKpDFetHmzyQlLUU9lxpfKSTv\nkdO8G8FY/u0YsWDWWDyPZyTCgRCrIhaw/EAqxHXXXadqNmF/qNvUtWtXpfloQxkMkdJ+f1qxadJc\n9f9AOmIX7TdD9zdNmic5OdlSt2sr087ZbtqfmJQkbw162bQI1GCiLvDea6pe6uN7fKt888tm1z70\nnLu2xnCSxodb952s/dT96DQoFA2z8cDUiAX0k0a0gGduZJcuXaRbt27KsJCZmanSIBCF8Pbbbwfc\nZ9myZdXP9u3bvb6+YMECtbjo3bu3eoyFBxYMWHTg/3bFVwukYheVk5xz2boK/uj1QvjzCKDIj7d9\nBONF0NvOyf196INdqGwp1/twzkUrXhjwnM0g0Lx5O/exHw2X40ePycAhL+purelZHGrkO++b6o3y\nVkPBm+AHslgf/Oew1Kh2g3ouUEsrX4S7toJ7/h8wWunYrPaVRootAV+ved7kGzk/z+38jcVzv+Eu\ndEWIOyjQeOedd+Z6Djc+ixYtkqeffloSExPV2gJrildeeUXmzJnjiAkMpJUgHNqP1omdenbzur1e\n7TdD94M5Z6dqPzR10uixpuu+HnCu/rX/v/nVc921OkqRKtLsqWlGtdFK3Y+09ntuE4z2U/ejg+MR\nMigEiy7DQosWLaRq1aq6dlimTBkxg7p168o333yjFh2wqqKwU0pKilSrljc/HEYNvM/dqhIpfBXj\nObznd9XSyV+RnmCjCXy1KRo/bKRMGjkmjxUdbZX0tDYKtrBQoDZWR/b+IRdLY5/bh4qeefOcr+ys\nc7Lnmx9F4uLk0/GTZO70mV7nOlCYot52UcHgK0Ih2KJFJUoVMzxGvWMwC8/8v+SkBMnOFsnMyrJl\npWM9oZe+XjOa6+hru8ef7h9wLJEqdBWJkMBCRRlxYVfq1asn8+fP1/VePS2s7aL9/jQQz4OwaP/w\nkTLuo+F5tg9GA8zQfbSiRMSC1a02I6X9Vuh+MGDufX+Pp+UqnKl3nJEo0uxN06rXqirbNv0ip9Mz\nqP0B1gt6U0CiVfejkeM2TW8IBl2rdNQ+8Kx/YDUvvPCC9OrVS4VdwFjx+++/y7x587yOA8UiX3zx\nRbEDvizdGyZ9JnXq11OFjfxZwE2JJjgzVWp2bikXX9s41z7adblHlyciWI+9v/fhnNdPnCPxCfks\na7elxyLvOV9rxkyXTdMXS6Oebf3OdaA2T3q9O3oJJiTRn8W6Y78OrjSIYMcYibBIb/l/zy/6Se5t\nVEUuLV3UdjUA8LnAWHy1gvL3mtFcx0Db+WtLpRFshIMTixQdObJH/aaBwX6gbsI111xj2v7sov2B\ntBIE8n6bof2IELz5pUfk+J9/59o+GA0wQ/f1nLOTtV9vm0irwDhQqBE1FTy/x7vc3881zmCueyQ6\nP3nTtOcW/iRXX1xKXrm9sS3r/4Rb+wNto0f7o033o5HjFhVcdEyNBTNBKkSNGjXy1Fh44403ZOjQ\noWrRAGMCvBxz585VdRYuvPDCgF6L4sWLR7wrBCzCp0+eUpbuzr26q64QH73+Zp7ntWgCPXmNenIA\nUUip84z35eypdEkrXkTlPGIfy9ev9lvp2Fveo+e5wCNwR7s28sSLz0vBQgVDOmd/3lm9IaGHDx2S\nJrUbSu3OLf3WcfDMB510533SoFvrkPM/9eajBkMw0QKaNXvc8Cly8lS65M+fKm173CX9BvaUomk1\nDY1RTz0FM1sW+cv/e2X5BvljUCf56NutUVEDwGiuo5k5kp6fGXgsuvezV0SIGSJ85IxzDAxOagH2\nz59HpE7Va02psRAqdtJ+Xxqoeb59vWam9q+bOEfuHDpIpSVouo/tQTDHCFX3A52zr+8ZvbqP9/22\nZ6+0v/nWgHUcrNB+K3Q/WLS5/2TUWDl56rSKVOjkJVoj0DizEn53Pa/HmWDWd5Ue3U9LSoia+j9G\nNJy676x2jEbWGn//ecRlFCpbpLrYGUu7QoSD1157Td566y1VxwE0a9ZMPv/8c5kyZYpqa+kO8jLx\nYxf8WbrNsoD78hasGT1dCpcvI590eFQy0zMkMTVFLrqusRL2UydO6vJEeDuXPo8+5CpQ6C180Mg5\ne0NvSKj2PqR8nEnPkPWT50pWeoY06tnOFR3hPm/u85V+9LiaGzO8DXq9O8GgtYPUch79Cb5mPW//\nQAdVUwHpDyWL1LFsjFa0LAqU//fXidNRUwPAaK6jmTmSgTwu0WLVx+swLmjb2NHA4KQWYGYvwszA\nTtofKLrNeu2foboxzOg2MJfua9sHowFm6L6/cw5V97X3xcXHy+Fdv6v0Bui+t3mzQvut0P1gCTT3\ngcbZuW97l1FBj0HB7O8qPbpfqXihmNZ+6r6zahUEEy15+MAJefeN4fLxuE/l1OkMKYC/p1695amn\nnrKd9geLbUePwo4//PCDKuCIVlR79uxR6RB43in4Ksbj7Xm9eY3ueOYAqrzG7Gw59tt+ady7vSvM\n78fR0yUhMVHtw2hrI70FCoM5Z2/oLTDl7X0wqmCBccX9nb3Om3aOWJRgQWJW/qfROTXDwKA9j7QH\nrVCj0THqCYW0omWRv/y/gsmJUrpgmszcsCcqcgGN5jpakSNpdaErO4QJutIjbGpgcEILMM+5T8+2\nn4HBTvjTOqu0H3oGowJ0v2z9S/PovlGdClX3A71mhu7/OGq6pBYpqHTf17xZof1W6X6wBJrfPOvE\nAmnKqICoxmDSHc3+rtKj+9rjWNX+WNd9pxU/1JOOefzfMcCoMHHUFBnUrN5/f0/DhqrXnnvuOXEy\nEUuFQHeHBx98UIVWrF27Vho1aiTJyckycuRIqVmzpkp5QBvKc+fOSalSpVTKRJs2bVSHClSMtmu7\nyVBQeZZDR0iDHm3yWMD9VRnW2hTlL1hArq9ZXxr0/K9QjytEctwsWb1ru8uiHWxro3CE/Rlpc+X5\nPrS3vKBWVflnyw7pfl8/efLFvH+g2P61ZwfJzI8/CXquA41f75wawdeNv2aAMDpGvakXVrYsGvTU\nGzLmo0kyuHkDV/7fC4vXSof6laTmBUVcuYB2udkKhSHPvyMj38cCrV6eXEd/52d0Oydjdt6hlh5h\nBwOD3VuA+Zp7veGQkSAWtf/3vfuk9fU3S8NebQPqfjA65TTdr9TkCilWqZysHTdLuvXvGzbtN6L7\nRusZeNN6b/vy9r4DRze6ohqrlipri+8qb5r2/KK1ctXFpeTV2xtHncYZ0fBY1P1o6aZwxG294U7y\nuVJyaY3q8qy3NKAvt8rmrduUjsVUKsSKFStk7NixXl9r2rSpdO/ePeA+KlWqpNIdPClf/rwl7eqr\nr5adO3eqnxMnTkjFihWlRAlnWyytsoBrFut9u/eofE1fBR09CzQF09oolEJFevMmteMUKnuBZGWc\nUS2svB0nUPGqvzf/qn7Dg+MNjOG5/70ihQoX0j3Xes4hmDnVi/txC6bpNyD4wnOMwRRqtLJl0T3d\n28rw9yfIy8vWq32l/BvWOmnNDskXHyedu7f12uXAieit5GzWdk7EqkJGdopgsGsLsHAWkdqyZYu8\n+uqrXl+rVauWDBwYXu+vE7U/OSVFsjIzdel+MDoVbt0vWaOyHP/jb1UXCtofrO7v+mKl7FqhPRs+\n7Q9F9wMZ8t1rGSA6UdNrzw4O7vvx9T5ENV5Zq5atvqugXceOHpdnx82Qc9k5KlKhQfni8tNvB+Wq\n9+ZR+2NM963Ss0gVQyzq47ioKXjSz9/TP//8k6vmoNMwZFjInz+/lCtXzvUYUQWrV69WRoC2bdua\nVh0akQne2ktGK3py5vxhJKTSyv0G00IL7/1kzHjVnnLZc++qHNFL77pZ1UwIps1VYlqK1Ol0u6yf\n9JlMGTNeHhjwhN/aEYHmOphzMJNwHldvOKSVLYvKliut9j3g+hrSsX4lVxjkS0vXydBVv8qg1/7P\n8XlnodY4cHJtBLvd1NrBwGC3FmCRqEqNKEX3tQQCKLdu3SpffvmlXH/99RIrhKL9TtZ9ULR4MUlI\nTJBFT7wu585murQ/MX9q0Lq/8ZP56vcnYyfYWvu1G35vKY447muDP5Axwz9xHbdnv04y8N9aBu6G\nA2/arT0X6H12+K7C+UDbZ0yZJ/dfUVWea1ZfFWw8fTaL2h9Duh9rXHDBBaqmgtc0oPxpKkrfyRj6\ndrz88svVjyd33HGHFCz4X8VgYgyjFnCrCgoZ3a/evEntvZNGj5XL+nTIVRvi76075IBKawjc5grd\nMGq3ay7lG9dRbT3Tz2YG9KoEmutgzsFMInVcf1jZskjb95D3J0hifLxr3+9/uy1q2yEZzXWMxhzJ\nSLVaiqSBwS4twCLZ5qpKlSpeIxcfe+wxyc5G1FlsYUT7naz7Wh2HHInL1f4R2o8ohF4P3Re07he7\nqFyuopV21X7PGkrazT+MCiM+nKRSNbTjDv9gonrt2cGP6jYSmNku2mrt73lfZ3n//QlSLC2Z2h9D\nuh+LBYcB0hxQqBE1FTz/nvr2v8+WaRARq7Hw9ttvy8GDB+WVV16RSOLUPMtgON9uaZ8K+dMqHwdq\n/xSKlT3Y/QaTnxkod/KuTh1k4JDBedpcvTXoZRk/fKRagMBjgVZSiHDYOG2hWmwkJSbK979sNryw\n8mxTdfrQ0VztO61qKRWO3NZgWlqGq1VhNLRBdBJ2E1079G4Odw2GcH7m/V3vYOY+HDUW0Fp68uTJ\nMm3atKC2i3bt18LyCxQqKCePn3CF5ztR9wO9f+3YWfLFxjXqHPXqPtptImJh26eLQtLJcGu/1uIZ\n6Q+XXnS9j/mbK5t3fxUxIzu1nziFSBrLg/l7QhrguDGj5cSp0ypSoXvPXrbuCqFX+00zLKCfNIor\nIpLhmWeekUgSzYsLfBjfeekVmTBspJw7l626QCQmJUmXfr3l0ef++0BaVUhQ735R76FZgyuk1fDB\nucIW/9m6U+b0e16WrVvl8hYEei/wFYb4+rMvyIQRo6V+19bKY6Habo2Z4fJ2hOJZ0MZV9ZZrZc/X\nP+Zq3/nrkm9ynYOZBDN34TIoeIIFkFVheVbumzhDdGPNwGDlZ97swlVWGxawJHn44YdVbSUUaw52\nbNGo/e7tlTNOp6uUQWicpy46Sff1vD85NUW69O3lVfcnjhwj9bq0yqX7KNysRTg6Sfs1w8KeXfvk\nitq3+pyPVZsXRdxrTe0ndsWJa5vTp0+rmgpIf7B7pIKlxRs//fRTeeGFF3I99/fff6u6CejqQKwD\ni4uxHw2X+Ph4uazvf2kD44eNkPj4OJeYWlFIMJj9BpOf6e+9+ZISpcWbA+TAtp1ewxAfH/ScxOdL\nkIkjRqkwSLSSQm2Org/0C7ntE8aF3M9dK36QRr3aeW3faQVW5cyaGSZpZVgeQ/6swYmiGynCnSJh\nxWc+UpWwgwEtpT2LPR8+fFjVbUKRaJI7LL9kzcqq45F72oC7LjpJ9wO9HwUca3du6VP3ocGTR487\nr/v/GlqO/rLXlJaPkdB+pENkFkjwM39ptmi5SO0ndsPJa5u0tDRHF2o0tcaCe15kXFycFC9eXLWM\nTElJMXN8xMNrAI9FfL58amGhhcpBgODlmTxqnCpIZEV4vpX5mb7eCw9EnfYtpGy9GuoHryEk0/0c\n4cVAW6kHBjyu2m6JxEm5iuVNm4O4uHhp6Na+U5trtPGyoqUU8i6typkNVDSKRH96g9NEN9LYochj\nNBoU/NVYKFKkiDRo0EAKFSoUsXHZTfdRQLDevXfKho/nedV+T12MFMFqV6C6CQ26tJL4hHxedd+9\nACNabZ86cdLUSI1QtD9Y/msP+Yu079FaPh4xLc/89XuwCyP4SExjVjofsaFhAdaVaLOwOAHkViIM\nEgTTAkpv2yczcD9WMC20PN+LyIMqN1+tcif1nCPOq2qN6qaeC85Db/tOT/RWY9ZaSqF+xC9//S4l\nSmWotlBG24+FUjTKbDzbZcUyVs+Fky32TsAJBgYnGRQ0SpYsKa1atYr0MGyN1maxaMULVUi+HbXf\nqO4D7flJo8aq96tIhXbNXdofSPe154oVL24L7Q+lVga0+cnnX1fvmTZu9r/zl6aMCugK4QSo+5wP\ns3GitsUyugwLy5cvl9GjR+va4c033yw9e/YMdVzECxCflLRUyczK8hoql5o/d6hhOFsW+juWnhZa\n7h4IRB60u6mFFKtUTnkrrEgFCEdKgr8bdq0Q0pihH8vp9AzVrxl9nNHSqW3Pu6TfwJ7S56n20uXh\nO+TgP4elRKliyuBg5nXz7IVtFtq5jR02RfW+zv9vRelYLMJo9VzQoBBe7GhgcNKia/PmzfLyyy/r\nem/t2rUjXq/JDmg6dGTvHyrPX48ehUv7Q9V9d+3v/kB/aVK7oUp/QKSCv/OzGqvSEV3FMEeMzqP7\nd/fp6bo+Tw1+S3o91lWSTp1zTK0h6j7nI5a1jfyHLoVB7QSELLqzcOFCVXCiZcuWkpSUJJ9//rkc\nOXJE7r77bj27JAaAON/bp6eMfn/o+XZM7qGDY2ZIj/v75RLwcLYsDHQsf9Z9d28HSE5JkU7duygP\nhtmpAHrGEKi9lRnjwM3mqA8myBXlS8jKvQdkUPMGcl3l0qqv7QvDP5VCSSmqd7EUEKlaqqxru99O\n/uIROmk/tHMb1Kz+f+f0/gT1mjqnGMKquaBBIbLYwcDgxEUXCj55riVQZ2Ht2rVy1113qbXGypUr\nZePGjcpJQXLrUMlLq+TRfm96FC7tD0X38+huaqq06tBOZkz4RDkUwqH74dZ+VYBz6HDvuv/R8FzX\nB46E8qWck6pI3ed8xLK2kRC7QuzevVuaN28ua9askYIFz7cARM9pLASeeOIJadGihUSSaK0Mnbsr\nxChV4Ep1hUhOki5983aF8NXGadPkefK9ie2SjLZH9PR2oLtFTk62ZGVmKW9BtRrV5Zdt29VrZrXP\nCjQGT89OsO229KRAIFSwdsXrZMD1NeR/KzbKszfXlyea1Ha9/saKjTLkyy2ycc/XPj0VWo2EUAwM\netM1gkE7t+durBX0OUUbVs4FhJeiG5tdJCK96DKzKwT2VaNGDfn666/VPjV69+6tIhYeeuihoPcX\njdrvqVOurhAF8ss9HnoULu0PpS2yP+0///8cycrMtEz3rdB+PfN1TdVLA+j+Vlm0eY4yKjipBhJ1\nn/MRDdpGItgVYtWqVdK4cWOXUQGgSwGMDVggRNqwEM2cL1b4vDww4An5fe9vaM4l5SpWyCPgWl5m\nMPmYRjF6LG/eDnhj0OIJaRCIwri3d0+5u1d309tn6fW4eBaJ8jWOYG7SkWuPsPjqpYrIyTNZymPh\nzvVVyshTC9ao9nO+KsVrx9EiGLzhy+hghUHB89yMnFO0wbmIHcyOYIiVIlVbtmyR0qVL5zIqgFtv\nvVWmTJkStGEhWgmmWGG4tD+U4+jR/g7du8jAlwdZFqlglvYHM196dB+pj1fWqiVOgloXvvkIVKDZ\nF1bpkZlEk7bFMglGCy59//33cvLkSSlQoIB6DhbmZcuWKeMCsZ7zxQov8fk6WiWhAKK3HEF4O44e\nPmLazbqRfESt0jVEPU/V5Ylz5NrHu6v/Tx0/UVrf3SHkMfpKu/A1Bs8q1P7abWk9qPWCAn7Iqdz+\nz1EpkJygwiAvq/DfHH21Y78UyJ+qq7WUr+MiokEzILgXa/S3jRlo5xbKOUULTpoLWKDRMviCCy6w\nfS/laDYwxFqKC9YSW7duld9//13KlSvnen7JkiXqNSJBFSuEvh07clRpvC/tz+dWtygSdQj0aH/W\n2Uz57JPp8vAzA0yLrnBPd/A3hmC0Pxhw7LS0lIC6X7fSf3PpFJykdU6dj1C0wZ8e+Su2GWt6RCJo\nWLjhhhukUqVKcumll8odd9zhqrFw6tQp6dGjh0lDI0bQwvcmjhit0iQ88zHRwhEhlO2aNjetoBOE\n9+6e3WX8MI+WUWNmSLf+3vMRA3k7Th44LEd2/yEZ6Rlyx9VNQh6rt7DHO9q1CVtUhztp/xbwG/L+\nBLmiQgl5YfFPat5gzYbwDFq6Xvo81DWklIH/Ihr+MzCEI6xSOzfUETD7nJyGE+YCfxevvvqqjB49\nRk6fPiVpafmlV6+e8tRT/6VVEesNDLG6gKtYsaJ06NBB6tWrJ23atHHVWEAkw+rVqyM9PMfgqW/A\nl/a3uOwa6dK3V0R0X4/2//DRx/Lrkm8lK+OM3FinkaorZabuYx1x190dw6r9SoMTRNr1bCNDhn9q\nme5HCidonZV4fsf7m4+ufTpL1hmR42eCjwIwqg3e9Ah/G+++MVwmjZuuDG2qhlv3dvLIk/1y/a3F\nkh4RczCkKvny5ZPFixfLhAkT5Ntvv5XMzEy55557pF+/fq4IBhI+3K3xQ994W4X3oef1ukmfKSs8\nvAAQy8S0FLmgVlX5e/OvcutbA+XAtp0hF3TSjp2VeVbV2UD/6dVnp0q+pETJPpetUjW8FUfy5+3A\nOLfPXSG7Vvwgl/XtYErxKW9hjzPHTFWRHe5jwGJm11er1NisrEKNrgBg7LCP5UxWtjy36CfJys5R\n1mwIsfZ6qIQzR1OzfD/4RC/1eMjwKSrsz+xzchLaOdt1LmBUGD58hDRr1VcqV28gO7evlWHDzhcR\ne+655ySaCUeURiADQyQMCnaLThk1apSKdFy6dKns27dPmjZtKjNmzJBSpc5HlRHfaNr6yZhxMmn0\nOKVvJWtUloWPv6a03lP7/9m6U3VdMKOQI459/OhRpfN6dT+Q9qNo4y+LvpZGPdtapvvYn1bHSRsD\ndP/0oaPy248bTe1A4d51CVo85OUnVWFmq3U/3ED7O9zTSjIzs2TImGm21DqrawJ4fsd70/6+/e+L\nqNHeXWNe+t9LMn7U1FzaP27kcEnJV5jaH4WcDqPuGyreaHeitYCTHms8ei836NFG6nW+Q1aPmCqb\npi9WRoZiF5WTw3t+VwuNOu1byGV9O4ZU0MlbManSdS6RW4Y8JhnHT0pascKyacZiWTdutiQmJXot\njvTGCy8pb0fDnm1zeVUuvv4y2fP1j2pxEWxRqGCLTK0bN0sZRBp0v0uO7t0vu75YKefOZqoCUt3u\n66PLU+K5eAhWkJFvV6BgATl54qRjWksFajNVvVZV2brxZ0nPOCNpqSnS877Otm03GY78wfT0DDnw\nz0EpWaqEpKaeL8wVCmbchEJoatSoKTfe3lOa3Hqv6/kVCyfKlwvGydatW2xx4xlNURruRR7DbVAw\n67zNLN5oNrGq/Zr+3vbWU+rmHNq/cdoiqd+llUv710/6TGq3a660P5RCjp7HhjHh0tY3SfWWTaVA\nqWIBdR94035EWMTFxUnjXu0s133sr0PXe2TiyDFSqlZVObBtlzIuIH20ToN68vGiuSF9FwRaE0SD\n7nvT/uSkBDl3Lkeyzp2zve5bVWTQ8zse2n/2aIoylNpFT6n9sROhmWXiesfS4o3EHngtgDRqmhzZ\n86d6vVHPdur3ho/nuUSzys1Xu54PFPrny+Pg89ijp8v6j+e6jBa4Uc/OyVbC7t37kJMryiE+IUGy\nz2XJzuXfq/eYFaboN/TyzFTp0P1emTluqhpr497tg/aUuNcw0Lo16DUwYDGhFfEpXqKoOBFvbaae\nW/iTXH1xKXnl9sa2bTcZTm9x0WSRskXEVsB6DaGBt8KdKtUbyoJpH6h2whddFH1hkJGM0ohkWGks\nR6dEG770d82Y6Up/ofGnDh6VteNmqZTIhJRkZVTQtN+o7vs6NhwC0G8cO7Dui1ftz8k6Jznh0v3h\nU6V9ty6ydtWPsmn9hly6j1QOnKORCAm9ToZo0H1f2v/8op/k3kZV5NLSRW2p+1Z3LfB8Ddov1H5b\nEIsa+GoEzpmGBYfir/gQxBoFkLCYgNDnS0mWTR/Pl7NnzqiKy/BoeBZ0QpVpvW2YAhVfatC1tXoO\n3n8ItrfiSN0f6C9TxoyXy/p0kEtb3yynDx9TUQ4bPl0om6fMl7g478WnjIQpBioyhfOaO21mLs+G\nr0JOwRgYnNIqKhTgeYG3AgsLrXUWihVh7l5ZvkFqlS7qeoyQwEcH9rOFZyZWc9rdOR8Sl18JTYVK\nl7qe37H9J8mfv0BUhqLD0g7LPURWi9LAuedIjowZM1Yef/xx23iVzCRWzzsa0aO/0P6iF18oScnJ\nciYjQ+rdc4c06NIqJN3Xc+w6HW71q/vQUuBN+9d9PFfWT5obNt0vXLSIamftb6xGCkfGgu7r0f6P\n2lxlO903SrS1QaT2x4YGno6Q7sebvkcSFvxZ4xHKv2b8LJVPibC/DZM+k9Yd26F1x/moginzXK9p\nBZ3QusrTI4Eb7VbDB6vf8Djg+UDHzjydIb+t2ihrxs1U4/D2HvSE/nXrdtc+sAgqVLaU+l2+cR05\nk54hd3ZoqzwH7mNdO3am6iMdrNjj/Vgg+drfyeMnfJ4PxgpPSTD4avMYa22VTpzJlL9OnHY9Pnnq\nfPinHUAOpJbrjtBFbyHq0Q4EBSFxS2YPV+kP+3ZtUb+Xzh4hPXv2iDqRDRSlcerUSRWlEY3E6nlH\nI3r0V9O3Vh3aKt1fO2F2yLqv59irRk3zq/vQUvd9uGt/xSsbqOgKp+p+rKFH++2m+0YMCviBQSFa\njAqA2h8bGvh3hHTfUMTCpEmTZMOGDfLmm2+aPyKiC7/FD5OT5NfPVsjGKfOVZb77fX2lz6MPyWfT\nZkjRSy7KU9Dp6C97Xd6AQG2YEGmQkZ4uKWmpPttZLXvuXTU21Cnw5S2oWrO6X2/CE4OelUKFC6lj\nYqzaecB7YgRtO2/7Q10Kb2NBISecp7tXh+hvq1QwOVFKF0yzddspzbiAxQOMC1YtHuxWME8DeXYA\n1mukPyBSoX//fq7no41Y9NTY+bxR/Pmdd96RmTNnRuT40ab9mv5aofu4ES9QqKDvY8fHy97PV/rV\nfe14vvaB5zt26yKfTpgUMd3XxoE50arlE2PaP3PDHlvqfjgjFKj99sCuGhiN52zIsFC2bFmZM2eO\n+aMhQVvj4VFwb/UEazyEE2F8sLhDyDVhRNsmvL+uWzFHRDPg/dp7AuUlNqndUEUUYPGwesSnkn3u\nnFzYoJbr2N369ZG7e3VXx1UdKnyMDz24/Y2/YKFCKsfR23kYAaGcvvaH19zHUrr2JfLj6Gny14af\nlQelad3GprTljEZ8tVV6ftFaueriUrL5ryO2bjtldUqE3ds5YgzIs0NIHKzXdiowZaWnBjmGCAeE\n5R4iiygNGFSi9dztet7ly5eXHTt2ROTY0aj97vprhe7jhrtajeqyxqOVJY7dvus9MvDlQX51Xzue\nP+2HTj/09P+FXfcxjj/WbpY1o2eo18xocx2L2v/C4rXSoX4l+ejbrbbRfT0tfz0JdT1A7bcXdtXA\naDxnQ10hMjIy5Prrr5dnn31WbrvtNomPt1dGRaxVhoY1HqF7sMYjxM+XCOp5v79KyigM2bB7GynX\nqPb5gk2o4ixxkpWZ6XVfgY4X7Pi9EajQlNG5RB0Kz4JO2qJHD6izEEyupdaqEV6ASIuw0crQ44ZP\nUWGP8FBccmlV2b75Fzl1OkM97t7vbltVhw5XjYWXXnopT+EcpB7gSz1aiwXZHW3BhygNhAPCco/U\nD7sYe+x+3mZ3hWjZsqVaR/Ts2VMSExND2he1P692mq370MJL69SWX7f/bFjXQ9V+q3Q/ITFRFZBs\n1KutIe2PJd33pv0pSYlyLidHtZ20g+776vgUyMCgpT+EArXffsSi9meZeM56td+QYWHcuHFy//33\nq4NgIeBu9ejRo4e8/fbbEkliZXHhLrLBWPcDvf+tF19WVny0rXS1gRw9XS6oXU1avvefuKqWVZPm\nybRlC6VcxfI+jx3oeMGOX2+hKSMcPnRIbqzTSOp2bRVSyyu9CwxvrRrhBbDTTbhetBZaWussz8d2\nwoyFQyBitaWTU8D1iYUoDbPP20zDwldffaWMCqdOnZJ8+fJJgQL/pZ3dcMMNQUdGUvvDo/vQwuXr\nV6saDaHoerDab5XuYxy/7/1N2t/cwmd7Sj3aH4u6D9y1HthF973pvHs9JU8Dg1kOB2q/vYlF7T9t\nwjlb2m6yRYsWsnTpUq+vlSlTxsguSQhA7LQ2THos+e7v15OXmJI/TbKzzknjXu29hkmmpKb4FdxA\nxwv0uje8tbzS2x7SH8gjzTidrvaJFp2nDx2VtOJFDLe8MtKuyQktmrx5WtxbaHl7bPbx7E6stnN0\nChDXWJx/O513vXr1ZPHixV5fK1LEZj3aHK79Zus+jAqh6nqw2m+V7mMcySnJrnQQq7XfqbqvV/vN\n0n1fxwsFzWAAA4O3iAYzHA7UfntjJw2MxnM2ZFgoXbq0+iH2wUxLvmdeIooXos7A35t/kdK1q4Xc\nBipUAhWaMtomCmBRhoKNq0Z8Ioe375Iz6WckOTVZilWvpOY00Lm697EOpV2TXVs0hdvTYsXxfIVH\nmk0sFgsiJBgKFy4s11xzDSfNBtofy7pvhvZHs+47UfsD6byVEYvUfhLLGPo2WLFihYwdO9bra02b\nNpXu3buHOi4SJFZZ8iF2CHnxV2wpkJiblQ+pEajQlF7PglcjQIJIlRoXy8/rt8tLLRq6vAnPLfxJ\nLqlfXbISfpcTZ/3vV2+Opb92TU8tWKPCCc20/DvR02Lm8cJVVyGchXPsWnGaED1s2bJF5X96o1at\nWjJwoLFuALGEFdofzbrvC4zxkpo1ZOu69Xm0v2aDeqZ1h3Ci7kdC+19+7h0ZP+xjGXRLcMcLt85H\nQvup+yTqDAv58+eXcuXKuR6fO3dOVq9eLTt37pS2bduaOT4SAUu+Nw/I3T27Sdd+feSTseN1t4Gy\nKh/SX7stPZ4Ud4OCpxEA3oTd23arhUUeb8KXW6R4fKpp3gR/7Zrs2KIp3J6WE8dPyMgPJsiLJhzP\nzPZRdmjnaPeK04ToITk5OddaAn/bW7dulS+//FIViCbh0/5o1309c7lj6zYf2r/VtNaTTtP9cGs/\nPj8wKoz6cKIMubWR7uPZwaBgtfZT94kTMPQtf/nll6sfT+644w4pWLCgGeMiEbTke/OATBg+Si0o\nUMBIb7ElK/MhjXhS/BkUIuFN8NWuyS4tmiLtaXnp2bflzNks044XicWGVe0cYVTw7DYB7whgtwni\nFKpUqSKvvfZanucfe+wxyc7OjsiYYlX7o1X3g5lLf/pmVo0Fp+l+uLX/fPrDZDmXnaPreHYzKFip\n/dR94gRMdW2hivPnn3+u0iFI+DDTkq/HA6JHXK3Oh/QsNOXPk6LHoBApbwLyBQGs8BBMHAOLC+15\nO+E+N7VKF5X9x09LmUJplswNPCQzpsyTlIR8jvLshKNwDsIgEakAo4LWbQI1HBByCe8IFjJMiyBO\nBmuJyZMnS9++fSM9lJjQ/mjUfSNz6Vv7zaux4DTdD6f2a5ERz95UV/63YqNf7berQcEq7afuk5gz\nLGRmZqrwRW+RDMRazLTkm+UBsTof0rPQVCBPit66B+H2JuA8kC+I0D67tGjyBcbVrW9Heea98fL8\n4nVyNitLkhIS5Fz2Oen3cDdTxw0Pyen0DLm3URV5YfHaXNfi+UVrpVP3NrqOF65CjeGEFadJNIO/\n9eXLl6uUSxIe7Y9W3Q8G7OfuPj3lhY+G59H+Lvf3M+04TtP9cGq/FhlxU7UL5eSZrDzaP2jpOuna\n5x7JOmN/g4LZUPdJVBsWPv30U3nhhRfyfOhR4XnkyJFmjY1EwJJvlgfE6nzIUFpV2tGbYGZrRitv\nrM9kZElcfII0u6u/KwR/0axh6nm9x/LsHe3PQ1K9VBF5+LpL5ZXlG9S1SE3MJ5IvXp59+TG/2xvx\nZjilIJLeitNOOR8Su/zwww95ij0fPnxY1W1CkWgSHu2n7ueeyyGjxv6r/WnKqGBGREQ4dd8q4vIl\nSLPW/2n/4lnDLIuMGNz8fJtmTfsT4uOkS6+75ZEn+5luUHCCVlL3SdTXWHDPi4yLi5PixYtLo0aN\nJCUlxczxkTBb8s3ygFidD2klTvMmhKtgIcR3yoRZ0vyu/nlC8D+ZOE6eG/hyQFHWekcHMi5okSOD\n30cV6nryWc+bZNG23+W9b7ZKv4e7SqFCBU07b6cVRMIc9+jRXYYPH5ar4vSSWcOlX7++kpSUJC+9\n9JJjzofELt5qLBQpUkQaNGgghQoViti4Yk37qfvmzeVvJw/rjpB0CkhRGDfyU2neOq/2jx81Vh5/\n+j5T1kieEaOtaleU/En5ZMjyjXJPj47y+pB3xUycpP3UfeIUDP3lIF/IrHxhYgxfrZzM8OCbFf1g\nZT5kOHCaN8EbZlr2zQjFw3hgXAg2cuTkqXQVOdL34W66IkeCOW+nFkRCcbuls0fKgswzkpiYLNnZ\n5xx9PiT2KFmypLRq1SrSw3AM/lo4hqr91P3Q57JgUjVV0wnGBRAtBgaVmnjqtE/tN7N4o7eI0W59\n75EXn81b5DVUnKiV1H1id+JyYBY0wF9//SWDBg1SoYy33nqr9OrVS2bMmCEDBgyQSJOenq6se+v/\n3C0pqc7zNPvDqlZOvhYxZuQxmrUfo0Doo0XggwGeezMNC4hYqFGjptx4e0+X1wKsWDhRvlwwTrZu\n3aIrjBCGBT3pEP8dNz2oyJFgztuscwon7mO+umk7OXHsoBQsXEK+/XyafDF/rHqPk86HOAvoa4UK\nFdTnMNUEfT1+/LgMGTJEFX6uWrWqDB8+XD1+/fXXVTRksGOLRu2n7juPYIpG2x1ocK2Lrpcbb+uV\nV1cWjpVNu780ParTXfdRU8GK9AcnaT91nzhF++ON7rxJkyZy9uxZlRZx5MgRqVy5ssybN09Wr14d\nyrhJALRWTnW63Cmthg9Wv5FugOetstqHagwwaz9EH7ixtqJoIUQWYYJLZg9X4rtv1xb1e+nsEdKz\nZw/LRFiLHLEiHcVfFMapUydVFIbdcB9zUnKKFC9VTv3GmPG8086HxC7wa7Rp00a2bNkit9xyixw4\ncEDVaoLjYurUqZEenm2g7jsPRC/gByCCQYticCIqRaFfJ6/a371vR0u0GfssWuS8UcEKnKb91H3i\nFAy5uL/66ispUaKEjB07VnkX1q9fr55v1qyZzJ49Wy677DKJ5pDDSGG0lZMdz4WYTzjaLyH3EKCt\nIUIgUSywf/9+ruedht6CSMEUevL3uhlFovyNGc+DYM6HkEixc+dO+fnnn2XXrl2qq9SqVavU882b\nN5dZs2ZJp06dwjoeO2plKC0c7Xg+sYZmXHBPkfCG0agGb06EYCIC9ewP3PdgLzmbkSUfT/hX+wvk\nl74P3SsDQyxqHcgJYtVaxmnaT90nUW1YQNXmihXz5p8lJiaqKAY94I8MIZDuuZb58uXz+t6TJ0+q\nsAtfrzs15DBYgm3lZOdzIfYu1OgLfG6Qe/j4448riz7E107hgsFGamhRGMirdC+ECE8MDCbu5xao\n0JO/14FZRaICjRnoOR9CIg3WEmXLls3zN4C1RDj03glaaaSFo53PJ9YNDP5SJsxwJGgFkoM1MARc\nQySLDBn0ptz/SC858M9BqVStXEiRCuFwhEST9lP3iVMwpDD16tVTNxZIgdDAHxaiFfR6Lj/++GN1\ng4K2UgcPHpTdu3fnKfz27bffykMPPSSbN2+WokWLysCBA+XRRx+VcIQcwjsAIUe7RKQaAFQLjiTB\ntoQK57nQMxJ5winOEDk7FXANxbDiKwrj4YcfVt9LmochUKEnz9d/2bxKhg4dJpmZmepGycwiUXoi\nR6IlqiRa2oWRvNSoUUO2b98uO3bsyPX89OnTpXHjxmGbsmjSfUDtd2YNhmAN4970TnvO3cBgdF/e\nKFukupQt8u/+z4SWchkJg0IgHcVNf8eOHZWmaFpiF+2n7ueF2h9FxRv79esn8+fPV4UcEFEAyxty\nI9F7OpiCS1hUYHHhaVjYunWrWli8/fbb0qNHD2W4GDx4sPoDtqqAE26Or6pWS9Ut0EIOwfop82Tz\n5Hny3c+bIh5O+NaLL6sFT4MebfK0cHRfAIXrXJzgGYmF4o1mF2q0kmCLN4YrUgMChSiMYsWKyXvv\nvZfLw9C1axcZP36CNG3ppXjVgnHy009rpEGDhqoQ1HW3dJLFs4bLd59PlzMZpyU+Pp/Ex8dJs1b9\npOntXU0tEqWN2VvkiL/XogUntQuLFswu3vjGG2+oa4gWk7/++qtcfPHFsn//flm7dq3kz58/6LEF\nq/3RpPuA2u9sg4JTdDyawHcZvnMmTJggEyZMzKUlcDDUrl3HZ5HHSGh/rOs+oPbbV/sNr7xQWwFe\nheXLl6uDoIhj3759g67i7AssNNBtok+fPiplAkYLPUaFcIcchgP3aAC9LaHCdS529vRoAh7NRgUr\nijRajWo5eWSPafsyOwrjpZdeyuNhGD16uGRmnvXZbmvbtm2uQlBYWHy9dKo0b/3f9gtnDlUFr86e\nyVCFFvG7RKnyriJRntEfhw4dUsbVmjVrSvHixQOOOdjXogUntgsjuXnyySeVUQERj0iJrFOnjjz4\n4INBGxWMEk26D6j9zuwSQYNC5IBWTpkyRUUueGrJ0aPH/LbZDkb7wfFjB6Vi5VpetZ+6rx9qv30J\nyaXTrl079WMF33zzjSoGifZTCHFNTk5W7S0feOCBPO9FuBGsV+5WlXCFHFqJv2gAFGzy18IxHOcS\nSlEpK4k1g4ITFyR2HTOMpPB+Y3GheSdQ2Ak5mItmDFUhjt4KPSHqCh4OvA5vBRYWebf/SJ5/sJmU\nLV9Z9v++U86eSVceDRTBff7555WHPSMjQ1q1aiXr1q2X7Oxz6vX69evJnDlzJCUlJWLz4rRrhQUi\n0vWi2WMTTTRt2lT9BIsZ2h9Nuh/r2m9HoqntZCxqycyZYyU1Nc1nkUe92v/sfTeiD46cO5clCYlJ\nKkUC0ZGAum/e9aL2O8SwgKiE0aNH69rhzTffLD179gx1XHL06FGZNm2aMjDUrl1bjaFFixZSq1Yt\nueGGG3K9Fz2vX3zxxZCPCSGEgMPjDoH0DDkMt1AGigbw50UJx7nY0dNjtBCSU7CLQSFa89r8taDC\njf6yuaMlPiEhT6EnRBUgbBJ5lVlZmT62z5brb2ovXy35WC6uWldub/+gKxoCiwx42GFUWL9+g9za\n9v5cHg88v3jxYnEK4fh8+LtW8CR5iwQhkQX1kl5++WVd74XuP/PMM37fY4b2R5Pux7L2241oNShE\no/YH0pKuXbvKlCneizzq1X7cAH+9dIo0uvo2KX1hZVk8a5hKuaTum3+9qP0OMCwgDaFKlSq5nlu4\ncKG6eC1btpSkpCT5/PPPVTHHu+++25SB4ZhNmjRRiwtw0003qXSLpUuX5jEsYPExYMCAXF4Lf+HD\n/ggm5NBKzPAIWH0udvP0+GrtZGdRN5LKEEmDQrTntQVqQdWly70yceI4r0UR8Rse1BEjRnrdPjkl\nvzS7s5ekFSgky+eOVYsLdyt79+7dVaQCjAreoiUQJmn0ey0aPx9G2oWRyIK8TM+1xA8//KDqKdx1\n111K91euXCkbN25UTopAmKX90aT7sar9diFaDQrRrP2BtAQRhfhu8lUQOVjtb9NlIHXfwutF7Y8s\nur4NUETRvUIzCi2ivgJyiwoWLKieg0UOCwF43swAx4Nl1J1Tp/BllldUcUyzjosvSHgF9IQcWokZ\nHgGrz8Vunh5/Bga7CbxdIg+CJdrz2gK1dMI5ojuNt+JI+HtDgVmAKAT37ZfMGqEKO6G+gmZVP3Hs\noBQvVc71GDdYiIrwFS2B79trrrlG7Ew4Px/BtAsj9qBy5cq5IhZgCEAo8YYNG1RRKI3evXsrvQ+E\nWdofTbof69ofKaLVoBAL2h9ISwoVKuS3zbYR7afuW3e9qP2RxZCZcdWqVerGXzMqgPj4eGnevLl8\n/fXXKmUhEGfPnlU9rNFqEhw4cEDlEMPbgIUCKrGixsLkyZPlqquuklmzZqkOElbVdPAEwhjJUD4z\nPQJWnotdPD3+DAyRTI/wF5HgJINCLOW1ebZ08uxJHagoIrwb+A4bM2aM2j4xMVktLJrf1S+XB6Ng\n4RK5rOxXXnmlqqngzQqP53FM9/aX0fr5CCbUVk/7LWJftmzZIqVLl85lVAAo3Ixiamg3HU6iSfdj\nXfutwt96wp9BwclrgVjQ/rxakl86deqk7kU0zNT+DWtWUPdDSLGh9tuXeCMboXLz999/r9pMasBq\nvWzZMvWaHtasWSP16tWTtm3bqg8SUirweNOmTer1a6+9VtVYGDZsmCrqhBoLSLe45JJLJBbQPAI/\njZmh2l79s3Wn+g2PQOde3W3jEdA8I2jJtWzdKvUbj50eGhcqWERoLSB9/TgNf3ltWoXjaACfXYgW\n0h5SUlLVOaMFFTw27oXi/G0P7wa6OiA3MycnW1LzF5I/9v6s2kwtmjlMyl1UXf76Y6d6DCt79+7d\nZNy4caq4E2oq4HlUksbvhTOGSqlSJeXqq6+Ryy67TGrUqKk6V+gZi5M+HzgfnBfOT+95/jfXW+TH\nH39Uv/E41r9/nALWC/g7+f3333M9v2TJEt1riWjCKbofi9oPg4LWutrXj7+1AHDqWiAWtF/Tko0b\nNyjdzs7OkYkTJ6hWk3r1NpD2o2tE7YY3yLefT1O637NnD2VYhb5T9/Xrfu65pvbbDUMKgBoHlSpV\nkksvvVTuuOMOV40FhC726NFD1z4QhfDXX3/5fQ+MDfiJVZzkEYi0pycQ4aq34L6AiDaszGuzW0Eo\nGBG8tZ4KJuwT5/Haa6/lyc2sX6+ubNu2Xd4b3M3lYUcqGcJMm991n/yzf7csnT1SFmSeUZEKF5Qq\nKYcOH5FbwhyCGuw1CfXzEUqobSy01oxGKlasKB06dFBOhTZt2rhqLCCSYfXq1RKLOEn3naD9dugy\n5fT1QCxpPwoqfvLJVNO1PykpWeLjRNZ8tyBXZB107+Chw6qg87LPRqv3QvfTUlMiovvBXpNI6j6g\n9tuPuByEGhgA1qQJEybIt99+q4qWYGHQr18/KVCggEQa5G3iw7b+z92SkpoqTgcFnSKZ9xktmJUD\n6eSQxlCBJRlf+s1a9/VafyAaCkJBVGE1v/H2nq6wTwCPw5cLxikLebALIOzTPTfT/THwPB56Xi/9\nbJSs/GK6xMXFmzoWK6+J0c+HFXNOrNNXpC7gmqEYoxnMnDlTFWY+duyYqrnQv39/Qzcr0aT91P3I\nYlbbai1y0elQ+0PXfuC5DnDXPeg+6i/8+O18+Xz++FyFnMOhh0a1n7ofG6Tr1H7DK3d8yNBW0ozW\nkiS2PQJOKejo1IKLZmJ2Xlso1mqrPB1WtDLytKq7P0bdBM/jodBTnUY3yhcLJ6rH4WyrFMo1Mfr5\nYPuo2AbRCvgh/0HdjxxaykMkuj7ZFWq/Odrv/n9P3YPuo6Bz2QqX+CzkbGU7RaPaT90nphgWEKY4\natQo+eOPP1QYr8Ztt90mDz74oNHdEhIWtEVDsMIfqwYFz7w2X9WRw1EQyuooh3C3MvJ3PDwPwjWW\nUIt0Gf18sH1U7PLLL7/IBx98oAxs7nm19evXV3/nhDiNaEyJpPaHT/v//nOXz0LOVrVTDEX7qfvE\nHUOr8L1790qTJk1U4UXUW0BHCA2EMRLihHSIaBT/cGFGXptRL7XVba/C3coo0PFAuMZiVuRAsJ8P\nto+KTVCX6frrr1dFu1CwOV++fLnqLxDiJGJhTUHtt177P583VurXrydLZoevnaIZ2k/dJ4YNC999\n950yLKDGAiFOMygUzkyNmrxHJ6PXS+2e8gDC0fYq3K2M9BwvHGOJZOQA20fFHijSWKRIEfnss88i\nPRRCdBV8juUaS+HUGc9Ux3C1vLSL9j/55JPyxhtvhG0ckdJ+6n70Yah4I1o/jhgxQqZPny52JJoK\nOBFzDQqA4m8f/BX90Somu6c8tG3bRiZOnCgPPz8+l/ihNSO6LKDloJm5h55FF63G3/HCNRazi3TZ\nfc5J5Io37tmzR1q3bi3r1q0z5TJQ+4mVBRxjISIhXPjSmb59+6goaM9Ux06dOsmVV14Zc9ofznFE\nUvup+zFevBEhi88//7ysWLFCbrzxxlDGSYgl0KDgDPxZqwcPHqwWFze17CnVal2uLOlTpgyTxKQk\nH1b1/KZb1a1qZeSv8KQvW2+42ipF2oPA9lGxAz7PSHmAsfCee+7JlVZJSKTWDJ6wcHP4dAY123Bz\ne83NHaVG3atl384t6jG6z/n2qDtD+wMVnPam/eHUw0hqP3U/xiMWxo0bJ/fff7+yXsBqkZSU5Hqt\nR48e8vbbb0skodcidqFBwZm4W6vxfQKjwogRI1Vl5OSUNLm6aTtpflc/+WrJFFk6G/UU4nJZ1RfN\nHCbxcXFy3339I9qqMhC+Ck9qYY92a7vJyAFiZcTCV199pQo+o9ZCcnKypKSkuF5D/aY5c+YEtT9q\nPzG6ZjhzPEtKpRb0+T5GKVivM9DHSy6pLhIXJ1mZZ13an5xaQL5eNEG6dLn3fDqEw7TfX8FpYMeW\n29R+EtaIhRYtWqie094oU6aMkV0SEhKxblCwqvViuHC3ViMcb/SYsaqHs1accfGs88UZazdsIgsy\nM6Vr167yyScjZcHZM5KYmCwNrrhFSpW52NQijlbMu6/Ck6hbs2nTZtfzv2xeJUOHDlNeGhhZIgE9\nCMRq6tWrJ4sXL/b6GmovEBI2g0KqM9cM0aT9jz/+hPLat7irfy7tr3f5zXLq1Enp0qWLJCYmKg11\nkvb7KzgNtNcqVL5Utm34Tml/OM/FE2o/CXvEgt2h1yJ2iHWDgtWtF8MNxLpGjZpy4+09XQWawIqF\nE2X53LHquW+WTJaffloj9es3kCuatJNmd/ZS/Z+19325YJxs3brF0kWWkXn3d26LZgyVW1r1kSa3\ndVELqe8+ny5nMk6rllPIOUXqmROvJ4k+zIxYMBtqPzESoeDENUMsaf/S2SMlMTFBtm3bqp6rXr2G\nY7Tf33l9sWCs5GTnSJPbekhG+gmX7ickJkmc5Mj27dulUKFClp0LIbaJWAAnT56UN998U7755hs5\ne/as6jmNPyxGLJBwEOsGhXC1Xgy3pT9Qy6Plc8eokMfjx49LevppqdOoiWthEUxrJL3eBl/vMzLv\n/s4NKR8XXHixMip8vXSqNG/9335Hjx6uvDR2vZ6EhAKicoYOHaoiF44ePSo1a9aUAQMGSLVq1Tix\nxDSiJUIhJrU/84zcfXdHtc3u3bsdpf2B1jTgn/27Zf3q5bl0f+HMoSpaEfdZhDgJQ5WSzp07J82a\nNZNp06apQo5IjdiwYYM0atRIjhw5Yv4oCXFbHGgLBBgU8IPFgZMXCEbxbL+Egkb4jfxDFN/B63YC\nln6kOcB6j771+I3HeN5byyN3kEsJ733v3r2UAdPf+/y1RtIzhkDvMzrv8DzgHHyd2597f1EeCywu\nnHA9CTEDpDVh8QznBDpEHDx4UBo0aCC//vorJ5gYWh94+4FBAesFGBWcvmaINe1PSkpWUXuB3mdH\n7Q+k+ykpqbJ+9bI8ut+iTX+ZMWOG7a4lIYEwFLHw7bffKvGHMUELh3j66afl9ttvl8mTJ8uDDz5o\nZLeE+IQRChK0JTyQ5T7c6LH0wzOAsEI8j/7Unm2otHoDCDn09T5UMfblidDrbfD3PlSvNzLviLJA\nZMLiWcNyjXnJrOHq+RULJ0hWVqZjrichofLHH3/IwoUL5eeff1Y3DBqPPPKIfPjhh/Lee+9xkomh\n9UEeHB6hEMvaD03XUgICvc9u2h9I91u0aC6zZ892zLUkxBLDwr59+5QlzzPH4vrrr5e9e/ca2SUh\nXqFBwTfulvu87Zd8W+6twl+IoaelH2DMEFpY+h9//HHXNnpbHnl7HxYcHTt2VMfD/vBb+07CfOgZ\nw6FDh2TkyFHS9PYeXt933333GZp3zEtqapqUrVhD1YvAmJOSUqV8pVry128/S+fOd8vYseNscz0J\nsRqsJapXr57LqKCtJSZMmMALQPwSq+sDar9/7QdYi8AYgRt7/A6k/QBrhVGjRvt8nxHt96r7yWlS\ntsIlcmD/LhUNMX/BAuo+iW3DAnIfEaFw+PBhKVasmCs9YsGCBXL33XebPUYSg8TqgiEYjFruI1HM\nKBgPC7aBBwFi76/lkfv79u/fr25EJkyYKMOGDVNCXqNGddm0ebNknj0rcfHxEh8XL+fOZfkcA/Yx\nZcoUtbDIyEiXz+ePkzNnTqs2l/nyJbjehzxw7HvRzKFBzTueRyoHKj7DmPD7nq1y9my67N2xQYWB\nY4GBWgqoqRDJ60lIuKhSpYps2bJFduzYof6vMXfuXNZYID6J9fUBtd+39ickJEpcfJzSfaQaqJbV\nySly5kyGT+1HmsX06Ug7OKXWCn/9sVOtFaD77u/DOgc15VD/QK9Ga7qPdVqT27rJ/t9+lW0bv5N9\nuzZLYlKSjBw5Uvr26SMjRlD3SQwbFi6//HK1EK5Ro4a0bNlS9Z9esWKFxMXFyb33/lf1lJBgifUF\nQ7Do9e5biZ4QQ28elrNnMmTDmhXqeW+Wfr0tj/A+GAQwB9oY5k/7QKVqtWhzn2tMi2adb+Hky9uA\nxYn7PtzbXN7W7oFc79u4cZNcVLWuLPtstJp3LGDq168XcN7xOlpLrl+fe2xLZg9X84gFDowLkbye\nhISLkiVLSq9evVRNhVatWqkWkytXrpQ9e/aov19C3OH6wPnaD90/fuyg/Lx5pU8vv1Hth0Fg/apl\nckurfm4aPkwurFhD9uzY4FX7Ub9hypRP5JbW/XKtFQoVKaF03/N9OA6KLaJTBYpK4vlA8669BsNH\ndnZOLu3HfCHNE/ug7pOYbjeJCAUssFFvARY89KPu37+/5M+fXyINW045Dy4YQgPhf/68+1bhr5WS\nZ+sneOQhojfd2VsO7N+jChZlZZ5VVvv+/foZbpPlOQYsXF54qJk0u7N3njEtnjnsfJ/stvfl8jYg\nwgIeD2/nAePBjbd1lc/njc3zPhzrxLGDsnblEtUGM1CrK73zFanrSUgk2k3OmjVLlixZIseOHVMO\nC4Qcw+gQLNT+6ITrA+drPyL1KlapI7/t2iJnz2ZIfHy8clDOnz/fcHtM9zFc3bSdDHr4Frn5jl5e\nW1WXrVBN9u3cJM3b9M+l/VgPNL+rv9cWl72feF/27tzs9X3Q/qWfjZJVX85UbTADzb2e+QLUfeJ0\n7dfVFWLr1q2qraT7zk+dOiU9evSQsWPHqoKNTzzxhC2MCsRZsMuDOWgW/nAtLPDFgrZPyEn0leJw\n6tRJJZIaMBzAKr/8s5GybuUSad66nzz8/HjlXYDBAd4Pbb/BVEL2TLOAN+TsmXSvY0J4I0Ijv5g/\nVt4b3E0JOsaESCtf54F9fbVootf3od1V8VLl5JJaV+Q5Xz1j9TVf4b6ehISrWCNuJDTglEAnqbvu\nuktGjBghU6dOlRdeeMGQUYFEH1wfRI/2161bR/b8ukGateqjdP/Wtg+oyD+juu+pp9D9MxmnvY7n\nTMYpufbmDkr/3bW/U6dOPgsmZ2aekaGv9vX5Pmh/nUY3quMH0n3PsfqaL+o+iQZ0mQkRlfDTTz+p\n1pIAkQrr16+X4cPPhzwREgrlCxST40fS1f+Z9mBvPOspoJYBIg70FDOCVwI5kdjW3WqvFUdCSCXq\nG6BHtbc6Db7wDLUsVLiEJCWneh1TYmKyJCYmyNq1P8mJEydcXp7zxR69F2XC83h/8eLF/b5PT4FF\nuxXdIiScoPvDO++8ozpIga+//lpee+01Wb58OS8ECdjlgesDZ2o/DIjbtm1XKQBm6b6nniJiITkl\nzet4klPyyz/796hx/fTTGpf2A7R09H4O+WXx4sXKQ+v/ffp0m9pPYgVdhgVUbYb4I1SxTJky8uef\nfyovw+bNm/O8F8Ucy5Yta8VYSZTDRYP98ZZTuXDGh3laKfkqZuS3iOPZM9Lk1q5Sp1ETn62g9Bay\nKn9xzTwFllAvISf7nPTv31cZCfDjbx/u56G9F+9D4UbPfS+aOUzq16sb0Gtkl6JbhEQCFH5et26d\nSn2oXLmyqqWA6Edva4kCBQqwzVqMQYNCdGq/FbrvTU9rN2yi6iN4tnVECgZSGTUtd9d+f3qMex+N\nUHTf21ip/URivcbC66+/LqNGjZLffvtNWR990bdv34hHMjDP0lmLCC1iIRoMC/5aLjp97L5yBJfP\nH6/SG5KSktRNAiz4PXv28Op18JdniJzGwR8uVyGGvnI1A3lTUPwIYYXwYlSvfols3ox8zjMqnzNf\nvnzq++mZZ57x6g3x3Ie388D4q1evoVpH/b5nuwqxhDek3EXVZf++7bpyLfUch5BorbEwceJE+d//\n/ie7du1S+/JF06ZNg45koPZHz3oAOGlN4FTt1zPuULXfKt33pqcofhwXF/+v7p/vCuEvEiJcuq/3\nWIQ4XfsNFW+E4cDOqRBcXNibaCzEpKflotPHjhzIyy67TOVHuocD7tu1ReUsog5LSkpKwCJSWhHH\nZq375rL8N7jiFunUe1Ce/f7444+6PZeehazweN++feo1fCHqEX9/xbDc56D0hZVV4caChUuoatSh\njpWQWCveCMOBmakQ1H5nrwtgVABOcjQ4VfuDGbcZ2m+l7nvqKcD/CxYsmCvl0Q66H+hYhDhd+w19\n63Xu3FnatWsXyvhIDBKNBoVg2i45fez+cgTxPGyUnkLpzRuSt01WfomPi5NSZS7ONa5g6w54OxZ+\nu4cz6t3e1yLBcw5QuNHIWLWxBbMYISTauOaaa+Tjjz+O9DAIiTntD2bcvrQfLSNRawHFVt3TC7zp\ncTh0311Ptf97jivQPrzd6Jup+4DaT6IZQ4YFWAFPnjwpgwYNUpZKpEagbQy+OFCDgZBYMShoogSr\nPwTaszARRBQFC+1qlQ5m7N5yBH/Z+qMsmTVM8iUkyHXXXefyejz55JPyxhtv+PSGYOGCfWtW+7fe\nekvtNy4+Lujcw1A9RsFszzxJQswDXk6kKL333nuqUNrRo0elZs2aMmDAAFWPgcTW+sBpKRBO1f5g\nx+2pe5Wq1ZOFM4bK7l/WSXZ2tjRo0NClmcCXntpJ94PZB3WfEIsNC+fOnZNmzZqpRUD79u1VfhUK\nOzZq1EgVYSpatKiR3ZIoI9oNCroKE037QImoXT3TwY4dgotFyaRJo9TryGeMz5egWka6ez2+++47\n2bRps19viLvVXluQQOCxXwg8Fhfa80Y8L0ePHpOXXhoccIESrMcplLESQnLTtWtX5aBAK9ciRYqo\n744GDRqoIo9Vq1bldEUp0bA+cKr2Gxl3nz59VI21xfPGyIL005IvX4JqGemumZmZmWpN4E9P7aL7\n/vahjdUd6j4hFhoW0H7y4MGDsmHDBleexdNPP61aSE2ePFkefPBBI7slUUI0LBiCwclthIIZe0ZG\nhrRq1UrWrVuvCiKhQFJ2do60aNM/j9dj0YyhckurPpZ7cfx5XiZP/kimT58uvXv38unFcKrHiZBo\n4I8//pCFCxeqNpT4LtJ45JFH5MMPP1SRDCS6iKb1gVO1P1Tdh1HBW+vIESOGSny+eGneur+lehqq\n7gfaB7WfkDAbFlAMDYVMPIs3XH/99bJ3794QhkOcTDQtGIIhkmFyoVaiDmbsWFysX79Bbm17v7Lu\nb1yzQr5YONGr1wMLkAsuvNivN8R97AiJNJKn6s/zghDNK5q087sf9+3PnsmQ48cOSqHCJfx6bpya\nU0uI3cBaAjVQ3I0K2lpiwoQJERsXMZ9oXB84Vfut1H38+IuEgNEi0rpvRPup+4RYaFhA7iMiFA4f\nPizFihVzpUcsWLBA7r77biO7JA4mGhcMwZK3MFEBS8PjzaxErWfshw4dUh4LLC406z6qI3+zbKpX\nrwfaPP39x27VV9r9eewb3xmoEK2NHcWfss5lqXSKYD0H/jwvaAfV7M5eklagkM/9YHscf/60D+T3\nPdvkTMZpSU5Jk3IX1VD79fQ40ctBiHlUqVJFtmzZIjt27FD/15g7dy5rLEQJ0b4+cKr2W6H70M6s\nrEyfkRBjx46VCRMmRlz3g9V+6j4hFhsWLr/8clWssUaNGtKyZUtJTk6WFStWSFxcnMqTJLFBtC8Y\ngsFbQUIrvRVmWs/1jH3r1q15PBHoO13vsptl4cyhubweS2aNkFJlLpIln42S+ISEPN4QhDe7j92f\nByRQnqovzwvGcN0tndQY/e0H29esWUN5ZBDaqc0lzql+vbp55sGpObWE2BFUk+/Vq5eqqQDPKGos\nrFy5Uvbs2aNSLYlziZX1gVO13wrdh+YeP3owz2vQ/jp1aqsbfTvofrDaT90nxGLDAv5Q3377bfn6\n669VvYVjx45Jz549pX///iH3tSb2J1YWDEYIRxshK63naBnpDVRqRxSCp4egZJmLJPvcOVkye6Qs\nyDyjvAUQ9sTkNFkxb7R8uSC3N+Thhx+W2rXr5Bq7Pw8ItkMXGvSR9hXy+Z/n5XwBqMTEZDWG5nf1\nC5jvCo/Mli1bpdmdvfPM5ZcLxqm5dj+mFTm1oaazEOJUjh8/rorCXXvttaoANNYWt912m9x3333K\n6ECcR6yuD5ys/UZ1H7WUYHjQdB+a+8WiyZIvPi6X9sMIMH78BNvoPuYRadtbt27zWivCU/utqqVB\n7SfRiCHDwqJFi+T777+XESNGSI8ePcwfFbElsbpgsBtmW8/1hFaiF3T9+vXyeilmj5DSpS+QgwcP\nSZNbu0iNulfL3p2b/41O6J/HG4KFgufYfXtAhkvdunVUKyt/IZ/unpfnn39ePv54iqTmLyR/7P3Z\nZ96ods6jRo2WjIx0+Xz+ODlz5rRalKAwla+5NDOn1sx0FkKcyKZNm9Tf7Oeffy533XVXpIdDQoDr\nA2dpf6i637BhA1VrDRqKG/NLal0hXy2ZIss/G5VH+zHuYcOG2Ub3tXOOi4+Xv/7YKefOZSnd9zWX\nZtfSoPaTaMbQ6rVWrVoyevRo80dDbAkXDPbCbOu53tDKOXPmqHBlzUsBTwYWHTNmzJB33nlHeUwQ\n2uieq3n27Nlc3hBfY4cHxN3LgdzH2rVrycZNm+UWnSGfEHcsMHC8GTP857t6O+fFs87v+7Z2D/id\nS7Nyas0uBkXvB3EaSKfcuXOnWmjTmOZMuD5wpvaHqvt4Hn+zaC8JLYR++tJ+vbqP91Svfols2LhJ\nmrfu53dc7rr/2muvqd/jx48JWvcXzRomhYqUULrvby7NrKVB7SfRTFyOrxgoP/z111/Ku1C7dm1p\n3bq1FChQwPVamTJlpHLlyhJJ0tPT1ZfM+j93SwpTMwzDBYN9QfFDiG2z1n3zWM+DuSnFzWiNGjXl\nxtt7usIBwYqFE1U44NatW/JY45E+sG3bNnVTAI+G+740D0VSUpJPbwie9zb2vn37qCKwCJlMTz+t\nFjAXV6snfZ/80OVN8DUubx6Adu3aKkNDoUKFdJ/zss9Gy423dZXP540NOJfu5xusx8LIvPuC3g8S\nbn2tUKGC+gyHmvqINErUWMjMzFTplEWLFnW9hnoLcGIEOzZqf3jg+sC52m+m7uvVfnRrGDFipNdx\nI0Vy8ODBMm36dEk/fVoSEpPkumbn0xqg/f7G5al/cEp069ZVab+7sdLfOS+dPVJ6P/G+W7Sl77kM\nRfeNzr0vqP3EjtpvOBUCRV3w8+mnn+Z6DakRqL9AnAsXDPbHLOu5kdBKLCquueYavzmmWPz48oZ4\n5kbmz59fjR0LD/fiTuejCIYpT4jmTQimFdSUKcOlcOHCeRYIgc75y4UTpFOnjmqx4y8KIJScWjND\nWtkGiziV9evXy7Jly9T/v/zyy1yv3XDDDcorSuwF1wfO134zdV+v9sNxgHF66r7mbPjkk6k+IwiD\nbf88evRwFUnhrv1+zznzjAx9ta8ySmgOEF/aH2otDWo/iXYMGRa6d++ufkj0LxgA6yhEbyVqq4oR\n+iswhRt2kJ2d4/oNjyXaUHnbZvncsXLzHb1UTqa3cQVb0MrXOf+y9UdJSEiUuPg4mThxonwydark\nZOeo9llm1z8wa97ZBos4meuvv16OHj0a6WEQHXB9ED3ab2UxQl9ajFaTXbt2yaX7gbbRtN/XuILR\nP3+6DyME5hWRkliHrF69WhV2xGNqPyHBEfQKGV7F+fPny88//ywXXnihyr0yckODok3fffed63Hn\nzp1VFVhvoM/1N998I82aNZNKlSoFfSyin/IFiqnfx4+k06DgAEK1nptdlEiPRR4hj57eiVGjhqkb\neF/b/Lz5Bznw929exxWsB8B3q6phEp8vQZre3itX/mWjq29TFaxDqX9g1byzDRZxMl988YWsXbtW\nihUrJnfccUeeEGsSeWhQiD7tt0L39ejR6DFj89RMgmHR3zZLPxsl3y6b6nVcwehfQN1v2TtXy8mL\nq9aV29s/GHLtI0+o/STaCdqwAEPCwoULpWzZsuqPFjcJWBgE+0V04MABFQaJLxWkUzRv3tyrYQEW\nyfbt2ytDxtSpU2lYIMTkNoVmFiUK5A3B89Onz8jjYcjKOitL54zyug1qLYz/4P98jsuI98XznLF9\nvoQEuaVVP69ekzZdBprSztPsebfK80SI1Tz66KPy3nvvKQfF4cOHZcCAAWotUa5cOU6+DaBBwZ7Y\nVff96dHPm1cqHYdRwVNfUWjZl4bFx8fLqi9n+hxXsPoXrO7DoWBWK29/46D2k2giPpg3r1q1SrWZ\nRAGXffv2ye+//y4pKSkyadKkoA984403yvDhw2XQoEF+3/d///d/0q5dO3UcEp5FBKIViH1BwR7k\nMaIAENo94Tce4/lQQitRNOjHH39Uv/HYaMi/ZpFfMnu4Kki0b9eWfwskjZC2bdt49TCgVRUqTnvb\nBrmZ/saF4/Xo0V3VY3DfFm2runfv5nUh4HnOixcvksyzZ716Ps5knJITxw6q/586dVIZVM3AjHn3\nN9c9e/YwZRFEiNngbwj6j0jE3377TTkaGjdurAwNJPJrAW09gJRI99bSTIuMHHbXfX96tHzuGKXv\n3vQV6wEUWvamYffee69s27bV57iC1X4juq89pvYToo+gvkG2b98ut956q1StWlU9LlGihPrDRzSB\nFaxYsUItPNasWSNvvvmmJceIdTy9EkyBsD/+ivXBoq55M0Awno1Q0yrcPSmwyKNugmf7J9RXmDFj\nps9oBuRgTpw4Lo8VX89iB2laqO6MQkyJiclqIaP3nDF2X56P5JT8UrBwCdmwZoUlUQChzrsVnidC\nrOTXX3+V+vXry9VXX+36G+jTp48q7EYiAyMUokP38bcUTFRDqPoD3I8HjUcksnvL5969e6naBb4i\nC9DBAYWWvWmYFdofjO67j5PaT4jJhgWtlZM7qOyKP06zOX78uPTu3VulSaCwij9wA+NutcU4SWDc\nvRKA3gj746tY0bnsczJs2DBXyyWtCCGs8WYXH/LEW7unmjVryLZt21XxIzzu0uVe1/H95XbCmzBw\n4MCgilJhTsaOHSe3tr1frm7aTnkZsCD49vNpMm7cOHnyySd1La68jQuVqetf3kztK9T8U7sX8iQk\nXJi1lqD2hw4NCtGj++7a69nmORza77nu6Nq1q6vlM9bxvnQfrxvRsFC135fuL5o5TNVY+OuPnabU\nnrAKaj+xI0F/06A1VNu2bV2Pd+3apXpRa8+hwCI8D6HyyCOPqNoKjRo1CvjeIUOGyIsvvhjyMWO1\nWCOjFJyDr2JFB/bvUVWW0RsZYrh+1TKVN+jp2TBSfMib90N7DguCV155RRVjvKX1+ePNn/aBrF+/\nQVq0uc9r+6dAHvZgPSh79+5Vc1Kh8qWqc0TxUuUMtW70HFdSUrLEx4ms+W6BI6IAzPA8ERIuUMDZ\nfS2Bv1NEP2rP1alTR92U+IPabxwaFKJL931prxXaD80Fn3zyiatFtLd1h3vLZz2RdZHQfm/jql+v\nrjLOvDe4G7WfkCCJy8nJOd/zRQdfffVVwHoK6D19zz33BJVeUaNGDdm9e7frCwBdIBo2bCivv/66\nJCcnuwwNsH7eddddcssttwT0WqC69Po/d0tK6n9tE0nehQUNC84Coo7cSiwkNM/F2TMZ8twDTaV5\n637Kaj/o4VtUiybtdYC8wy8XjFO5hXqt7p7eCHggkM+oiTCiEVCQCWGHCYlJcl2zTnLjbd3kpcdu\n03V8nEsoHvY83pJ/x9D8rn6SL1+CoXP2HBdgFAAhkktfK1SooP5OUg3qK+o0vfXWW37fU7NmTXns\nscf8vofaHzw0KESf7uM5PLZa+5FyMHLkKBUpkZOdrYorXlytvnR/6M2w6b5V2u85LjPGSUgsan9C\nsD2n8WM1+CPu1q2bWnxonDt3Thkf8OMJPKGB0iVI3oXFmeNZcjyTaSN2wT0KAKlA3vIjvYXuIfc/\nK/N8AaLjxw7KmYzTulsvBpvTOXz4MLW4uLhaPdm7c5Pc3LKnlCp7sfzz525Z+tloOXzwT93HD9XD\njo40nu2r0B7y+NGDcsGFlQyHL3qOi1EAhJgLnAmjR48OeT/Ufv3QoGBf3KMAKlasGLTug3BoP27o\nke7QvFVfl+4vmzta5k59J2y6b5X2e46LEYCEGMP8pCud/PnnnzJ37lx1IwU+/vhjFWWAiISLL75Y\nVYx2Z/LkySrFwj10khg3KJRKLSiSyroKdkCzvo8aNTpXFICv/EhvLZMSk5KUuCJiITklLeT2g75y\nOrGoWTJ7pKq+XLFybVmxcIJaUMBLABAKifFb2f4Q84WFxYgRI1Vupef4Fs34SFJSUm2fukAIIZGo\npwRYUynyQMuQzoObeC0KAFret08feeaZZ3TrPrSvUOESlmv/wukfqf9rup+UnCoFChWXH7+ZL3Hx\n8Za3Pab2E2J/ImZYOHnypKxfv179v2/fvqrlFH480xw0evXqJZUrVw7zKKN0YUGDgq3QvAPlK9VS\nUQAIbfSXH+mtYA/CijVvRu2GTZT13luRJL0WfF85ncoDkXlG/V8bq5ZbiTBELc9z4cyhIR0/0Hxh\n4eOrfRUiKpYsWSzVq1cP+ViEEOJUaFCwN9CyoUOHKWP8bW0fyBUdgDSDYHQf2lfuohoha68/7c/J\nyZZ9u7e41ijQ+j2/bpDb2j2g1gGhrjv0zBe1nxB7EzHDQrVq1fJEJfjj3XfftXQ8sQDrKdgPzTvQ\n9PYeygug5Uu6ewngocBiwlt4pBa65+7NQL9lhAgvmzNSFpw9ozwGiHzo2LHjv+2VAov8+TQM722Y\n8iUkSU72uVw1HWBU0Mbd98kPZcQbD8iiGUPVzb+ZhQ+1+bqpZU81X748JMgDI4SQWIQGBfsDLUOU\nYny+fD50f0xQuq9FMaDwIOoLaIUIzdL+nzevVFEJ7jUdft+zTRWKxONz57KkUJESrraP6PJiZtQg\ntZ8QZxAxwwIh5D/vAPIVQ8mP9ObNAPv375cJEyaoHtJoS6W3/ZSvNkxLZg2X7HNZgpqvvmo6IC3i\n9vYPqorKs2fPlgYNGpjmsdDmq1qty+XMmdOqHaSVHhJCCHEKNCg4B2gZUh+BmbqvFR40W/uXfTZa\npWr4qukA3UfkQvU6V8nQV/vK4sXmRg1S+wlxBjQsRDks1GhvNO8AiiCZkR/pWXBoypQprnZQwbaf\n8taGqV+/vqoSOzwtemo6BGtU8Nbeypc3BVESYPnc8+NDOGnfvn1YV4EQElPQoOA8oGWpqWmSmZXl\nQz/zG9Z9PDZb+2FsGDtuXMCaDnt3bg46ajCQ7gNqPyHOgIaFKIWFGp2Bu3egQuXastjEHEV/RZh8\npVfo8YYA5H+OHm1eTQdv7a28eVc8vSk4NgpILZ87Rnr37qWKOhJCSCxAg4JzgZZBsz788MM8+onI\nwPvu62848s4q7Uf7dzNrOujVfW2+qP2E2B8aFqIQFmp0Fpp3AOKalZWpuzZBICu/3wKMQbSf8tZ2\n6fnnn1d1HHzVdAg2t9Jbeytf3hVv3hQswtgBghASC9CgEB1As1BseMSIEaqbEf6flJSsIgON6r6V\n2q+npkMw2h+M7ns7PrWfEPsRl4Nk6SgjPT1dfSGu/3O3pKT+114plhYdLNToPLBggOAXLFhQTpw4\nkStCwIiVH/urUaOm3Hh7T5fXAqxYOFEtBLZu3RJyLQJtzFrIpmdkg959GBmn+7FZU4GQ8Okrwpzx\n95dqM32Ndu2nQSE6wd/Svn371P/xtxWK7odD+z2114gWhzJGaj8h9tV+RiwQYhPcvQPFixcP2crv\nqwiTmQUOPT0aerwgZnlXvEVSEEJItEGDQnQDLQtU6DAY777V2u+tpkOwWhxKVAW1nxD7QsNClC4+\njh9Jj/RwiAUEmzvpLXQw2DQFPaGXoeCvtWUwhSsJISSaoEGBGK2ZYHftp+4TEp3QsBAFcPEROwRr\n5fdXgDEQwYRehkI4IisIIcQpUNNJqN59u2s/dZ+Q6ISGBQfDxUfsYdTKbyR0MNjCSqF4QczwrhBC\niJOhphOzvft21n7qPiHRB4s3OhAuPmKbl156SYl8s9Z983j3QxF9q4s/6fGCsCgTIfaGxRvNh5pO\n7KD7kdJ+6j4h9ofFG6MQLj5ij0h5981qVxWsF4RFmQghsQI1ndjNux8J7afuExI9MBXCQYuPM8ez\npFRqQfX/osmshh/NBLLwG82djFRhJSPFpwghJBqhQYEY0X6rdR9Q+wkhoUDDgpMMCqk0KMQKkfbu\nm11YyQovCCGEOAkaFIjdvfvUfkJIKNCwYHPKFygmxzPTGaEQQ9jFu29m6CVbSxFCYhUaFIgeqP2E\nEKdDwwIhNsMu3n0zUy7YWooQEmvQoECCgdpPCHE6NCzYfEFy/Eh6pIdCwozdvPtmhV6ytRQhJBag\nQYEYgdpPCHE6NCzYDC5ISLR698NRdJIQQiIFCy2TUKD2E0KcDg0LNoEGBRIr3n22liKERBMstEzM\ngtpPCHEycTk5OTkSZaSnp6ubl/V/7paU1FSxMzQokEDFnOjdJ4TYSV8rVKigvptSbaav4dZ+RigQ\nq6D2E0KcqP2MWIgQNCgQPdC7Twgh9oIRCsRqqP2EECdCw0KYoUGBEEIIcR40KBBCCCG+oWEhTNCg\nQAghhDgPGhQIIYSQwNCwYDE0KBBCCCHOgwYFQgghRD80LFgEDQqEEEKI86BBgRBCCAkeGhZMhgYF\nQgghxHnQoEAIIYQYh4YFk6BBgRBCCHEeNCgQQgghoUPDQojQoEAIIYQ4DxoUCCGEEPOgYcEgNCgQ\nQgghzoMGBUIIIcR8aFgIdUEiIkWTLzL/yhBCCCHENGhQIIQQQqyDhgWdcEFCCCGEOA/qNyGEEGI9\nNCwEgAsSQgghxHlQvwkhhJDwQcOCD7ggIYQQQpzHibM7JDNf8n8pi6lMWSSEEEKshoYFD2hQIIQQ\nQpzL2RNZckFCERoUCCGEkDBCw8K/0KBACCGEOJ+SKQVYVJkQQggJMzFvWKBBgRBCCIkeiiRXjPQQ\nCCGEkJgjZg0LNCgQQgghhBBCCCGhE5WGhZycHPU7Iz3Da1EnLQcT4ZLJkigp2Reo59LT08M8UkII\nIcQ5ZGRk5NJZW2r/v2MkhBBCSPi0PyGaT/6KyjUiPRRCCCEkKnU2LS1N7Kj91apVi/RQCCGEkJjT\n/rgcO7odQiQ7O1uOHj0qKSkpEhcXJ04DkRPFixeXQ4cOSWpqaqSH42g4l5xLO8LPJefRqZ9JLBmw\nsChSpIjEx8eLnaD2Ew1+x5oD59E8OJecy1jQ/qiMWMAJFytWTJwOLjANC5xLu8HPJefSbvAzGd65\ntFukgga1n3jC7wZz4DyaB+eScxnN2m8vdwMhhBBCCCGEEEIcBQ0LhBBCCCGEEEIIMQwNCzYkISFB\nXnjhBfWbcC7tAj+XnEu7wc8k5zKa4OeZc2k3+JnkXNoRfi7tO5dRWbyREEIIIYQQQggh4YERC4QQ\nQgghhBBCCDEMDQuEEEIIIYQQQggxDA0LhJCwc+bMGTl27JjhXvUHDx5UPXXDxdmzZ+X48eMSTrKy\nsgzPESGEEGI3Dh8+LJmZmYa2PXnypJw+fVrCydGjR+XcuXNhPSZ03+gcERJpaFgghISdSZMmyS23\n3GJo2z///FNKliwZ1pvuhx56SEaNGiXhJCMjQ2rVqiW///57WI9LCCGEWEG1atXku+++M7Rtv379\nZPDgwRIufv31V6lfv74y8oeTd955R55++umwHpMQs6BhgZAwgJvgcFvaiTns3r1bZsyYIf379w/r\nlBYoUEB69OghL730UliPSwghxJxIt0OHDnEqHcqgQYPkgQcekOTk5LAe9+GHH5bRo0fLX3/9Fdbj\nEmIGNCwQEgbgnf/f//7HuXYgEPg77rhD0tLSwn7se+65R6ZMmSInTpwI+7EJIYQYZ+HChXLhhRdy\nCh0IDEKzZ8+Wzp07h/3YRYsWlRtvvFHGjRsX9mMTEio0LBBiExD6HqiegLf3uNct0H5raQLIDfT0\nmGAf3tII4F3B8dwJtL3n+AKlJ2hjDOYYgfC2T3/n5A5CHAPlT3722Wdy8803ux57nrNnJArG7rlP\nI9uAqlWrSrFixWTFihV+x0gIIcR5QBO8aZh7PQF/OobXNF1xr1+AmkCe6wW87i2s39u6ItD2nuPz\nF5HpPsZgjmFU9wOtlfS8vmjRIqW/pUuXdj3nfs44vrtm4/y8zYGRbcBNN90kc+fO9TtGQuwIDQuE\nRJglS5ZI9erVlZUaPwMHDnQJq1ZP4PXXX5eyZctKkSJFpG7durJjxw7X9jNnzlSvYVvk5D/66KOu\n+gWbNm3K4zEZP3683Hbbba7HP/74o1x++eXqBrZw4cLSrVs3JYZ6ttfG9+qrr6oxVK5cWU6dOpXn\nHOfMmaP2gzHWrl1b1q1b53pNzxi98f7778sFF1ygxnzttdfKnj17dJ0TWLVqlVxxxRVSqFAhKVWq\nlLz88steFz7YZuvWrVKnTh3Xc9o5IwcS54z941ibN2+WW2+9VYoXLy4FCxZU4wtlGw3keGK8hBBC\nogMYzrt06aI0CLoIPdiwYUOuegK9evWSJk2aKO2APmAdoAFjfMuWLdX2eP3//u//pEqVKq76Be3b\nt5fhw4fnOmalSpVk5cqVrhvrJ598UmkPfi655JJcBuxA22N8PXv2lOuvv17KlCkj7733Xp5zPHLk\niNx5550uncXaxt1AEugY3oDOX3311WrM0PePPvrI9Vqgc0pPT1evlyhRQm2L/ezcudPrcaC57rqv\nnfO9996r1g6Yc6zHUHsJPzBA4DpeeeWVcuDAgZC20XT/p59+Cnt9B0JChYYFQiIIRLJVq1byxBNP\nqBtyCCpC31G8xx3cgKKQECz8EN4XXnhBPb9r1y4VqvfBBx+o7T/++GOZPn267uP/888/ygiBG28s\ndFAocP/+/fLMM88EdR5r1qyRn3/+WXnl8+fPn+ccO3bsKG+99ZYa48SJE+XTTz+VUPn222/ll19+\nUYuXcuXKqbQBPeeE/+N1LHhQ8RnjgycBiw5P/v77b2VwwILAExgLsCiBhwXbY+GA3Eic4/z589U1\nxVhC3QbHxpgJIYREB/jeh35BD6BT11xzjTIUuHuwkUoBYwJex/+hYZoBHdvjRhp5+H/88YfSQfzo\nBUZuaCiMGdCf1157Tdq0aaM0Ty/wqMMoDyPHU0895fUcsWaBfkGHoW3BjNEbs2bNUusfGP3x/8cf\nf1y++eYbXec0YMAA5cj5/vvv1etvvPGGa1tPMK/edB+GiqFDh6rzevvtt+X+++9X+9SuY1JSklrr\nhLoNjo3oE9boIE6DhgVCIgiEsUaNGsozER8fryzsjzzyiDIQuIMCfrhhhwC1bdtWtmzZop5HDmDj\nxo2VeMbFxaloBngR9DJt2jS56KKLpF27dkrwIWQPPvigioIIBiwu4FHxBvbVqFEjZVzAGGGJD2aM\nvnjllVeU5z81NVXVr4Cn5rfffgt4TjC8VKhQQS2EMJ8YN+bXWw0FXBPgLU0BhZ1wbGzfrFkz5f3Q\nIkWQH4nii1g4hroNPBb58uULeb4IIYREHnjtYVxHhwN48hMTE1XUHwzOX3/9tet9HTp0kMsuu0z9\nX4sM2L59u9oeBYWhJ/CAo7hgsDWcUDsIUQ4pKSnKIYCoP4T+L1u2TPc+sO7Adr7OEVoMbYVXHmP0\nvHk2AuodQTvBDTfcoMYwderUgOcEB8HYsWOVsQGdKbAWueqqq5QDwhvQfm+6jyiLBg0auM4f6wsY\nfKDdOC4iLeEICnUbLVKB2k+cRkKkB0BILAOrOG5y3cFjTw81hFkDAq3lB2J7eOvd8XzsD9yIw0iB\nVAx3ILr+ahN4gvB+X8BbUL58+VzPeT42gvs+kEqBMWPeAp0TvDsXX3yxrmMgVDEhIUGdAxZ1vq4J\nDBTuj7XnPPM4jWyDY2uLS0IIIc4GkXL4nnfXfnz3Q2/ctd9THzTth8EcEW/uWo/3ekYL+gLboyAw\nDPya8VwDN+Rm6D7OEWN01+lgxugLb2sJLXLR3zlhPIhS0Kv92K+3Vs+eGu7tuUC6r2cb6D6cEEjZ\nIMRJMGKBkAhSsWJF5YFwBzfF8Ljr3R4pCO647w8iDuu4VtAJ4MZaAzURUPMAwuv+g5BFiHOg7fWO\ncdu2bYbH6Av3fWJhAY8EjhXonPA6vAOBijYCCHu9evVy1YQIN2vXrlWeFUIIIc4HN4uIWnPXMITJ\n40ZWj/ajfgC83e7aj23d6xtBV90f46Zb6y6k1XNCRIGnTiJ9IdD2Rs8xmDH6wttaArof6Jy0193r\nWABvtZUA6h5EWveRKulpJCHE7vATS0iYQO6kp+AhRA43vChqhHoJ6EDw7rvvyn333adrnwiVRCHH\nIUOGqNxLiCoKH2pAcLEIQZjlvn371P6R66fRqVMnNQ4UfIRgY19jxoxxpSoE2l7vGHfv3q3CPvEb\nIaDBjNEXjz32mCpuBCMBCiShZgKKOQY6J7wOT0rv3r1VYcb169er6+Ct6CRA6gkqREdqcQEDCEI+\nCSGEOA9P3YfWQLOQ848UPhgIevToocL2r7vuOl37hJ6hECEKFUPnoGfuIEVy8uTJ6kYaN994v3sU\nItYcyPVfvHixivJDCga0WruZDrS9HjAmbYzQWqR8BjNGbyCtAQUbsd4ZOXKkqj2h7TfQOaGGEX7m\nzZuntkc7R896VhrNmzdX0SO+ijtaDc4Law9CnAZTIQgJA8iDRH4fftxBRwSIH4oOIcceN8Zvvvmm\ndO3a1ZVfh5tud6s1wiG1MDoU+Fm+fLnKK8TNM+oXPPDAA/LDDz+4QuxgbIDg4mYeFnC8VytYBI8C\nChmhGBJEDHl9uIlFzQQ923sbnyeowIziRVhgYB/INXz++edd1ZoDHcMT7Zh4z0MPPaTEH5WzURRJ\nzzlpryPHEYUzUacBx/YVotm9e3dV5AlFlHBcb+eMbVH52h28Rwt7NLINmDBhglqcuT9HCCHE/kCr\nEVngmZaHG2E4A6BFuBFGIUJ4yJcuXerKqcdrnpqEKADsE6CoI9YNKN4MDevfv7/STO11aCMM+a1b\nt1Z1HOC1hyEe9RwA9BipF6hVtHfvXlUUGg4NrRZAoO29jc8TOAugedoYca7Yp7aPQMfwBMeEwwDv\ngUEA6x84Ii699FJd54S6SphDvI5Ug6ZNm6r1ljdwbliHYU2COhHezhnpldqawD3KEefqPuZgt0Fk\nB4whwRTiJsQuxOX4igMihDgSWP21QkXEHLCwwM09PEzhAjmhqBSOKtcwTBFCCCHeQNcoFCVE3SU4\nKEjowPhw0003qdaT3oo7W8Vzzz2nnEeIyiTEadCwQIjDgZUelnd4ReDxQAcEtDJi+DwhhBASfcDg\n/Pnnn6vUPrQsRDcpRMBFKm2PEEIAUyEIcTgI1Uco/8aNG1WVabRVpFGBEEIIiU5Q0PfLL79U7QtR\nhweplEg9IISQSMKIBUIIIYQQQgghhBiGXSEIIYQQQgghhBBiGBoWCCGEEEIIIYQQYhgaFgghhBBC\nCCGEEGIYGhYIIYQQQgghhBBiGBoWCCGEEEIIIYQQYhgaFgghhBBCCCGEEGIYGhYIIYQQQgghhBBi\nGBoWCCGEEEIIIYQQYhgaFgghhBBCCCGEEGIYGhYIIYQQQgghhBBiGBoWCCGEEEIIIYQQYhgaFggh\nhBBCCCGEEGIYGhYIIYQQQgghhBBimATjmxISnZw5c0bmzZsnl112mVSoUMH0/R85ckQ2bdok586d\nk5o1a8oFF1yQ5z3Lly+Xo0eP5nm+cePGUrFiRYkk2dnZMmvWLGnQoIFUqlTJ5/v+/vtv+eabb+SW\nW26RggULWjaeGTNmyJVXXikXXnhh2M714MGDsn79ejl27Jg0bNhQLrroInEK2nVp3bq15MuXL9LD\nIYQQR/Dll19KcnKy0hsz2b59u2zevDnP82XKlJGrr75aIs13332ntPDaa6/1+745c+ZI9erV1Y9V\nrFy5UuLi4uTyyy8Paix6z8HJ6J0bQqwkLicnJ8fSIxDiMP766y8l6JMmTZJ77rnHtP3+9ttv8sQT\nTyjBa9SokSQmJsqqVaukW7du8u6776oFi0a9evWUYQHvc+eBBx6QG264QSJJRkaGpKamyrBhw6Rf\nv34+37d48WJp0aKFbNu2zdKFBoR0+vTp0rZt27Cc69KlS+Wuu+5SBoWSJUtK//79pWnTpuIUtOty\n4sQJKVCgQKSHQwghjuCaa66REiVKKA03k5dfflkGDRokrVq1yvU89H/gwIESaW6//XalhXB4+AN6\ngvE+++yzlo0FOp+QkCBTp04Naix6z0EDxnesLXDNnYLeuSHEShixQEiY2LBhg/qBVbl+/frqudWr\nV8t1112nbl7ffvvtXO+HAWH8+PG2uz7wcrdp00YqV64s0Y63c33//feVF2nJkiURHRshhJDoICUl\nRUXf2RHcXJ89e1acTLDn8OKLL6qbdBjiCSH6oWGBEJ0gVBEhi7ipRERDsFSpUkWF4xUvXtz1HNIt\nmjdvLtOmTctjWLAitaN8+fLKsIFoiCuuuEKKFi2a5/2ZmZny448/yqFDh1T4/6WXXprnZrtjx455\nDAvwgH///feSlpYmV111lc/xIIJh9+7dKgWkVq1auSI1QgWpCcuWLZPr/7+98wBvovzj+K+ltKXs\nISBDBJS9BRy4UAHByVIB2du9UBQZojj+4hao7I0yREVkKcPNkL2RjaCAbGihpfk/3xcvJGnG5XJJ\nLsn38zx92iSXu/fepPd97zdvu01FEzhGofz888/KU587d+5s6S4438OHD6sQQnik3J3rqVOnVLQC\nvgMlSpSwLwJheIBnQ8/cuR531apVsm/fPuWpQkQL3nv//fdLZmam+q7kz59fbauBbdavXy/XXnut\nVKhQwe0cwCuD/eAzxnbBjBYhhJBYBtdZeMGRDucaYWiFlI2dO3cqzYUWVKxY0e32eP3PP/9U6wGk\nW7pqMtIHkEbgCB7DMYLUTuhmoUKF3O4buoq1EzSyRo0aTusfs9AzFnfn4GlsS5YsUa9B/zWdL1Wq\nlFoz/fHHH2r9AjBPSE2FzrumFTp+Bnv37lXHKVu2rEp/dQfWKHA8wcmEaEisU1w5ePCg2gbHwnli\nfUCI5UAqBCHkMocOHUJ6kG3SpEn257755htbrly5bE8//bQtKyvL1Om69957bQULFnR6rmbNmrb7\n7rvP9t1339nmzp1r27dvX8DnM2LECNtdd91lu+2222yVKlWy5c6d2zZjxgynbX/66SdbqVKlbFdd\ndZXtzjvvtOXLl09tf+TIEfs2aWlp9v1pzJ8/31agQAFb+fLlbQ0bNrTddNNNtsmTJ6vttmzZorY5\nc+aM7Y477rAVKlTI1rRpU9sNN9xgu/baa23ff/+9LRBatmxp++2339Tfa9asUcdcsmSJ0zZz5sxR\nz+/YscNpTkaNGmW7//77bbfeequtcuXKak4WLVrk9lzxGeBY+KwwP/gbPxcvXtQ9d9pxU1NT1Rzg\nuNjHhQsXbOPGjVOvLVu2zFarVi1bo0aNbPnz57e1bt1afeeGDh1qq127tu2WW26xxcfH2z766KNs\nczF9+nRbkSJFbBUqVFDvx2fSokUL27lz5+zbYI5wTJwbIYQQfTRo0MD2wAMP2B8fOHDAVq1aNXVd\n/vvvvw1P4+uvv25LSUmxLV261Pb111+ra7SmK4GMc/DgwbYaNWoozciZM6ete/fuTvs9duyY0om8\nefMqzbr66qttxYsXt/3www9O+7vnnnvU6xr//POP7brrrlP6gjVFxYoVlW5CP3EuGm+88YY6Lxwf\nx4E+DhgwwBYI0MEPPvjA77G4noO3sb388su2okWL2ooVK2bXeRwXYC2gPdesWTNbiRIl1Nph/fr1\nbj8DjBXrOawFcuTIYXvuueectoM2d+nSxZaUlGSrV6+e7fbbb1efA9acGtDqzp0725KTk20333yz\n7frrr7flyZPHNmHCBK9zQ0g4YMQCIT4YO3as9OzZUwYPHiwvv/yy/Xl4r+HF9kXjxo0lX758bl/b\ns2eP8rBjG1fg/U9LS5Njx46pQoHwjI8ZM8ZwIcSPPvpIJkyYYPeAP/XUU9K5c2e7dx8FCR944AGV\nmoEICtSAgKUdERodOnSQ7777zu1+EamAWhTwtI8bN07i4+Nl06ZN2epToE4BrO2aZ0SzwKOQZSBz\nGkj4KNIa8PnC0wRvxj333CPPPvusigrQohA0EO2BY2FbRAFMnjzZ/pq/c4fjIs0F3iFXhg0bJsuW\nLVPnh88daTPwdMD7sXr1arUNvofIHe3YsaPda4F6HW3atJF+/fqpME7w119/KS8LntMiYlC/w6oh\nt4QQEgkgcg2FiRGJiJoLmi5D+xAB4At4rh2914hSe/HFF1XEHyLZUIx46tSpqnCwEaC18MLjt+aF\nv+uuu5Tnvn379uq5rl27Kk86toHGYAzQEBT2xfl5isyERmJdsnnzZrUN1indu3dX73dc20CjZs+e\nba8dgYi+L774wr6Nkbl6/vnn/R6LK77G9uabb6oICKRCuGolai051pbC+xDV2KVLFxUp6AjWNog+\ngI6DkSNHqvdi3rXzefLJJ1WB6B9//NG+NkPEI/Rc45lnnlHbYEzVq1dXz2EtiP1Az/E5u5sbQsIB\nDQuEeAECM3DgQBk1apQqsujI3Llz1Y2bLxCy5s6wgK4Q2CduaLUbQY0BAwbIfffdp25QweLFi1UY\nP0Lgpk2bZugzg2g5htW//vrrMnr0aHVur7zyitovBHro0KH24yLMD2L13HPPKYMAFlGuTJkyRb3v\nrbfeUkYFgNDAO+64wy6omhEB4X2OBQORUoAfs+bUX7AfLXwVY8fi4KGHHrIX8NSLv3OHxYY7owLA\nwk47NywakC7y2WefqTFpwGjz9ttvq8WHZpTCsWH8wPdVA4tTLFzwWb/77rvsAkEIIQHy22+/qWKA\nKNoLA3NiYqLTzaQewy10Rru5hEEaN7ua5hw6dEgZuZEmuW3bNrcpi76AgR66rtGwYUM15g8++EAZ\nFqDHMIhAt2FUALiRfu+991RaB5wQ7gpHIkUAxQFhHNfGi/D9xx9/XK0FNLB/4JiSCG10dDgYmSsj\nY3FFz9i8AWMCjCIHDhxQqYeYLxgpzp4965TCcObMGac5xHoPY/vhhx/U+WD8cDD079/faW2GlIxm\nzZrZjQwwImAbzagAYFR455131BoO50+IVaBhgRAPQHB37NihBAOC7C4CIBAee+wx5ZkePny4vZij\nBroOOIKb9B49eihv9ieffOJUB0AvuJl1BJ5u1A7QPAbwWhQoUCBb7QTtBhhefHeGBbwf9RIcDQTA\n1dOCisWffvqpEkfcPGNRhvxDx9zEQOfUX1znXWtZiQWDP4YFf+fO9biOaN4HDcwrFkso7uX4HHA0\nwsCbgTHDs+EIFiZY4KCWg7aAJIQQ4j+41kO7EIkG7daM6RrwXuPHH2BYcATXcRgAUMAZEXCImPQX\nGPcdNUPTZERIohkcNAm/XetCoPZP0aJF1evugKEDzhBXDXN9jGOhDhBukKH3MIBj3hzrAhiZKyNj\ncUXP2DyBqMpu3bqpzx3GAUSYwOiPucRvxzUAPgPN0QBggIIxA+sLsGbNGuVg8laTClGKiL5AfSZX\nIwwcEDBwEGIlaFggxAO4WUZ4HUIF3RkWAkmFQIQCwuLQYgrtCvUA7zXEC4WDjBgWcHPq7jlY2QEq\nJkMkXdGeg7C549y5cx737QjEEyIJ6zvaQ+LccSMPz4jWrjHQ9BJtkefaRddTNWjXfWieJ0/n6gl/\n587b5+c6JixM3D3nul+EgGoeHFeQRgNvFCGEEOMg4g6aCYM6rrmuRfaMpkK403uwa9cuQ+P0pMnQ\nDNyoaproSbe86b27/UM7HTUGRg0Yu7HOmTdvnvLMwwgADz7SSs2YK71jcUXP2NyB6AQYQhBBgdRO\nLV0SUSsotuy67nC3RsHYtLn19hlo4DsGUHQbhhRHECnhqYgzIeGCK01CPICweHjZn3jiCSU6jmGF\ngYTtp6amqpvqp59+2ils3Rf//POP+g3PuBFcFygQQYRfat4ShO7D4g6xdhQ6VJXWXncHvOCYB4ik\nY0ioVjnZEYT1wxMDUIOgdevWap7xtxmpEPC0AKQlOOJuLGbi79y51m8wA3ShwKKX9RMIISQ4wCMN\n4y2iCJGeiPo5jul9gYb3B0vvNR1EuhwM05omYTtEDmrgphcarNUecEWLesO+HCMD0LHIta4BIgD6\n9OmjfqCNSPFEWh7SPBHNF+hc+TMWV3yNzZ1Go+4EOk+0a9fO6XU9xhFPmg1gMEA9Jm/boG5EINEd\nhISMsJSMJCSCukKggn9cXJxThWGjoAsDKvp36tTJY3eJo0eP2s6fP+/0HDoqlC1bVlU8dgTdEGbN\nmqXrfK655hqn7gBffvmlev7XX39Vj//44w/1+MMPP7RvgzGikwMqJmdkZLjtCrFp0yZ1TmPHjrW/\nD10O6tat69QVYv/+/dnG9uKLL6ouEWaB8aKa8xNPPOF2LK5dIRw7f4CVK1eq59Hhwd25AlSgbteu\nndP79M6dp+MCrSuEYxcJd9WswenTp7ONa9iwYbaEhAR1Dq64m3tCCCHGukJA99A1AFX6cT0OhL/+\n+ivbc71791brjnXr1jldx7GGcLe96zjRZWDx4sX2506cOKG09vnnn1eP0R0CawJ0K3DsFPHpp58q\nbfn555/dahB0DZ0m0BHBkYEDB6quB9o66fDhw9nWMStWrFD7dhxXIOgdi+s56BnbQw89ZLvxxhud\nttm1a5faZsqUKfbnjh8/rro4OK4v3HUQ0ShTpozqLqaBjiI4h/T0dKft0O1Co06dOqqLlraO0MjM\nzFRrCkKsBCMWCPEB8hsRYo/f8PKjiI4R0P8YxYEQ/o+iTLNmzcpWVwHHgVUchXnwGGFu6DiA4n1I\nEXDtLoCCfSjA5Ms6DxB9gYJQsP4jZP5///ufyhPVvBXIO0QoIAoOIh8fHgLk66M7xZw5czyGFmI7\nFAhESgeKFCI8D6kO8OagurUGuhIg/BB5jfCWYFvUjHBXIMoo8CKgsjb2iYgGHAfnAO+L41jMxujc\nmQmqTaOoGIp04bNAdAg+Z1SbRh4nokEIIYQEDq7xS5cuVddb6DlC6o12bEK3BqQKQkegFd9++636\nQfFox5o7P//8s6oJ4NjNwBONGjVSGg9NQPFHREpifJreYq2BtEToMX6wPti+fbuKKEQXAk8edGgs\ntsE5Y42CNFGkOCICwLGmA6IR4GXHfitXrqyiApA+gGhDbzUF/EHvWFzRMzZ8ruichWKWWEeUKlVK\ndVjCMdCJAtqK+lCTJk1S3bX8iT51BO/H54/ijSjuiIjHBQsWqM8dka0A3UGQ/ol6GFizobgj6n8h\n2gNptQ8//LDBGSTEfGhYIMQFCBJy0h3D1yFCeP6bb75RQu0YOqgXtGXUajXgxtsVtCuE2EPQUdQR\neXv4DXDjjvA7CIojWjtBPaCQEPaBvMITJ06om3qt7ZRjwUqIHBYuqFyMPE8UVHQsPAgxxfw4Fin6\n8MMPlSDD8IFQQSyIsEBCjQotZQGGBXQxwBzOnz9f1bDA9q6FqwIFN/fFixdX4gxjDI6LPEUUStJC\nVt19xqBQoULqea0GgrtzRT0Ircijv3Pn6bgABhm8hraSjtx8883ZakRgbl3Hhe8OFik4NhalOH8s\nhmAQ0ypME0IIMQZuNh2NB2g7DOMCnA2ozo8bTqMFIbEmwNoCaXy4qURYfs2aNbPpPdIY9LSghHah\nMwKKPcMggRaSSOt0rO8D7UUNKbRcRucppF3AEI4bdW8ahDQQGOpxzlijYF7Q+hhGddyoa9ug7oC2\njkEdBHRIatu2bTaNCwQ9Y3E9Bz1jw5oPc4jtUN8A6yz8wCEERw8cRVgvoEYUDBModumYEuP6XdGA\nFmv1M7R1GdZJqPOANQrWSzAwoH23RsWKFVU9CowXLS2RmgunE4xZLMhMrEYcwhbCPQhCiP+ggBQ8\nEVgIoJ+2J7TWibjp1NtOiRBCCCHWATebMMijRbQ3cBMNAwKiGQkhJJQ498khhEQMaFkEq7o3owIh\nhBBCIh8YFVyLSBNCiJVgKgQhEQrC45DO4Atv4feEEEIIsT6+IhV8heETQkiwYSoEIYQQQgghhBBC\nDMNUCEIIIYQQQgghhBiGhgVCCCGEEEIIIYQYhoYFQgghhBBCCCGEGCYqizeix+uJEydU0bq4uLhw\nD4cQQgiJCtChOj09XfW8j4+3lm+C2k8IIYSET/uj0rAAo0LhwoXDPQxCCCEkKvn333+lUKFCYiWo\n/YQQQkj4tD8qDQuIVAC/79wiybku/U0IiWxOX/hTSuUpGO5hEBLTpKelS+XSN9t11kpQ+wkh3uA6\ngpDgan9UGha09AcYFZJz5Qr3cAghJpCRI0ly0VBIiCWwYpohtZ8Q4msdcfTiOfV36TzWirgiJBq0\n31oJkoQQQgghhBBiMnkTK6gfsP/MMfVDCDEPGhYIIYQQQgghMWlgIISYAw0LhBBCCCGEkJhCMy4Q\nQsyBhgVCCCGEEEIIIYQYJiGc7So+/PBDWbJkieTIkUPuuOMOef755yVPnjy6XieEEEIIIYQQQkgM\nRyw89NBDkitXLvnf//4nr7zyikyfPl3at2+v+3VCCCGEEEIIIYTEcMTC/PnzJWfOnPbHb731lrRs\n2VIyMzMlISHB5+uEEEIIIYQQQggJP2G7Q3c0GoDNmzdLyZIl7UYDX687kpGRoQwOGmlpaUEbNyGE\nEELCD7WfEEIIsQ6WKN64Zs0aefPNN+Xdd9819PqQIUMkJSXF/lO4cOEgj5gQQggh4YTaTwghhFiH\nsBsWVq9eLY0bN1apDq1bt/b7ddCvXz85d+6c/QeFHwkh0cHpC9vVDyGEOELtJ4QQQqxDWIsV/Prr\nr3LfffepAo1du3b1+3UNpE24pk4QQiIbR2NC6TyFwjoWQoj1oPYTQggh1iFshoUffvhBWrVqJcOH\nD5c2bdr4/TohJPqhQYEQQgghhBDrE2ez2WzhOHCpUqXkxIkT6rcjy5Ytk2LFivl83Rso3ohaC2sP\n7pbkXLmCMn5CSPAjFmhYIMRapKWlS9kidVXaIVpCWwlqPyHEX7jWIMQ87Q9bxMLSpUudOjloaIUX\nfb1OCCGEEEIIIYSQ8BM2w8I111wT0OuEEEIIIYQQQggJP2HvCkEIIYQQQgghhJDIhYYFQkjISDt3\nTvbt3qN+E0IIISS6oe4TEjvQsEAICTqol/Lea2/ITRWqSeM6N6jfeOyujgohhBBCIhvqPiGxR9hq\nLBBCYoePhrwt40aMlOu6tJQra1aWQ+u2yLjhn6nXnh/4ariHRwghhBAToe4TEnvQsEAICXoY5ORR\nY5VRoWabe9VzRauUF3S6nTJ6nDzW5znJlZLCT4EQQgiJAqj7hMQmTIUghASVI/8clrSz51SkgiMl\nalWRc2fOytHDR/gJEEIIIVFCpOj+6Qvbwz0EQqIKGhYIIUHlimJFJVfuFJX+4MjBtZslJU9uKVL0\nCn4ChBBCSJRgdd2HQUEzKpTOUyisYyEkmmAqBLFEyBys20qIGBIfdeAzfbR7F1VTAekP8FhgcbF6\n7Czp/FhPfuaEEBKDUPujF6vqvmOEAg0KhJgPDQskrBWDUdwH+fcImYN1G0L0dL++kpDAr2Y0gc8U\noKbCitTPlccCiwvteUIIIbEBtT82sKru06BASPDg3RsJG6wYHDvAUITuDyjUiNxKhEEyOoUQQmIP\nan9sQN0nJPaIsyFGKcpIS0uTlJQUWXtwtyTnyiVWJZbDAHHuN1WoJjU6PGDvFADWTp0jGyfPkV+2\nbcg2J1aaLyuNJRph7iMh1iQtLV3KFqkr586dk1wW09dI0P5Y1w5/td9K82WlsRD/4bqCkOBrP4s3\nhikM8L3X3lDi2rjODeo3HuP5WMGfisFWmi8rjSUa0QoqIVSR4YqEkGiB2uGf9ltpvqw0FuI/LNRI\nSOhgKkQYYBigc8XgolXKe60YbKX5stJYogl6Eggh0Qy1wz/tt9J8WWksRD8s1EhI6KFhIQyhdChW\nCIHSwgAhrshIQYEb5KDHQoid3orBVpovK40lWqBBgRAS7VA7/NN+K82XlcZC/IeRj4SEFqZCWDgF\nINpBZWAsJJBX+VWvAeq3a8VgK82XlcYS6TDlgRASK1A7/NN+K82XlcZCCCFWhxELFk4BiHb0VAy2\n0nxZaSyRjFZDgRBCYgFqh3/ab6X5stJYiLE0CEJI6KBhwaIpALEEzrn01WUsP19WGgshhJDIgNrh\nn/Zbab6sNBbiG9ZVICS8sN1kGEAlYRQDQn4eQulg9W7XrbMKA4Qln1h3vqw0lkiFEQuERC5sN2kM\nakfkzpeVxkLcQ4MCIdbQfhoWwgiKAnlKASDWni8rjSXSoGGBkMiFhoUA54/aEbHzZaWxEGe4riDE\nGtrP4o0WCAOkQAU+XxD8fbv3qN+R+NmFevyEnDuXJnt27VO/CSGhgbpv3nxR9wnxD+o+CTY0LJCI\nBiGK7732htxUoZo0rnOD+o3HeD4SiPTxk8gD360hAz6Q6mVulRuqN1O/8ZjfOUJIJBDpuhnp47ca\nLNToG+o+CRVMDiMRB7wUaAGFas3D331fxo0YqXpMox0UKjejyBJA1Wmrg7zNSB4/iTzeGfyJjPpk\nggxqXFtuLV9cftz5twz8eIJ6rd/gZ8M9PEIIyQZ1n3gyKLDLlG+o+yRUGKqxkJGRIX/++accP35c\n8ubNK+XKlZPcuXOLVUhLS5OUlBRZe3C3JHvJAyGRhVZAafKosaqvdHJKLrmYeVHqdG0pNdvca99u\n7dQ5qi/2L9s2WDrNBAsleCpqdHggIsdvFOZChjcMEhEK/e+oJi80rG5//t3F62XI0k2yfs+PkpLC\nayYJTY2FrKws2blzpxw9elTt6+qrr5YCBQoY3h+1P/qg7hN3cB2hH+o+sWSNhYsXL8r06dPl7rvv\nlvz580uVKlWkQYMGUqNGDbUQuOmmmyQ1NVXOnj1rygkQ4sm7jxvxB1MHS8UH75SMCxeUp98RtINC\n5WYUWbIyiLqAgSRSx08ij8N/H5Gz59JUpIIjt11zpZw5myZH/jkatrGR2GHhwoXSunVrKViwoFSo\nUEGtH2rXri2FChWSmjVryv/+9z9lbCCEuk9IYFD3SSjRZVhYu3atVK9eXd5//31p3LixLFmyRE6c\nOKG8DadOnZI1a9ZI586d5auvvpKKFSvKN998E/yRk5gC3n1EKiBlAN79olXKy3WdWkiOxJwqfcAR\n9JhGOyhUbrYySOXIlTslYsdPIo+ixa+Q3Cm5VPqDI8v+PCR5cueSK4oVCdvYSPSzf/9+5ZB49tln\nlVNizpw5yoAAxwWcElu3bpU+ffrIihUrpGrVqspZQWIX6j4hgUPdJ5arsYCwh3Hjxsn111+f7TWk\nQlSrVk39dO/eXYU1bt68ORhjJTGMO+9+QnKSlGt4g6wcNV2Q0QNPP27KV4+dJZ0f62n5NAKM79Hu\nXVRNhUgcP4k8kObQpXdbVVMB3zlEKsCoMGjhWunxVEemQZCgAuNB//79VeRj9u9miopewM+jjz4q\n//zzjyxevJifSAxD3SckcKj7xPI1FqwO8yyjD0/1CNZM/lrWjJstORMTJe3sWeXpb9etszzdr68k\nJCRETP7olNHjVPpDpI3fCMyNDP93DoWcxqVOVekPiFTo3KutvDTgyaj9zhFr1lgwG2p/dEHdJ57g\nOsI/qPskVNpPwwKJGNCOCd79Ol1aZvPuP9bnOVWTAOkDkejpxwIqksfvD6zkbJ2CTqipgPQHFmwk\neqFhgYQS6j7x1FqS3SD8h7pPLGtYmDlzpgwfPlwOHDigai1otG3bVgYPHizhhF6L6MRK3n3H1lfR\nbggI1tzQwBDaxQQKOCHXkkYEYiXDwrJly+Sdd96RXbt2qWu8BmoxTJgwwc+xsSNUtEHdjwyCvSai\nQcF/qPskYgwL69evlxtvvFHlSqJYY1xcnP21smXLqqrO4YSLi+gmnN5919ZXKL6IOgnRnLoQ7Lmh\ngSH4aQ9jR0xV3SBy/1djgWkPxAqGBRRuxJrh8ccfl3r16kmOHDnsrxUtWlR1i/BvbDQsRCvUfWsS\n7DURDQrGPhPqPgmX9hv6r0cXiAceeED69u0byBgJMQSMCaWvLhPW1lfoToFCkujogPQM8PzAV2M6\nusHI3IC8iRWcFg/EPLC4GPXJBBnUuLZqMYluECjcCPoNfpZTTcLKtm3bpHLlyvL222/zkyBRpfux\nov1G58YXNCgYh7pPwomhiIVffvlFXn31VdV20orQa0FCWUBy7dQ5snHyHPll2wanxUMsRTf4Ozeu\nsBBTcMIgq5e5VfrfUU1eaFjd/vy7i9fLkKWbZP2eH5kWQcIasXDo0CGV8rBjxw6naAWjUPuJFbQt\nVrQ/UN13Bw0KgUHdJ+HW/ngjO8dCoFy5cvLMM8/IwoULZenSpfaf7dvpeSTWFsJ9u/eo32a0vgIo\nJImaD0jPcGfJh+g+mDpY/YYlH89HG/7OjeMigtEKwQE1FZD+gEgFR9BiEt0gULiRkHBy5ZVXqjbV\naC/53XffOa0l1q1bxw+HRJzux5L2G9V9X6AoIwszGoO6T8JNglGvwIYNG2TVqlUyadIkpxoLnTp1\nkqFDh5o5RkICxgwPggpnzJ2iQv2KVilvfx7dKVBIEjUfNLCAwbEQHqhZ8vEeBAih+CS6WERTaKQ/\ncwPolQg+KNSImgpIf6h/1eX5X/bnIdViEt0gCAk3K1askK+++krmz5/vFLVw2223yaxZs8I6NhLZ\nhFr3Y037/Z0bEnyo+yQiDQvwLJw8eVL++ecfueKKwC4cR44cURf4ggULun0dxzl+/LhcddVVEh9v\nKMCCEFPyALEYwKIE78MiwbXlpeNiwZslf0Xq58qSH6580WDgz9ywWGNoSPmvUCNqKuAzQaQCjAqD\nFq6VHk91ZBoECTuISkB0AlIhrrnmmnAPh0QZodb9WNN+f+eGBB/qPolIwwIuINdff31ARgV4KNBV\nAsaJ8+fPq9SK8ePH2ztKXLx4UXr37i0TJ06UvHnzSlJSkkyePFluv/12w8ck1icYxY4C9SA4jgme\nDoD3YZEAqzwEVHs+li35eucGMMwxNK2h0P0BDEmdKi/PXaUiFWBU0J73F7avImaCazCKN9KoQMzW\n/nDofixqvz9zQyJT94M1ThKdGDIsoAUUijeeOnVK8uXLZzj88YsvvpAqVarIhQsXpFu3bvLII4/I\nli1b1OvDhg2Tb7/9VtVsQLQCqka3bt1adu/eLXny5DF0TGJdglnsyKgHwduYsCjx1vIyFi35+Jzg\nBfI1NyR0raHwfnR/eLZvL1VTAekPRhYFbF9FggH0H1GLe/fulTJlosOLS6yh/eHQ/VjUfup+9Op+\nsMdJohND34rNmzeriIKKFSvKjTfeKImJifbXGjVqJF27dvW5jzfffNP+N97/0EMPydSpU9WXGF9W\nRCd06dJFGRXAs88+K2+99ZbMmzdPGRhIdBGslkWBeBB8jclXOGOsWvLD2RYsXARizTerNZS3MeBx\nmbKl/RqXGWOkl4N4A44DRCQiUvHWW2+VFIebrurVq0u/fv04gVFOsLQ/XLofq9ofqO5HahFnoxpn\nZktIT2MIVPeNjpO6H9sYaje5cuVK+frrr92+Vr9+fbn//vv9HgiKPu7atUt+/PFH9RiLjZEjR0qb\nNm3s29SpU0cefPBBGTBggNN7MzIylEHCsbhk4cKFZe3B3ZIcYDssEpkti1x577U31OKgTpeW2TwI\n7hYvZo4J+6IHPzpbSwZqzTejNVSwPQpGxkgvR/RiZrvJnTt3yrhx49y+hvQIrAu8Qe2PbIKt/eHU\nfW1/1H7vRGoh50A0zqyWkFbTfup+dKNX+w198+rVq6d+zOLjjz9WaQ+//vqrk3HANeUBxgackCtD\nhgyR1157zbTxkNDmR4ai2JG/HgQzxxSLHvxYIVCvg7fWUMiNRBijL4+DmZ4Pb2MsVzivnLuQKSmJ\nCT7HGOwxBQo9KtagfPny8sYbbxh+P7U//FhZ+8Op+4Da751ILuQciMaZofuBjiEY2k/dJyDsbRbe\nf/99tThYvHixVKhQwf48ukT8+++/TtvicaFC2S9ACJeEwUH7cX0fCR6wUMIrACt/4zo3qN947BhB\n4k/IoiNmFjvS8gDhcVi0Zrn6jceerLqhGBOJbHBzCk8BRB3WfLR0xO9BjWvJuNSp6nV/WkM5orcl\npBlj8Ab+j8eP+kJyxMfJQxMWS4lBU+WVuask82KWxzEGe0yBns+QAR8oL8wN1Zup33jsz/WKWAdq\nf/iIBO2n7lufSDQqBKpxgeq+GWMwW/up+0TDcKzMzJkzZfjw4XLgwAHJysqyP9+2bVsZPHiwrn1g\nO6Q7LFu2TCpVqpQtpQJpEVoo5OHDh2Xr1q3qeVdy5sypfkhstHMKBL0ehFgrwBQqIjWPMlheh0Bb\nQ5nl+fAEPBDjU6fIkGZ1L3tE5v8hy/celt/3HXU7xmCPKRCs7lGJRaD/77zzjkqFdLwpbdCggUyY\ncOmz8QS1P3xEkvZT963L/jPHIs64EKjGmdES0mraT90nARkW1q9fLx07dlTtIlHAMS4uzv5a2bJl\nde2jb9++kpqaKtOmTVOPYTTQ8iphZX7hhRekadOmUrduXVXYCUaI6667ThV4ItYg0HZOVi92ZMUx\nWbGdZyyEPfryOsBbYMTroKc1lLewfbPG4A5XDwTAMfD/3X/eH9LtiQ5u21cFc0yB4O18MPeooM02\nWqHl6NGjcu+998rjjz+uij7nyJHD/lrRokVDPBqil2jWfquNJxq03xN5EyuodQGMC5G0NjBD4/S0\nhIwk7afuk4AMC2vWrJEHHnhAGQeM8ssvv0jx4sVVtwdX70WxYsWkYcOGMmPGDPnwww9VVMMNN9yg\ncjHj48OevUGCkI9oxZZFVhyTFdt5xppBwUyvg7fWUFrYvrfCTGaNwR3ePBCZWTbp3OMRt9+dYI4p\nEKzsUYlVtm3bJpUrV1btpEnkEM3ab7XxRLL26zUugEgyMJihcd5aQuopgmg17afuEw1DVxVEFYwf\nP14C4aeffvK5zX333ad+iDUx2s7JaMhiuCzu0VCAKZjtPGOlA4Q7L4Ier4NeHFtD4RiD+v5Ppk+a\n7TNs38wxmOWBCNaYAsGqHpVYply5cipqAe2rHaMViLUJpfZT9yNT+wMxMMSS9rvqPvaP2gZIQ4g0\n7afuE8PtJgHCFtGloVmzZpKYmGh/vkSJEk5FGMMBOkqgHzbbTQYff9s5RaPF3eqEop1nLBgWvHkR\nLlzIyOZ1COQYY4ZPkfPp52XIPXV1t6PCosSMMTiCiImRH6MmQa1sHhE9NQmCMaZwng8xt90keOut\nt1R6Zfv27ZVuOxZwRhqkP1D7o0f7qfuRpf1GDAKaMcHqa4Zga7/r/lEwEbUNIlX7qfvRSVDbTUK8\nN2zYIKtWrZJJkyY51VhAscWhQ4caGzUJKsGw/IciHzFSLO56CbUHJhTtPGMBrfDfK3fWkEpFC8jW\nwydkiIMXwYwweu0YTzaoLO8u2eBX2L6j58OsVouBeiD0jilUWNGjEuusWLFCvvrqK5k/f75T1MJt\nt90ms2bNCuvYooFg6U2wtZ+6Hxna72hQ8NcgsP/M9oiIVgi29jsWFUZrR3RhiGTtp+7HNoYiFiD2\nr7zyivz8889yxRXWa7tHr0XoLf9YvAQjHzGc3nazCZcHhhELgQNBrnbVLVK/RAFZdeConDmfKXmS\nEuS6UkVk1aETsmHvTwF7CnAMtD/sf0c1eaxBZdXe6dVGl4sn+fJa+EJP3makRB4ESrSdT6RGLKxb\nt05uv/12WblypUqxDHxsjFYMtd4EQ/up+2L5eQzEoOCIVlfBjH1FovY76j60/tyFTGp/kKDuWzhi\nAbaI66+/3pJGhWjCLE+DGZZ/X2MJVh2CSPW2u5uvcHlgQt060/HcjTe0tRaw8p9LS5ff9x6RQXfX\ncWq/dD4zy5TCf44Fk1ISE+SxBlVkwLw/5OiZNGlWpbQsx7EDKMwUSKtFq3kgAiXazidSwfUIxRvN\nMCpEE2Zofyh0P1jaT923rvZ7Mij44w23siEh1NrvWigR2t/9hkoyYN4qOZ+ZKY0qlgq4KCO1/xLU\n/dBgaNl/0003yauvviqnTp2SfPnymT+qGMdMT0OgbaHCnecYjCJRwcTTfPV49mnT2nNZNWVFO/ep\nI8cooUxJSZbWXVvKkDf6RHwtjLz58qq8Ryws3LVfypM3T8DHcCyYVKdkYcmy2VSa2dClG+WDZZsk\nPiGH9HiivaGwfbZaJFakSpUqcuTIEdm7d6+UKWM9A3GoMUtvqfuhxaq6b6b2ezIoBBIJFwkEW/td\ndX/A/NUy6retcuGiTQYvWCsD56+R3CnJhlP2qP0k1Bj6r9+8ebOq4lyxYkW58cYbnYo3NmrUSBV2\nJMYx07MdqOU/3HmOofa2B4qn+Tp14mRYIy9C0UIL5z5pWKoMauLgEU/9QvIlJkd8YbzTp07LxSyb\nx/ZLZ06fkcJFCgZ0DMd2TfO27FcRCq85ekgWrFHtdo0s1thqkViR7du3qyLQKNJ46623OhVvrF69\nuvTr109iCbP0lrofWqyq+2Zov6+Uh0C84ZFAsLXfne47RUYsWC3tuz1seC6p/SQiDAv58+eXhx9+\n2O1rTI8IjEA9DUY9/u5CHs0ei5W97Wbgbb6+mTTLEpEXwUpZwbkjUgFGBVerPor/oFdzJOeyh6pV\nITwSGRmZMurTiU5VoQOdS7ZaJFYEeZroLOWOkiVLSixhpt5S90NHJOi+Ue3XjAqe0hViwRseCu30\nqftjpsuL/Z8wNJfUfhIRhoV69eqpH2I+ZucW+vL450xMVG2j3IVeWiXPMRTedjPwNV8Pd+4gs8ZM\ni4jICyPn7pgnqKeScSTh6FXA5+fafsmsxRO+6526PyypH08wdS5DNX5C/KF8+fLyxhtvcNJM1n7q\nfuiIdt33VgMhFrzhodDOYOl+qMZPiN+GhX/++UcKFy6sOwT3r7/+ijlvg5VrCnjz+HsLvVQeEotY\n24PpbQ/VZ/fCoFclX/58lo+8MHruofDox0KrQqMeBl/Fs9hqkYSb48ePS1JSklPKgzcOHDggpUqV\nkljAbO2n7oeGWNb9WPGGh0I7A5lLaj+xErosBWgr2bdvX+ndu7c88sgjUqJEiWzbnDhxQmbPni0j\nRoxQoY2DBg0KxnijnmDUFPDk8dcTehlJ9Q2s/tnlzZcvIiIvjIDzaNujqwwclhq1VnH8HyHPEeGd\nwWxV6K+HQW/xrFCNnxBP7Nq1S5o3b67qMLVt21auvfZaty0j586dK6NGjVKplZMnT46JCTVb+6n7\noSEadd+xroI3YsUbHgrtNDKX1H5iReJs+AbrYO3atfLmm2/K119/rQwLWBCgI8TZs2dVRWcUYbrt\nttvkxRdflCZNmkg4ifRe1lqFYdzcnztzVlm423XrbHonhn2790jjOjfIg6mDnSzthzfvlK96DZBF\na5bLlaVKhmQsVmzTaWSf3j67jAsXTB+XlcC5vzv4JZk19ks5czZNWdk794qe6tChRFswjEud6nMu\nhwz4IHvxrAVr1GIkGopnkcjsZe2J3bt3q7XEtGnTVOFGdIYoWLCgpKenqwiFTZs2Sa1ateTZZ5+V\nNm3aqM4o+sdG7fcFdd987Y8W3fdVqDFQrSLmziW1n1hR+3UbFjSOHTsmy5Ytk40bN6qwxjx58qju\nEDAqWCVkMdIXF46CFkwLN/Z/U4VqUqPDA/aIBbB26hzZOHmO/LJtg1MhR6tb24PRGtPoPh3nC3Us\nwtmyM9QLk8LxuegRNwmEOHrzkOD16mVulf53VLMXfALvLl4vQ5ZukvV7fowarxGJDsOCxpkzZ+Sn\nn36SNWvWyNGjR9W+ypUrp7pDuItk0Dc2ar/POaLuB037I133od96DQr+ahUxdy6p/cSq2u/31a1Q\noUIqlBE/JLJrCvgTemn1+gbBao1pdJ+O84XimOFs2RlqIISRXrApUuYyFopnkegETommTZuqH+JM\nMPWWuh887afuU2tCtYai9hOrYk2zqcUJRqh9uLBSK8dA5jXQVl3BarcZjJad0fT9I/pxV6ApmAWf\nCCHRed2l7gdX+4PVqjuY30G9dRVIaPGk08Eq8kxIoMQHvIcYAqFx8D4jfQC1CfAbj/F8pKIVeELa\nA2oq4DcehzJUz4x59dbyCTmPCE/099hG92nGuPwdK4le8PkilxIpDzdUb6Z+4zGetxd8WrBGpT+s\n2HdE/R4wb7WcP58hH7ydmu374W1/hJDov+7Guu77On6gum2m7vsaqxkGBS0FwmgaBDEfXzrtUfvn\nr5aKVa+VxMScfu2PELNgxEKYQ+2t4i0xEnpp1pjMmFdfLZ9y582jila5jvW9QW/IJHgWujof+9SJ\nk9Lz+aclITGn233mTErU1f7LzBZi4fz+GcXVOk5ruf+gmFO24owfT1CvoTij1vLqtU8mSvrcVZIr\nZw55pE45qVS0gAx22E7v/ggh1rnuBlP7Y1X3g639ZrcODcZ3UItQCJYxwVHrAb3k/qFHp6H9v/60\nUl6d94dczLJJ7sQEaXB1Ufl19Sb1fuo+CQd+F2+MBIJRwMmfgkdWL0hopTGZMa/aeMYNS5Usm03q\ndmttrxfxx5iZUq1mDdm+ZavTWB9/6QX56I23ZHzqKKnf4+Fsx145crrYxCaSZZP4hBxO+1w1ZqbE\nS5ys3LNN12euaiwM/0zqdGmZrY6F3kVBuL5//qJ5PvCZvD34ExmTOk3Ne3JKLqlS7VrZtnGH15aI\nxFiBJmxX7apb5PEbrpX+jWtLSmKCx+1Y7JFYoXhjJGh/OK+7VtN+q+m+NqZ2Te+XDWvX6dJ9LQXk\nvUGvB137zdB9M+fKzEKN3nDV/oScCZIjTuT8hUzqfhB0H9u9dFtleaR2OSmeN0VpP3WfRFTxRo3h\nw4dL7dq15cYbb1SP0SYKz6GNVDTiLbQNtQkQ2haMYkuh8pb444Uwc0xmzCvGM3b4Z1K1ZRM5e/S4\n/DF2lqy48LnyLGBxsWnDxmxjXfHzr7Jx/QaxXcxye2xbVpbUbHefrP/iO8lVuICsHDVdbZszJVnK\n33Gj7Fjwk+7P3Ix81nB9/4yChcVnn05SiyqMefln02TDHxvl9abX0UvuB54KNF1f5grVjmr/ngNS\nscq1artzaenSvMbVdqOCuyKOvgo+7dtzQCpVMVaRnxAjzJo1S3LkyCEPPvigvVtE//795e2335ak\npKSwTmo4r7uh0P5I1n3NQLBh3XopWvUaWT1+tqxI/1zicsTLFVdc4Vb3NSaOGht07TerjkUka/+x\nXftl7w+/yWtN6lD3/cCXTq9ZuV5q16th3+6uCiWlXOF8hnVf2x9rLpCw1VjYvHmzpKamyvXXX29/\nDq0m0Z96xowZEo04hrY5YjS0TQ+uBYAQUoffuGBDrPB6oPibu2f2mAKd19OnTqlIBbD+87my75fV\nUrV5I6n+SDO1WIXHwt1Y169ZKzXa3is5cyW7PTYWEdd1bC71urWWC6fPSq1H75eE5ERp8/kHUrBs\nSb8+czPyWcPx/TMKrOjwVmCeMd+FypWSY1t3KaMCrO8oNITfgxrXUv2asT1xj2OBJpB5MUtembtK\nmo1cqB43u62NypMsVKSg03aeCjm57s9xuxzxcdLs1kv7Y94lCQUnT56Up556Sho3buzULQI/77//\nftg/hHBdd4Ot/ZGu+xjnO68OUlEHWRmZ8u/2vVLlgTvloSlDlVYfOXLE7Vgnjxonk0aOkdodHgy6\n9ptVxyKY38H9Z45JsLQf67B9P66UwXfXoe77iTudhva//O1KpdMtm3VVkQrjR30hKbmSA9L9BIf9\nUfuJGRiKWFi9erXUqFFD4uOd7RLXXXedrFixQlq3bi3Rhj8tmswiFJZqf70QZo/J17yC7ZshqHFS\n+uqrss3z0IGvq/QHLAK08SNcsVzDGyT9vxtWt16Ji1lSpHwZqdqikdre8dirRs+QIhXKSkJykv28\nilxztWSmX5C1n8+VzTMXGPrMA2khFo7vnxHPV66C6XL4zBmn78i5f0/I+bTzbInoAW91J+wFmj6e\noD73zf+ckOlrd8lrd2eP/HDcDp4ILBoGLVwjnXs/at+v6/607QbOXy1t65SXqsULst4CCRlbtmyR\n0qVLq/QF17XEhAmXvtfhJFzX3WBrfyToPmoj5MmXV86cOp0togLjR30EpDI46j5SF66oVN5jNALG\nCkrVrS6Zaekh0f5AW4cG6zuYN7GC+r3/jPFaC67ahb+17wl13//503Cn0zAq/LrnsAxpVvey9o+Y\nItVqV1EFHI3oPoo831y2mLx5bz3WWiLhNSyULVtWGRAyMjIkZ87LlUd//vlnueuuuyRaCXWLJrML\nAJnREikYY3I3rx179ZDMzItSv2wlybhwQeLi4yVHjnjp2LuHPNv/FWX1x/i/mTFL6nV/KNv4kQ6B\nvP64uDi3Y0W45PG9f0ndrpeMYGsmfqWOjeeLVb1Wju7YI5np5+0ejGN7Dqgx7Ph6cdjacVqpRZhr\n3u3UkWNUqF1KSrJ06Pawmntt3lMKF5CkXEmGWiJGM8rrNvgTGTtiqte6E1pxxjdGTJH0tPMy5J66\n9rxLzCe+70NSp8qaPxer5/A3whvhicjMssnEUV9IzpwJ9v1q+9O2Q7HHZ26rpjxLCTni7ft7tm8v\nhkaSoHL11Vcr48KxY8ekUKFCTmuJMmWsEd4djutuMLXf6rqflZWlIihwgwo9hpEAetK+R1f1HqwH\n1Pi7Zh8/dLxa6xzqfe7Gmit3brHZstRr1H7jBgZP2vXkC93s3xNELFD3jWu/q04jUgFGhWzav3ST\ndO7dToaMme6X7mM7GBXm97xb6b7jWoLaT0JuWEBdhQIFCkizZs2ke/fuKg/yq6++kl9//VXGjBkj\n0YoW2gbhhYUeYhpMT3GwvSVGvBDBGJO7eR3+7vsyfsRoqdutld0jsXL0DBn7aaqKlMH2Xsd/4XNp\n1e4RyZc/v9ux1qhdS9ZO/Ep5OK6+pZ7kSE66tChp0UjKNbxRvuo1QFaN/1I2zpgvxapdK+smfS0P\ndXxU+r4xKGzRAaH+/ulBeY6GpzpXLh4xRarWrqIKaGnzXqhSOen/3R8uVvW10uOpjjF786q3OwM+\ndzxu8fC90rB+c4+RH8f+PaG2y8jIlLEjJsurd9VUuZeu+9X21/Lhe+T2+i1kbvfGcmv5Kz3mZxIS\nLIoXLy5NmzZVDolnn31WChYsKEuXLlWplqtWrbLExIfjuhtM7be67o8bMcopkgK6X7Di1faIitYd\nHvU6/vVTv1X67qg/rtEQ2nlQ+50NDFqniEC0q2uvNpL6yUQ1v1fdWk/6z/uVum9A+zWdxk0+aiAg\nXcGT9nfu8Yh6rEf3HfeHSAUYFVz3R+0nITcs4Mbu22+/lRdffFGee+45OX/+vKq3sGzZMilcuLBE\nO4GGtlnFW+LNC5GcO0W1ajIyJteCUHoLRGnzqnlUYFRwF4mA17AY8TZ+FG58YdAAyfXfTavrWC91\nhXhbxqeOtBdmqtj0VqnTqYVs/voHFZ2Amg2oDH1i+15p372LtO7YPqQtQH3NU7jB+SNSAeLozore\ntVdbmTR2hpr3XLlzSfW61dTzEC5EKsCooFnRo7kNpbtzw3PwVridOw8egzJlS9nzJD1FfmC/k0ZP\nl8FN6vjc71VlS6v9rdh31MmwEOuRJCS0wBmBYo0DBw6U48ePqzTLBQsWSKVKlSz1UYT6uhss7Q+W\n7rtqIjCi++4iKWD0r9n+AXXszk/09jh+RCp0QGTDqy/LsHeGehxrZkYmtd8gvrRLi5wb+9k0OXfm\nUleIQd5z2hwAAIojSURBVIvWyXk3uh/N2u/pvPzVfvyNworetD9P3jy6dV/P/qj9JBAMd4UoWrSo\njB8/PqCDk/B6Szx5IZBnmJV5Ue6sWc9tOylPY9IKQmntqBC+WLFKZbctn7wVMPIViZB2IcPuVfHm\nRcmbL696n6f5e+mNQZKVdVEmpI6WixmZsuWbxbJ9/k/q3Fu1byvP9n9ZTh0/IV+MnySfj58o44Z9\nps6hbddOqubD1DHjLNEGLFzgc/JWabhLj0fkpf5PKOs3hEprj+T4GN8ZFAzylQ4QbeGOvqo0u/MY\neMqTdIz82LNrn+796tkfIcEmd+7cqlCjFYo1xoL2m6377lpR4mYyLi5epS2Yovupn0uhq0vJuTNn\n5ezpMx51v1OvHkrXgbe5o/Ybx5d2IXLu1cHPynN9e9m1/tLne1n3/UkFjDR8nVcwtP/0qdN+7ZPa\nT4JJQP+9SH/47bffpHr16nLHHXfI3r177e0nSWR4S1y9EKrGQLUKqijSPxu3+2wnhYucp4JQqsXg\n2nWqBoI/7am8eVRyJOaUxJw57Tmder06jvPn6FVJyJkoORISnFIusMBCCkWhwoVl3KcjZNLoS14U\n7XWVjpEjh9N7/Gm7FWikQ7giJVzB8X1ZvSFgroLm+FhvOkAk4u3c4EHw12MAo8zDjz6oUh20fEpX\nD5Bj9Wc9+3XNu3TnUSIk2CxZskR++OEHVWehW7dusnz5cmnUqBEnPkja7073r6hYVup2f1j+3b7b\nL9131f5juw7IrsW/qRoIZum+VudIq+dgRPfDrf1m6LYVtF+vxrhqvevNcrRqv6/z8lejtciHp17o\n5lGrL1zI8Hs9Qe0nwSLO5qoQOmnTpo2sXLlSihUrpgwLH3zwgdStW1cWLVokJUqUkHCSloYicimy\n9uBuSc5Fr5suC2u/gUpEEfqPNkzoloDiRuunfycbJ89RrZI8eSewGHikUweZNm6C1Oz4oApjROHD\nSQ88JnU6NbeHNYK1U+dk2587EPkwdliqXNe11WWPCnImL2ZJt6ceyybiEFxfXh3XcSOi4mLmRanj\nUATKcYzfr12hvDc1Ojxgfx3nNeHenlK3a6ts79kweY786uW83M2bP5EOgb4/GOBzmjgMNRZqZbOi\n+1ocQDDR4qj/HdXs4Xvg3cXrVcrE+j0/RqzXXM+5ffB2qoz8eILPuXPnAWnf7SFp36W1XFmyWLY5\nQgSInv26jtfVo0SIO9LS0qVskbpy7tw5yWWCvr766qsyevRolQKBwn1YQ9xyyy1qTVGvXj2/9kXt\n948j//wjt1erI7a4ONW2UdP+nLlzyaapc33qPvSnx7NPya1VaimdRMG+QHUfN+poV+io+6hzdGTT\nn8p44Kj9enQ/3Npvhm6HSvu1Ggu+Cjga0ZhY0H6956Vn/rwVyDz+74lsWm30M6H2E7O139AV6fff\nf1dGhfXr18vEiRNl7dq16iAowoQWUS+//LKR3cYEVrA4u47n7X4DZeaUz+X6no84tW4CKG7kWtDJ\nXauqSaPGSOaFDKcWgxlp6YbbU0Ews1DZ9rNRsuL85/91hcghHZ/o5TbHFHOJxYW3uXUd9+4fV8i6\nqd96HOPG1WuzhWbivC46nKc/5+Vviy+z3x8MtM9iyKixyoqe2w+Pt5GQwEhBz7l58xg45mfCAJHN\nAzJiiqr47G7BYMQT4epdIiQUHD58WIYNGyabN2+WTZs2ydtvv606+Tz66KOqgKO/hgWrYkXdx3g+\ne/8jgWfJXbtmpB340n3oz6kTJ51aDAaq+2CyQyQFnAmoc+QtIgHng/aUVtR+M3Q7VNqvFXDcf+aY\nVwNDoN7uaNV+veflaf4QlYB0Ro+67yWiw+hnQu0nZmPIsLBjxw5p0KBBtt7TiFQ4ePCgWWOLKqzm\nbXYdD1IMzp86I0UqXO1UMAndEhzbSXkrsLRy1HQ5sGqDvcUgvB9G21NhTvq81l+eeOl5ObB3n6pp\nUKpMacMeAXfjLlSulGycucB9Eaj4eHmyY1fJmZjo9DrOC3PlqXCUp8JXRlp8mfn+YOGYd7vnwAop\nUrSQXFtUX8SSvyGBkYSec3Os0qxFCyQm5nTxUiTLhQuZyguht8iju/1GoveHRD87d+6UKlWqyJVX\nXqkMC45riX///VciHavrPjSreI2KUr11U9UhybFIMsaqR/e/mfSlvb0wIhYC1X3H2gjQU9RU8BSR\nYHXtN0O3Q639WvtJzcDgzrgQqMZEq/brPS/X+StYuIB8MnS01Cp/B3WfRDwJRntPI0rBNYsCkQxo\nG0Ws7212Nx4tSqF+z0fsVni0ZOzyeC+7cHkrsATPwpoJs9UCBY+LVCqrjA2BtKfCdtdWrhTw3Lob\nd0JykvLMuI4R83BNowZq8bHisy9U3qXj6yj0uHKU83NamgYWQcjPNKPFl5nvDyYqfDJBpNTVJXT1\nwI6FAkL+nJujxwDhjK5eCrTp3Hr4hN+eHXoiiNXBWmLbtm0qtNJ1LVGhwqUbnEgmEnQf7RxXjZmh\ndN+xSPL97R7RpfvQn4c7t5dZYz6/1MLx1npqn4HqvqZn7vQ0UrTfDN0Oh/brbTtpVGOiVfv9PS9t\n/qj7RGLdsHDTTTdJUlKStG7dWvWdRpRCnz595JdfflGto4i1vc2+2jrV6djcuXWTQ/ihtwJLufLk\nllZtH5YZE6f912IwRarXqqnyD81ulenv3Hoad4EyV0p8XLxsmPSNGiM8EjUeaqrqS8BAknXxoqwZ\nN1s2TLp0Drly55YcOeKlaLVr1VzhORSWKnxNGTm996BHj4y3eXP15LgLm/Xn/aHCcfHhj0EhFAWE\nrNDCyt9z89aGavDCNTKs5U2SkpgQFZ4dQgAiFVD4uVmzZqrw87Fjx1Q6xIgRI2T16tURPUmRqPu4\n4XZs16zhS3+eeKmPIKfim0mz/usKkVPWjPtSpTEGQ/cjRfsD1X1/92GmAcGopodT+6n7hESoYQG5\n7nPnzpW+ffvKwoULlbcBeZFLly6VvHkvtfgj1vU2+xrPqvFfyuaZC5xaN/lqVQWPRNUa1VWthvRz\naSo0EgUdnx/0qmo5ZXarTH/n1tO414yfLZ0f7yn3tW4l9ze4XZoOfUlK1Lq8r5J1qsnKkdNl9rLv\nJTlXsjqH4e++r/ZT49H75NiOfbLv97VyZMtOFTqJ19yFuXqbN82T4y2sU8/7I8mgEKywfSu1sPL3\n3LzlZ6bNXSWvL1wjzWtcHRWeHUI0xo0bJ4MGDZIZM2ao9Id58+bJ/PnzpWzZshE9SZGm+/uXr5eT\nf/2drV2zHt1vfN0Nds16uHMHeWHQq+r6FyzdjxTtD1T3sa9gaT90PNjGg1BpP3WfEOtgeKV9xRVX\nMDpB71xZzNvsbTzIL9zx9WKvHgZ3rZ6wuNi4foOqmOxY0BH9rBGSGKwFlD9z661FldZvG4sEx8WF\ntp/CRYvImVOnnfYzbthnkmXL0t1O01eLLF9hnXpbbJlFKL0ZZoXth7qFlTcPieNres7NW35mcmJO\nGb58h7y7ZINHz44VvDWE+Evu3Lnl3XffVT/RRETpfo54WdT/Q6+aolf3Z42ZJvny5wuq7keS9geq\n+3r2EcmYof3haF2pV/t9aXGguu/v8QixbLtJhCwOHTpUfv75Z7lw4YLUrl1bXnrpJZUzGW6s2HLK\nXRslzeIcjlxLT+Np2a6N9H1jkC4ruNbqCUWLXFsz+dNiKtRz66lFlbv9/DFmplSrWUO2b9nq0mLr\nabm1Sk1D5+zu+HjupgrVdO1Pb4utYLecshqhbGHlzUMCjEZNeGsZ5cmzYyVvDYl+zG43Cb3+5JNP\nVPTjyZMnpVKlSvLCCy9IzZo1I177I0X323fvKm27ddalKVbR/UjT/kB139v4IzFiwSxC3boyGNpv\nRPd9jYXaTyKq3STE+/bbb1c7fuihhyQxMVH1nkZrKORFli4deW1igoFjzpyrxRmpAu3/C3cLB94s\n4O4uSO7y/7QCS2jz5Csk0VcryGCdi6/CUL72g4XFpg0bvbbY8nTOnrw17o7vT9isu/eb0c4sUg0K\n4Whh5c1DAox6T7zlneL/0t34jXpr6OUg4QZ+jZYtW6ruEJ06dZL8+fOrwo2o47Rs2TKpW7euRBre\ndB83py3bPqJuTmNB97VWkNT+wHXfce59fUaxRKhbVwZD+43ovq+xeDoedZ+4cup4mughPS09eBEL\n3333nbz44ouyZs0ayZkzp/15GBkQufDyyy9LOAm318JTztzjL70gH73xtkwbN0HVIQh36yk9FnC9\n7Zw8Wd1RuPGRjo/K5+MnBb3dllnWfD0eGRR8AmZ5a/z1XLh+PlNHjrFbrNv+V3BT7/xGukEh1J4L\nX8fJyrLJgDurBzQGHENP3qmRc6aXg1glYmH79u3KiPDnn39KgQIF7M+/8sor8s8///idbhlO7fem\nlbi+Dx30hnwzfaYl2k6GQvd/3LRWRn7wUUhabUaq9hvVfTO0nxEL1tN+vbqvZyyux6PuE28GhYJJ\nV+vS16uuuio4EQvYKcIUHY0KAN6F06cv5aL5YsWKFSr0UePJJ59U3gpH0G0CRZy08Mi7775bFYm0\nOp5y5lb8/Ktb7zcIJCzSiMXa8T3e8iD15P/hmBUqV8reumn0DClSpIhMGj0uJO22PEUi+Dtfej0y\nKFKFfFLtnPevXK/ac6KThr+LG6MFmvD5TBqe6myxHpaqXvMUCqqde2bCgagwKIS6hZUvDwkI1Hui\nN+/UiLcmHPmohHhaS5QvX97JqKCtJVDMMZLwppVg1tTPLaf93nQlEN1HNygYFULVajOc2l+0yjWy\n77c1smnmAqfW3HrHbbQwYyDar+13/5ljEb8GCGXrymBrvz/1JvzVfuo+cYceg4K/GDIsNGjQQHkU\n9u7dK2XKXLqYw6CAhcAHH3ygax8XL16U9PR0VQU6NTVVHn30USfDwk8//SRNmjSRpk2bqmMgB/Oa\na65RFaPRlcKqeGqBlHkhQ1aP/1Lq93jYtNZTerwKgbxHbzsnbLdt8xYpVr2CUxsmPD60bqvU7dLK\nEu22/J0vX8WhUPkaRaqwP5wzCmChnzWiM9Buy1/PjL8pHZh3eCvctSccMmqs0/y6nnty7lzSrVdb\n6RtgW0erEaz2lXqLLeF48Fp4es3s9pC+xuJ6PG8tLTFnyOdk8ScSKqpUqaIcB6tWrbKnPaBm04QJ\nE6RVq1YR80F400o8D8xsOxlM7TdD97dt3qpes0qrzWBof568eWTCiFFqLWvLylKdIXDtx7GCqfuI\nNEg/ly5TRo42pP2u5479RbqBIRS6H8naT90noUTX1e/777+X0aNHOz0XHx8vlStXVkYG1FhAXiQi\nGLZt2yY333yzz32iZzV+tm7dqgwLrnz88cfy4IMPytSpU9Xj5557TtVu+OOPP6R+/fpiVTzlzBW6\nupS66TSz9ZQer0Ig79Gb/4ftkNpxfc82UqhcKTl37KSkFMovx3YdkK96DZCCZUqads6B4O98+fIm\n5M13qfJ1ZkamTMJCrGtgnhmIPLbHokBPWCfm3ZvF2nF+3Z176icT1WuvRpGX2uz2lUY8JCAU3hM9\nY3E9XqjzUQlxZOPGjfLGG284PYcW1TfccINaS8C5gDpNJ06csLTOu+JLK0GkaL9Zum/2OVtN+7EG\nlvg4qd/tYfs+J6SOlPj4uKDpvpa2mPNMppw7l25Y+x3PPW9iBfu+YWCIRONCKHQ/krWfuk+M1lUI\nmmEBYo9oAUdcH6O2gtaG0qwWVFlZWZcHmpCg0iDwvCsZGRnKKuuYBxIuPFm6j+05oDzaeloj6QlV\n9OVV6PxEb9UiyXEfej0Rvs7Fdcyu2+UrUdS+Hc75+N6/pKzU83rOZuBt3rxFkkwaOUbNV6HChf32\nJmC/n4+fqIwKZnlm9IZ14jw9W6xT7PPr7XMf+9k0ec5CXmrHwkLAaAsls9pXBuIh8eU9MVpEyfV9\n/nhr/I1wIMRMkJfpbi2B6ESNGjVqqN8lSzobpN1hFe33ppV4HoRC+yePGif3tW4lpa++Kls3Ab3a\nb4bu58qdW2y2rJC12gy19vu7ljJL92EEgAEgI0+CpKQkB6T9ruPU9h0OXDXNqDYGW/fDqf3u3qNX\n+6n7xGhdhaAZFtDtAT+h5PXXX1fpEe3bt1fFItB14q233pKqVatm23bIkCHy2muviRXwZOleN+lr\nqVG7lmph5CmXzp9wPV9ehYbVr5PzaelO+zBSgVhP/p+37XDOqDsQn5DDr/xBf9Azb67nnpV5UVaN\nmSEbZy2UzPTzckeNutLeTeEjX94Ef+fUTDAOFGtCXqWrxbqDQ66nrzFawUvtWlgoKTFBYFfMyMy0\nZAslXx4Sb68ZLaLk7X16vTWhzEclxBXUU3CNWAgEq2i/L60E3nTUTO2/v8Ht2d7vb9ehQHVfzzlH\nsvYf3H8gbLoPA0DeRJF2PboFrP2O4wyHUcGdplWqdq1s2bBdzqWlU/t1rBf0aD91n5wKgUFBwxqr\ndDfs2rVLjhw5orwW8EIgj23Tpk1y/vx5SUpKctq2X79+8tJLL9kfY/vCbizQocKTpRtdIYa9M9Sj\n99ufcD1vXgVECVR5uKkUKV9GRQto+1DWaR2eCG/n4qldltFzDrQoFbZ7u99AmTn1c6nbtZXHeXOd\nLywsNsyY7/U9erwJer07wUKbR+RVXrJYp6iFheP8eh9jiiEvtdkti9wVFhow7w9pX/caqVq8oGWL\nC3rzkHh6zWgRJV/v0+utCVU+KtEPW4AZw0rarydXPtjanyMxp9z2Ug85deiw0/v91alAdV/POQeq\n+9hu2phxPotDB0P7w6375mj/5XH60xnKzGuVO03r/90f0qBsUXnz3nqWLiwcKu3X8x492k/dtybn\nTF5LeyPYBoWA2k2aCWosoFbD7t275eqrL590uXLlpFevXqqtJUChR4RMDhw4ULp3727pdpO+WiC5\ne95I26H3XntDCWKdLi2dKjLnKlxALpw6Kxlp6ZIzV7IUqVRWTmzfK79u3yjD330/23s0L4K3vMDT\np07papflzzm7Q6/nxnU7LKiqt75b6nZtraIj3M2bNl812z8g66bMkes6tzClVaS7z0HPnJqJr/n1\nNMZ2PR+Sd97W3x42GC2LvLVNevP7dfLXoDYy7OfNpraMjLSWmMFopelPaysSHCKxBZiZ7SbNxgra\n7+1aHGzt1zzyrrqP9xvRqUB139drgeo+nCjFa1SUe957Wem+p3kLhvZbQfcD0X6Ms8fLD+k2KJh9\nrdKj+ymJCaa3iw4XRjScuh+++gL5CuaKKu0/dTwtYMNCUNtNhgJUhnbs/oC/USwHz0cKnrzc7p43\nElLv6i1Izp2iFhVpx05KvW6t7db4laNnSFZGptqHvxWINUZ+8LGudln+nLM79Hpu3G2HhRXO/4bH\n27mdN+0ckVeJEEizwhiNzqmZ+Jpfd2OEUaFX365+HScYLYt8FRb6+/S5qCkuaLSIUjCKL4UiH5V4\nhy3Aog9v1+JgaX8cCgkibbV7aylRu2o23cf7jehUoLrv6zUzdB/nueKzz5Xue5q3YGi/FXQ/EO3v\n0qeFX4Uazb5W6dH9coXzxbT2U/fDlw5w/PieoBoY3glh2+9gFmq0lGEBrSonTZokR48eVY8//fRT\n1cu6c+fOKv2hT58+8vLLL8uWLVukWLFiqsYCDAutW7eWaMRIaJ1rDmCOhBxyZ636yqjgWqgHvaZz\n583jd+cBEIxCRYEcx9t2OE8RmyTlzyu5XOZNO3cUa0JepVlhjEbmNFQ45k3CO9Hh6fvl6OFjUqRo\nIUlOSfZrYRGslkWXCgu5L0SVNymnFM+bIrPW7YmK4oJGiyix+FL0wRZgxAzt/3Pbdml9V1O5vucj\nHnXfiE5Fou7X79nmUvHIEGi/lXXf0zj3HFihtP/aoiXCfq3ypmma7muPY1X7qfvhqy+gbRcMA8O5\nELX9DmVdhYANCzt27JDly5e7fa1ChQq62kSh4wPSG/LkyaPyJAEea5kZTz/9tNxxxx2yZMkSOX36\ntDzzzDPSvHlzy4VemoXegkneLNb7du/x2NISz589fcZe/VhvBEEoCxT60+bK03Y4zw2zFqrfNa+r\n7XbeMAco1mR2YSl/5jQUuM2bzCN+LyqC3bIIF9CWbe6T/uNmORWiGjh/tTxcu5xKgxi0cI30eKpT\nRIdCBlJEicWXog+2ABP5+++/VTtrd1x55ZVy5513SrQTqPbnL1BAJMumS/f90alI1P0j2/fIPxu2\nS/VaNUOm/VbTfY9rgQSRG6tVM/T+YHnOL2nheCctHDBvtdxUtqhs/Pv4f9oYu9pP3RfTbq6N3lgH\nw8BwOARtvwM975AbFtatWydDhw61P0ZhRRRbTExMlBdeeEGXYaFs2bI+q0NXr15d/cQKgYbWeW15\nFUBBoUALFektyITXk1Nyye4fV6i+2AnJSbraXDmOJ2dKstRoc6+smfCVbNu8RR3b3TH9nWu952A2\ngR7XzJ7UwbSev/rGczJtwmwZvHCNpM1dJcn/5ctOWvWnJMTHSbcnOkRNcUGjRZRYfCm6oDfqUuSi\n41oCC+49e/aoNpIdOnSICcNCoNofDbqP4xxYtUGS8+eRlMIFlPYb0f3V47+UYtWule1btsa89vtb\nlDEc1ypoWkZGpvT/dKJkZtlUpEKd0oXlj/1H5aaP5lD7qfumYMbNtauBwR16jQ5FQ9T2OxxGBcOG\nhVatWqkfR5DScM8990RtqkIoCDS0LhDPRzD2608LLWyLwpIXMy/KuqnfysaZC6RcwxukQJkrZc34\n2braXP0xdpYq4Fi6Xg3V1jM983J+qdG59ucczCRcx/VGMK3n+fLllV5Pd5TPPhovfRpWl6aVS8lP\nu/6WId+vly6PPSqD3uoj0YLeFlFmvY9YE3qjRK6//npZu3at07ygMFSLFi3k4YcfllghEO2PZN0H\nORMTpULlSkq/UScJRoUrKpeTwxt3SBeH1ol6db96q6ayqP+HMa39ZhkUgn2twvlo2j52+GR5oWE1\naVSxlCzadoDa7zBH1H3r4Olm/fj5PfYogXw+DAzRrv2mXR2LFCmiFgOzZs1StRGIcfSE1sEbsH/P\nPpVXiG01YQxWQSEj+/WnhZZ9264ORZlGTZf4uDhp2a6N2zZXmRmZMj51pAqDhMdC6wqxfvp3qktE\nYs6cPr0qvuZaG1et9g9IwTIlndp3BrPysz9zF0qC6TXX9jEidaq8u2SD2nfPpztFTaSCWcUTWXQx\nemAUSnbQ1QGdnyZPniwNGzaUWMKXHmlRAHny5ZUzp07bowEiVfe17Tdt2Cj1ez7spP0VK1fyW/fx\n+NieA7oiKqJV+2FUMDNSMRTXqldff1Zy5kyQd1OnysD5a6j9bqDuW7tgoT2i4bw+A0Ow/p9Cfd5B\nbzf57LPPSnJysrz11lsS6y2ngmnN/uD1N2XCiJFy8WKW2LKylMW/Q6/u8mz/l+0Wbb3tHf1F7379\naaHlbVssMLCA8GS1f+fVgTLhs9FSu2Nz5bFQbbfGzFTv6fbUYwEtADCuG6+tKgUrXi1Ht+52274z\nGGkRRtqPhXKBEexWhWyDSGKNSPrOh6Ld5LBhw2Tp0qUyY8YMP8cWndqvebHR1SD9XJpqs+hOFyNJ\n931tv3LkdElKTlI1Edzp/sSRY6RWhweddB9pEEc2/Rlwy8dI1v5g6j6g9pNwYOSGOVypAJqBQcOb\ngcHX/5OVzjuo7SZ//fVXmT4dVXgvs3//fvn222/lxx9/NLJLohMsLsYOS1UdMhwt/ONHfCbx8XF2\nMQ1WQaFgFH7yVZSp0evPyKmD/7i12j8/qL/E50iQiZ+NUmGQaLuF1qQdn+gVsLcG48Ii7vCmP6Vu\n11Zu23cGY47NLJq1/8yxoCwygmk9p2WexBqx+p3fvn27DB8+3Om5f//9V0U+jh07NmzjshqaF/uK\nKuWz6ZGjLkaS7vvaHk6Tis3v8qj7CTlzyuTR4y7p/n+GFtz0mxGpEana79gFKlhQ+0koCVdng1BF\nMKR40P5IPW/DhgUUa0QHB424uDipXLmy6u5Qp04dM8dHXKzZ8FjE58ihxM615dLkUea1gAoUfwo/\n+SrKVPr6Girv0l2bK3gx+rzWX5546Xk5sBepIXFSqkxpU+YA4aZYsLiba8c2XmYTaNEsjbyJFdRC\nA8YFEEwvBiGE+IvWHcoRtJueM2dOzBRu9IXWZhEh+eumzHGrR2a2fwyldvnS/us6NpekvLnd6r5j\nnQRoMbpfmBWpEQzt13PTn6tguiTnzuWhEGeKJBc46XM/1HkSiXjzzEfajbUvA4NeIvW8DRkWbrnl\nFvVDQotmRQf+WLRD2dHA8Vh6Cz95KsqE0MYaDzW1d4fwdo7Yx7WVK5l6Lshh1du+0yiecmbNKsYF\n4wIIh4EBIV5oq4MKuFYP7w42nAtCslOpUiVJTU3l1OjwYiPPHyH5VtR+I7qvpyAjtN+X7mvPBarF\nwdJ+d4UUNT3Imy+vnD51+rJG5hHp1qutpH4yMdv89Xqyg+FW0aGEWsf5iJS2iKGkYJSfn9+GhW3b\ntslvv/2ma4cVK1aUG2+8MdBxES/tGDMyM91btHM7ewRCWdXY3bHadu0sHXt1l2ljJ/gs/ORaJAre\nguI1KqqiTEY99oESrDZejvM15bPRci4tXXLEx8nFLJtqQdO2R1d5/KUXTC3G5WhgCDY4t3cGfyJj\nR0xVvXpz/1cBF0VpwtXRIlxwLgi5zKFDh2TBggW6pqREiRLSuHHjmJ8+TYdQPBB5/nqiAUKl/YHq\nPtCeR8QltkfhZa0go6fziwTtd2dQ0PRgzPAp2XRf08i+/xVvG/vZtP/mL0UZFbTnrQq1jvPhD7Fi\nUIhFdCnMpk2b5NNPP822QDh48KBcffXVkpiYKDt37pS8efPKiy++SMNCkIB1HoWMRn88XOX6OVn4\nx8x0as0U6s4C7o41IXWkWlCg4JCvwk+uoY1TR4+VSaPGqkrPZrXP8pdgtfECqhDX8FS5oXQR+X3v\nERl0dx25tXxx1dd24LBLHrxAWo+GEyycRn0yQQY1rn35nD6eoF5D26RYgnNBiHMtJte1xLFjx2T3\n7t3KkJA/f37ZtWuXqiHUuXNnGhZcdOiKqtdk0353ehQq7Q9U9121/+1XB8rMydMkMW9uObp9T1h0\nP1Dt99bqUdMDt7rvoJGvDn5WnouwtsLUuuiYj1B1FaBBIXox1BXixIkTctNNN8no0aPVb4DFwH33\n3aeKOlatWlXCSbRWhnbuCjFK1bpQXSGSEqVDz+xdIczoLKCHQI/lGEYJ8HfBwoVl5AcfKY/9uTNn\nlceiXbfOQesh7Slk1B5ZYOI4cKybr60qL91WWf63eL282qi2vNCwuv31dxevlyFLN8tPJlaedoxU\nCHbF6OplbpX+d1Rzc06bZP2eHyNikWQGnAsSjZjZFQI1FrCGeP7556V160se6sOHD0vz5s1lwIAB\n0qRJEz/HFp3a7xoZYO8KkSe3POqiR6HSfjOO46r9h/46KDMmTJIvJkwOie6brf3eOjJoeuBd9yNT\nI6l1kT8fkVwskISGoHaF+Omnn6RGjRp2owIoV66cPProozJ79uywGxaimUvFCgfIEy+9IAf27hcR\nm5Qqc1U2ATezs4AvjB7LdbGEtpk2W5bqU62Fbi7btEZOHDseNI+9r5BR10gKM8aB+UKKQKWiBeTM\n+UxlzXbktmuuVH1tjXxG3lIdQlFbAXmjODdP5wQPTKxUv/c1F7u2H5CrypQK2/ishLd2TKHyoHgb\nAwkOSLPENVgzKoCiRYvKM888I9OmTfPbsBCt+FOsMFTaH8hxfGn/I506yMOd2kuxElcGLVIh1Nqv\n6YEv3Y9EjaTuh3Y+gqWJNCgQM0gwekFGSKMr+/btk2LFipkxLuKDS8UKK3p8HcUAExJzuq+2nJQo\naefSlKXeDNE22sXAXRglwjyvbXKLFCpXSoUhZmZkSJuunSVY3gm9IaNmtvHCsZFTufXwCcmTlKBC\n5OpfdXmOlv15SPLkTvErp9Rb+GUoQREqnJv7c8qlwjpjBW9zkTd3ilxbqo6kJFkjvQUW6H/++Udd\nv+HxDSWe2jGFOgfz+PFLfadpYAgdWEv8/fffcuHCBZVS6biWwGtE/CpWCH1LT0uXhMREt3qM583q\nZhRI9yJf2j9x5Gg5d/as9B3yWtCiEkKt/ZoeeNf9yNRI6n5o5iMYUQWO2k9I2AwLDRs2lKeeekra\ntWsnbdu2VQuCRYsWyaRJk2T58uWmDIwEZoVHW8rMCxnZ8jHRaSErM1Pub3C7aQWdILwo2DR+hEtl\n5zEzpVNv9/mIWgstiLprO6c1E7+SBs90kH2/r5XxqaNk3LDPAh6rO+8EvCLTxk1wO4Zgtu9S89Wj\nqwwZlio3XFVEBs7/Qx0T1mwIz6CFa6WDS70MPYTbqABS/itChVxC13Pq8VRHy4X/mY2rJ6Fd54dk\n4Mgp2eaiZ+/HQn4D7+n/4q233pLRo8fIuXNnJSUlt3Tr1lVefvlyWlW42jGF2ntiH8d/BgbiHty4\nmkWVKlWkePHicu+990rv3r1VjYXff/9dfSdnzpzJj8CgvkmcuNf+jEy5o0ZdVaspHLrvS/tXT5gt\nuQrkRQ9z+WL8JPlmxizTdR/76/HsUx7HECzt17RxyMcTPOp+pGpkrOu+mfPhKxrBLF20gvaT6MTQ\nt6dAgQKyePFieeWVV6Rbt26SkZEhtWrVUsYFpkGEHkdr/PB331dW+KotG8u6qd9K+TtuVDfqCE1E\nX2g83rHgJ2n0+jNy6uA/ARd00o6dmXFB5cuiTdSKC5cqO2ddzFKpGu68Bb7CKJenfi6HN/0p9Xs8\nbErxKXfeiUmjLhlfHMeQmX5e8pcsrnIqzUwX8VQJe+rIMXI+M0v6z/tDMrNsKlIBRgWj3R+s0Gbq\nyRe6qcdDUqeqsD9Y6CGmqHgdrXjyrr/26tuSnCO/vDlmtJoLRCrAqADxtgJYWKSmfiaNH+wp5SvV\nkZ1bV8uIEZeKh/bv3z+kYwm1IcFTlEa0h4MGGp2SlmVeGG6OHDlk7ty58uqrr6r0h5MnT0rlypVl\n6tSpTIPQ81n8p63TxoyTSaPHKX3LV6KYLOr/oUftr9j8LlMKOeLYp06cUDqvV/f1aP+m2YukbtdW\nQdN97O/UiZNOY4Dun/v3hBStck1QtV/TwLEjprjofmRrJLT/4UcflIyMTBkyZnrM6L43tPPWuw4K\ntVHdStofSxGasXDOhoo3Wp1oLeCkxxqfceGC1OnSUqq1bCKTHnhM6nRqLlWbN5Jzx05KSqH8svHL\nhWqx0f6r4apHNAotbZg8R371s6CTu2JSaA/ZZMhzkn7qjDrWhpnzZc242ZIzMWe2HEaM01PhJ3gt\n4iROjd2M4lPeikytHDVdruvcUmq1vU9WjZkhm75cpHqF43w69eohzw96NajWW4zNV86sFYoy+ttm\nqlK1a2Xz+m2Sln5eUnIlS9fH2kVlu0m94fq4qKMoHfLHrSJkGFPlylXkjnu7SsNm7e3PL/5uoiyd\nO042b95kmbGaSax6asw6b70FnMJBrGq/pr/3vPeyikj0pf3424juuzs2jAlVm98lle67U/IULeRV\n9/E986THf0yYLavHfSn1ez4cdN3fMGmOqulQo/39cuHMObvu41zibCK//blJ8ubLJ/7irXij6404\n8uzz5M0jZ06fiZjOD760PykxQS5etEnmxYtRrfv+oH3Wnj7jcETpUftjR/szTVzvBLV4I7EGbvMU\nR02X43sOKqNB1RaNVPijFqa4a+ly9bjGQ03V64EUdPKUI7l2yjdSv+cjapsTew9JFsS7wwNuvQ8V\nKldS43UN1yx9Qy3Z+9Mq04pPefOQoLL2mgmz5a8/NqoICUdPCSIaEnJeKuAUrpxZKxsUvLVV6v/d\nH9KgbFF58956EdNmyR/8zf/HzQ5a81oJWK8hNPBWOHJNpetk7vRPlCHEamM2g1j11MTqeUcjnvQX\nxnHoL7TfMRUC2o+oguqt71baH0ghR3fHhm7HJySoY/vSfWieO+2HQwFdrkKh+9jfw53by/RRUyQ+\nR7zU7db68rmMniEjP/g4qLqPG0yteF/hIgUlUnGn/QPm/SHt614jVYsXjFrd95eCBYpI5nmRU+fd\nvz/UUXLU/tjRwLfCoPs0LEQo3vIUsYC45fnOUrfrpUrbWphiXHy8FK9Z0f48gKDD2+GuoJOnUEZf\n9RHqdGyuntu15Hep1/0htzmMnZ/oLds2b5Fi1SvYwzXhLcCNPowKGJORolD+FplCu65WbR+WyWPG\nq7SLUNZaMIpVCjVq1nh4K7Cw0NoqoVgR5u7N79dJteIF7Y8REvhs314R6ZkJV0HBYHIpJC63Epqr\nyl3u5PPn1j8kd+48Kroi2oClHZZ7iKwWpYFzt4lNxowZq9oeRmOURqyedzSiR3+h8f9s/vPSzfvF\nLKWtMCpo2m9E9/Ucu8bDzbzqPrQUeNJ+rFFCofvY3xMv9ZEvp3whdbqGtsZStOBL+4e1vCnqdB9Q\n+yOTWNTAc2E653jT90hCgjdr/MULGbJq/JdydPseScybW8Rmk3taPqi8Af9s3CHrp38nhzfvVCGB\nKqLhYpYKw3cMnXnvtTdUGGHjOjeo33isVen2duyMc+myf/l6WTVulhqHu22Qw7hj81ZJP5cm1/ds\no0IzH/n8A+n47Wdyz4evqO1atHlEFYHCGLWxrh47S/WR9lfssT1CMd3tDz3A2/fqoebA01jhKbEa\nVjAq+GqrdPp8hvx9+pz98Zmzl0ICzRB5I54D7X2B/GBREQ0LCwBBQUjcgtmpKv1h365N6vfC2Z9J\n165dok5kfXlqzp49o6I0opFYPe9oRI/+QuOPbPpTWrR5WL2G+kdYC2BNYFT39Rx7+ajpXnUfWop9\nuNP+ez/qp9Yoq0Kg+9gfjCRIyYwk3bcSerTfTN0PB45rDWp/ZBOLGvhPmM7ZUMQCCjfu3btXOnc2\nrw0gEdOs8WgnuePrxbJ+6rfKMt/l8V6qCvIP8xZIwYpXOxV0KlbtWjmxfa+TN8BbGyZY8dPT0iQ5\nJZfbY8MLgsJRGBt6U3vyFlxbpZLT+POVuOQdPbz5T/V63yGDpFCRQspzgLHiuc6P9TRc1FB7n7v9\nYXHhbi73r1yvztOs9lwx11IxKacUz5tiWistV6+B3vaAVvA2WLVYkFZEEtZrpD8gUqF3716WKS5p\nNrEYpWHl8960aZPMmTNH+vaNvGK1VtR+TX81fYPuz/3yK1N0HwYBtLH2eOz4eNn7w+9edV87nift\n17o1fTFhUth0X0Uy5k5RhgezWnLHqvbPWrcnIltomrlmoPZbA6tqYDSesyHDwqlTp2TJkiU0LIQR\nzRoP4XfKUxw7SwknFgKwuDsWA0SLKWxfs/0DUujqUnJszwFZN+lrtb22jbdQx/EjRqo2lvA2YPGw\n4rMvJOviRSlZp5r92Ch42LZbZ3Vc1aHCw/hQT8Db+FE4CTmO7s7DCChS4ml/eM1xLMWrV5SVo6fL\n3+u2KQ/KnTXrmdKWM1Ac6ypYva3SgHmr5aayRWXj38cDbjvlSeRd2wO6GhisYFCweqFAjAF5dgiJ\ns1pxyWBGaSDHEOGAsNxDZBGlAYNKtJ67Vc8b14xvvvmGhgWTtN9Rf4Oh+7jhRn0E1CFwPfZDHR+V\nvm8M8qr72vG8aT90+qlXXgy57mMcf63eKKtGz1Sv3d+goWktuaMRT9o/cP5qebh2ORn28+aIbjcZ\n6JqB2m8trKqB0XjOhrpCwPPWsGFDWbZsmVxxhX95b6Eg1ipDwxqP0D1Y4xHi50kE9Wy/b/ceFQb5\nYOpgJys+wgi/6jVAara7T8reUs9e5AjdGzIzMtzuy9fx/B2/O7zlgwYyl/EJOVSeqOa5QTiltugJ\nBt7Ow0qFGr1Vhh6XOlWFPcJDUbHqtbJ143Y5ey5dPe7cq63X6tBm9G4+fn6PofcFk9dffz1b4Ryk\nHuCiHq3FgqyOtuBDlAbCAWG5R+qHVYw9Vj9vM7tCoFV17dq1ZcaMGarNZKBQ+7Nrp9m6Dy2sWqO6\n7Ni6zbCuB6r9wdL9hJw5BQviut1aGdJ+vV0hXFs0w/sfiTffrtqfnJhTLtpsqu2kHt0PBa5rC18R\njtp7Al07UPutRyxqf6aJ56xX+w0ZFr7//nt58skn5e+//5YGDRpInjyXQ8UbNWokXbt2lXASK4sL\n15aFeq373rb31qIJRSCRC6l1lNDaNk1f9J2UKlPa47F9jc/f8XtqtWmGZ+HYv//KHTXqSs2OD5rS\n8krveUwdOcbeqrFtj65O5+HvYsUqbZV8tVmySlRBsIjVlk6RghVbgEbCeZtpWNi4caOKfMTvW2+9\nVQoWvFwhv3r16tKvXz+/x0btD77uQwu/X7vCa4tkPbrur/YHS/cxjgN798tDjZq6nQO92q9Xq921\naIb3P9w34UZx1HrgS/dDhTsDgeaA8GRgMKNAM7Xf2sSi9p8z4ZyD2m4yf/780rJlS7evWTGCIdpx\nbFmox5LvuL2719yFKSI6ofydN9kXF45tm5JzJXsVXG/H0/O6O7zlgwYSVXDm1GkV8ol9Zqafl3P/\nnpCUwgUCas/l6zwmDU91atc0cNilVjDBbHcVKO48LY4ttNw9DsSgYNU8RW/EakunSMGKLUBj7byx\nOGnSpIn6caVkyZJhGVO0ar/Zug+jQqC67q/2B0v3MY6k5CR7Ycpga7+7No2R0ppRj/Z70n2zjhcI\nrimU3rYxCrXf2lhJA6PxnA0ZFurVq6d+iHUw05LvWvAoV+7cEh8XJwWvLuG0ndE2UIHiLR800DZR\nWJShYOPyz6bJsa275HzaeUnKlSSFKpVTc2rmueI8EKngrl3TkFFjLdnuKlBPi78GBS2Ma+zoUXLm\nXJrkwfG6dY+I0LVYLBZEiD+UL19e3njjDU6aBbQ/lnXfDO3XWwPJW5tGK7dmDHWURbCPF8wISWo/\niWUM/Xfu2LFDli9f7va1ChUqSP369QMdF/ETMy357goeaQWZJC7OY0GmUOGt5VWgngWcS8UqlWXz\nmrXyetPr7N6E/t/9IVXq1DL1XHEento1vTx3lenREeH0tBhNeYBRYWTqCBnUuNbl440Yrl6zeo2C\nWCwWRIg/IJ0SqZXuuPLKK+XOO+/khIZI+2NZ90FmwgG5pnJZ2bZ2azbtr1i7knr99AXv+9CTBuGt\nTSN0H2kEZnr8zSLUURaRHNVB7SexjCHDwrp162To0KH2xxcvXpRdu3ZJYmKivPDCCzQshJhgWvK1\nEhze2jaFqtCSnnZbgXpSMNY/N29RC4ts3oSlm01tP4Xz8NSuKY/J0RFmYMTTEkgNBaQ/IFIBRgXX\n4705ZrTqZGD1m/Ngt3OMxBQRQjTQttpxLYH/7T179qiijh06dKBhIUzaH0u6r0UapJ9Ll91bdnvQ\n/k1SOD6XKZEE3to0WrU1Y6ijLC4db0rERXWESvup+8TKxBt5U6tWrWTt2rX2nw0bNqgFAqIVWrdu\nbf4oiWFLPiodw5Lvbwjae6+9oYo5oVI0fsMrgsUEChgtWrNc/YZ3w1tImrv94DGeDwQtHxTVmlFY\nCZWr8RueFFSWDmQR4y2K4MzZS4WmzALjRKHGgQvWyLuL18uKfUfUb7RoatO9i+XSILx5WlARGp4W\nd8CgYCTs8ODBgyr9wd3xTp+9VIjG6mjtHFGoceXKleo3Hgcayon/IVSdRnFIRIjhNx4H+r9FSCi5\n/vrrndYScFocOnRIFXJ8+OGH+WGEUPtjWfcRaZDzTKYhfTPcptGN7qOLghVvmI1qvxHwPRnU93+q\no1QojhdJ2k/dJ5GAaYlRRYoUkRYtWsisWbNM88aR8FjyfYVW6g03DFahpUA9KVaKItDGi5oKCIPE\nMTo83ivg8wgGjp6WasULyqFT5+TKfClB87RMnDhRcsTHuf0s8uZOiagaBWYXzkGKiGsbS6RcREKK\nCCG+/le6d+8ukydPVm2tSWi0P5Z1P9SRBKgTAOB9v6T7uaTHUx3tz8ey9iMF4ouJX0pyQo6IiuoI\nhfZT90kkkGB2vmRycrKZuyQ68FTR2UgupFmhlcEutOQuH9QMD789imBYqhorLOQQM3gTcMNvdhRB\nsM4jGMCT0qnnI9Lvo/EyYP4auZCZKYkJCXIx66L0erqT1zQII6F+kyaMl1vKFpOB81c7fRYD5q+W\n7j16xmzoP+Zm9OgxyqigtbFEcUjUcUDYZSSkiBDiay1x6tQpTlKItD/Wdd8pkuDjCdm0Hzf9ZkYS\n4DxQJwAh/VZpzWim9geacvFakzpy7Nz5bNrvz2cRyPrDilD3SVQbFn799VeZPn2603P79++Xb7/9\nVn788UezxkbCYMk3q0BSsAstBdKq0opRBME4j0BxJ8zn0zMlLj5BGrfobfeUz/tyhHre3fZGKy+j\nbgDSIAY3u0O+3bRf3vx+nfoscicmSMbFLGnfoYOYTaTkLeptZRUp50Nil+3bt8vw4ZeKsWr8+++/\nKvJx7NixYRtXrGk/dT88kQTeWjKHC0835J60P1gpF3VKFlbPadqP6MV2nVv5/CyM1HWKBK2k7pOo\nNiygWGN6err9cVxcnFSuXFn69esndeo4L3ZJaDDLkm9WaGUwCywGGytFEehtYWUmnoQZ4jt1wpdy\nd4ve2Tzl0yaOk/593zBNlCHwaC356+7D8uY9deXVRrXk79PnZNrqnTL0p21SooRzC7RA0FpaIgoA\nN+xoEYluDlZtaemrlVWhQoVUvYVIOR8Su2RlZTmtJUDJkiVlzpw5LNwYQr2i7kdmJIGZeLsh96b9\n40aOlR6PdZFiJQqano6iaf/rC9fI8OU7ZNDbL3rUMCMGhUjSfuo+iRQM/efccsst6odYj0A932aF\nVpqZnhEuwhlF4GhQ0NPCygx8CbNei7kZwEDRpVt31VrSMRTyzR82SM/ej5nqVYi0vEWce+XKleS7\nWcOd2ljOmzVCateqKR999FFEnQ+JXSpVqiSpqZe+myR8ehXrur//zDEnrbViJEGw0HND7kv7jxw+\nKrlyXUqDzlcwl+npKB//vEVXCoS/EZKRpP3UfRIpxNm0vkIG+Oqrr+S3336T6tWryx133KE6Q9x4\n440SbtLS0tQ/4dqDuyU5V3Ram81u5eRqxUUBJoRWorI0IgxQdRmhlf5Ycc3aTyxhRYOCo9cC3Qfu\nuLer3WsBFn83UZbOHaeqHpt5w695E8aNGa26QKBgY+eu3Uz1JoT6nMwac6VKlaVEmcpyYM9WOZ9+\nVpKSc0upqyvJwb1bVARZJJ0PiSygr1dddZX6HuYySV+XLFkiP/zwg4q26datmyxfvlwaNWpkaGzR\nrP3U/fBFB4ZKj0OJpv2+bsj16uTx83vsrxk1MED3UcBxXOpU1QEC6SjoloEUCG+6j3Pxx7AQadpP\n3SeRov2GV+dt2rRRLVQQnnPy5Elp2bKlWggsWrTI1DBl4v5mHQWSUMMA6QbwEJh5s25WKoCVUgoi\nCasZFDQgsggThEXf0VO+cPZnqj+z2SKstWtCMUJEQ6ALhNnHCGUUhpljTks7J/c+9KQUL1leTp88\nKnnzF5G//9opHw3upLaJpPMhsc2rr74qo0ePlho1aqjUiGeffVZee+01KVCggNSrVy/cw7ME1P3g\nkzexglejg2tUQywYFPzVfm1/MDDgGEaMC0bSUYwUaow07afuk0jB0J3o77//rowK69evVy3h0H8a\n1oumTZvKhAkToqLdZDA9A4FgpJWT0XMxKxXAioUJoxV/Bdbf0EHtfxvdByC+yOnHwiKY//Nmt2r0\nJ2/RXUtLX4WevL1uRpEo1zEXLlrKPmY8D/w5H0LCBRbvw4YNk82bN8umTZvk7bffVhE3jz76qEqR\nCLVhIZp03+j5UPc9Gx0cDQzuCJfRwVH3vd3M+2tQMKr92L9j9IIR9KSjGKmrEKnaT90nUW1Y2LFj\nhzRo0CDbPwciFQ4ePCiRTCg8A0bxt5WTlc+FmEsgCwarRRGEEn+iMHwVevL2OjCrSJSvMYNQRZUQ\nEgg7d+6UKlWqyJVXXqkMC45rCXSHCBVW1kojLRytfD6RjNWiGlxvrLVIAVcDgxnrAytpfyAGhUjV\nfuo+iRQMKQy8h4hScC3PgEiGu+66S9c+pkyZIu+++6798XfffZctheLYsWMydOhQ1cIS1jpc1GrV\nqiVW9AyEAn9bQoXyXKzq6YmVaIRgGxRCGUUQajx5Yp5++mnZvXu33cPgq9CT6+vbNy6X4cNHSEZG\nhuTMmdPUIlF6vEehjCoJN5HQLoxkB9eQbdu2qc/PdS1RoYLnmziziSbdB9R+a0Q1GDEwBBJ16JqK\n4G6bUGm/MnQcDyxqwdf+A8WdjuKm/5FHHlHXJE1LrKL91P3sUPujpHgj2k2iSCOKOBQsWFBFKcDr\nMHXqVNm6davkzZvX5z6OHDkif/31l+zZs0eaN2+uFvCOFyvUbahfv75UrVpVevTooZ578803lZEh\nWAWccHN8U4VqUqPDA3bPAFg7dY5snDxHftm2Iaw3zf6ML1TnQs+IuWBR4roYCdYCgWQXKHhiUEAO\nnRUcPQwdO3aQ8eMnyJ33dXNb6OmPP1ZJnTrXqUJQtzZpI/O/TJVffpgh59PPSXx8DomPj5PGD/aS\nO+/taGqRKG3M7rxH3l6LFiKpXVi0YHbxxoceekh9T7GmWLBggXr8zjvvyOrVq6Vs2bJ+j81f7Y8m\n3TeyvVGo/eYWYg5V1CHJDq5lhw4dUqncEyZMdNISOBiqV6/hschjOLQ/1nUfUPujrHhjjhw5ZO7c\nudK3b19ZuHChOgjyIpcuXarLqACuuOIK9ZOcfKlFjStYLGJfs2bNUvsG6DxhNc9AKHCMBtDbyilU\n52JlT0+k4KkaNQ0KoUXzxLz++uvZPAyjR6dKRsYFj4WetmzZYi8EhYXFjws/l7ubX34/WkPu27VJ\nLpxPl8SkZPW7SNHScvbsGbdFohAGjrxzGGwLFy7sc8z+vhYtRFK7MOKecePGyaBBg2TGjBnqez9v\n3jyZP3++30YFo0ST7gNqf+Thb0cDYi7QSjhGEbngqiUnTpz0WuTRH+0Hp04elTLlq7nVfuq+fqj9\n1sWwSwdGgTFjxkiw+Pbbb1UkwxNPPKE8F+XKlZOXXnpJVY52BeFGsF45WlWMoML4c6eom2PkMGpA\nxNEqEV0NQok7j0Dbrp2lY6/uMm3sBLXowbiwuEDuZKjPxUjuJ8luUDh/KlOK5rpskKNBIXzASArv\nNxYXmncChZ2Qgzlv5nAV4uiu0FPlypWVhwOvw1uBhUX29w+TAU82lhKly8uhAzvlwvk05dEYO3as\nDBgwQHnY09PT5cEHH5Q1a9ZKVtZF9Xrt2rVUa19PRthYxdtnhQUicoGj2WMTLeTOnVulRTqmRurF\nDO2PJt0P1flQ+2O3RWWsacmsWWMlV64Uj0Ue9Wr/q4/BMWqTixczJSFnokqRQHQkoO6b93lR+yPE\nsIAcyN9++03XDitWrKhCGgMFaRIIR4bXqW3btjJ79my13w0bNigjgyNDhgxR7akCBTfB/ngGgo27\naIAJqSPVWBDO6K2FYyjOxaqenogzKORi+KNV8NaCCjf6i74ZLfEJCdkKPSGqAGGTyKvMzMzw8P4s\nue2uh2TZgilS9tqaql2kFg2BRQaudTAqrF27Tpq1etzJ44Hn4cUlkdsujIgKN0a6gx5Qc6lx48Ze\ntzFD+6NJ90N1PrGu/f6mOZDI1pKOHTvK1Knuizzq1X7cAP+4cKrUbXCPahM9/8sR9nsc6r65nxe1\nPwIMC6jW/Omnn2ZbIKC2AhZuiYmJqrozUhdefPFFUwwLSUlJcv3110ufPn3UY3ShQIjk559/Lq+8\n8orTtv369VPRDI5eC2/hw97QPADwuPvyDAQTPR4BX8Id7HOxmqfH6kSTQSFaC+b4akHVoUN7mThx\nnNuiiPgND+pnn410+/6k5NzS+IFukpInn3z/zVi1uHC0snfu3FlFKsCo4C5aAmGSRq9r0fj9MNIu\njISX/fv3Z1tLoEgzaizBkJA/f37ZtWuXxMfHq/8HX4YFs7Q/mnQfUPuDAw0K0an9vrQEEYW4Nnkq\niOyv9rfs0Je6H8TPi9ofAYaFFi1aqB+NEydOyE033aRyIvEbYDFw3333yf3332/KwBBeVKRIEafn\n8Pj06dPZtoW3Dz9mgHBk1AaAgPvyDAQTMzwCwT4Xq3l6rEo0GRSivWCOr5ZO8C6gtoy74kg4/8GD\nB6u/EYXg+P4FX36mCjuhvoJmVT998qgULlrK/hhRYYiK8BQtgVzOm2++WaxMKL8f/rQLI9YABZlX\nrVplfwxPHtYQKNbYunVr9Rz+t5AGqWctYZb2R5PuA2q/udCgEN3a70tL8uXL57XVphHtp+4H7/Oi\n9ocXQ1eDn376SdU60IwKAOkJjz76qEpZQCeHQMG+EJlw4MABKVWqlKxZs0aWL18ur74amoKAWFSE\nM5TPzGiAYJ6LVTw9ViSaDAqxVDDHtaWTa09qX0UR4d3AzQ5q0OD9OXMmqYXF3S16OXkw8uYv4mRl\nR6QXaiq4s8LjeRzTsf2lFTHj++GPR0xP+y1iXZBmiRsWzagAsGh/5plnZNq0adKkSZOQjieadB9Q\n+61tUPC3tWQ4iXbtz64luaVNmzaqK4SGmdq/btVi6n4AkTDU/ihrNwnjwdChQ+WXX35xer53797q\nS4EKz76A16Jbt25y/vx51aISEQpIqZg0aZJUr15deTKwuEBhM4RIIu0CYY96FoxG201ajfdee0NF\nA9Tp0jJbNIDVOi4ghDOcnh6rF2WMdIOCduGvXLmKx7ZLgbRNtBq42YEHYty48ZKenmbIO4P5wkJj\nypSp0qRFL7tVfd6sEfYaC5qVvWfPHir8W4WJx8VL05a97dt/N3O4FCt6hZw6ddrSnqJAvx+BeMRi\npcVWtLWbRM2kpk2bqohH6L/Ge++9p5wJkydP9ntska79kaT70ar9ZhoUfBkPImFtEEvaf+rUKaX9\n06fPkLS0c6ZqP7pG1L6+sRQrWc7uXcd+a9euLYePHKXuG1jfUPujpN1kw4YN5amnnpJ27dqpwopY\nECxatEgZBRBVoIdKlSrJ+PHjsz1fvvwlKz0W2R9//LEqzIQFY5kyZWKuKnokRQOE29NjBaIxQiFU\nBXOslreJG1x3raf88c7gPN5+++1suZm1a9WULVu2ykeDO9k97DCkwht0d4vH5PCh3bJw9kiZm3Fe\nRSrAqPDvsePSJMSeIn8/k0C/H4F4xGKhtWY0gnaqxYsXl3vvvVc5JvC/8vvvv6vvwsyZMyUWiSTd\njybtd+3wYKZBIdLXAbGk/SioOG3a56Zrf2JiksTHiaz6Za5TZB2udUf/PaacDYu+Hq22he6n5EoO\ni+77+5mEU/cBtT9KIhbAjh07VKrCzz//rIqW1KpVS/V/N6NwY6BEg9ci2j0C0US0RiiEwmthxbzN\nYJynq1Xd8TFwPR56Xi/8epT8vmSGxMXFh9RTZPQzCWTeYskjFumYGbGgLUyR4rhw4UI5efKkil7E\n2gI1m2JZ+6n7oTco5M8w9zsTLesAar852g9c1wGOugfdR/2FlT9/Kz98O96pkHMo9NCI9lP3Y4e0\nYEYsgGuvvVYVbyTBJ1o8AtFGtEcohKJgjhXzNoPhnXG1qjs+Rt0E1+Oh0FONunfIku8mqsehbKtk\n9DMJ5PvB9lGxC7xio0aNCvcwLAd1P/QGhWjW70Cg9pun/Y5/u+oedB8FnUtcVdFjIedgtlM0ov3U\nfWKaYQEWC7R//Ouvv1QYrwaKOt5xxx1Gd0uI5Yk1g0KwCubgGgLLOETMtb0i9o8KzN5uRoMVQhnq\nVkbejofnQajGEuhnYvT7wfZRscvFixdl/vz5ysAGj5kGPCOO3agIMQMaFIxB7Q+d9v9zcJfHQs7B\naqcYiPZT90nAhoUzZ86oYiMIi6hQoYKqh6ARFxdHwwKJSmLZoKCBcDhvbZf8waiXOtjpE6FuZeTr\neCBUYwk0csDo94Pto2ITOCXgiNi8ebNUq1ZNcuTIYX+tTp06NCwQ06BBITCo/aHT/h/mjJXatWvJ\ngtmha6cYiPZT94kjhlbhyIXMkyeP6qtuparkhAQDGhSCUzBHr5faNTIhFOkToW5lpOd4oRiLWZED\nRr4fbB8Ve2zcuFF1hdq5c6fqFU+I2dCgEJna7y4iMZa0v0+fPvLuu++GbBxmaD91nxgu3vjll1+q\nlpPoAmFFoqmAEwkfsVKUMZyg4CsWBo2b98xmldcqJjtGJnTs2EEmTJgYsiJ/oW5l5O14oRqLt88k\nFHUv2D4qdoo3rl+/Xnr27Cm//fabaWOj9hNAg4K18aQzWutl14jEp59+WqpXrxFz2h/KcYRT+6n7\nMV688ZZbbpGBAwfK8ePHpWDBgoGMkxDLwQiF0OHNO4Be0lhc3HVfV6lQ7XplSR81aoRkZmaErKhR\nsFoZeasP4cnWG6q2SuGOHGD7qNhqN4n/hT///FOuueaacA+HRAE0KEQGnnQG6VG4ub250SNSuWYD\n2bdzk3p84sSJoLW8DJUO+aoL5U77Q6mH4dR+6n6MRyx8//338uSTT8qxY8ekQYMGkpiYaH+tUaNG\n0rVrVwkn9FoQIzBCIXw4WqtxPYFR4bPPRqrKyEnJKdLgztZyd4tesnjuBFn41Si3bZjmzRyuvB0D\nBgywbIqWp/oQWtij1dpuhtJjQ2IvYgGpEJ07d5bt27fLbbfd5vQ9q169uvTr18/vsTFiITahQSEy\ncdQZ6GPFipVQrE0yMy7YtT8pVx5Z9t14tb27iAWra7+3ulDAii23qf0kpBEL+fPnl5YtW7p97Yor\nrjCyS0LCRjREKASrQ0KocLRWIxxv9Jixynig5VHO//JSHmX16xqqv12LGi34MlXKVqilxDlnzpwh\na1Xp77x7yhH95ZdfZMOGjfbnt29cLsOHj5CMjAxlZAkH9CCQYIPFSZMmTdSPKyVLluQHQHwS6waF\naNL+559/QXntm7bo7aT9ta5vpG66O3bsKFOnRp72e6sNAbTXripfVbas+0VpfzhbblP7ScgjFqwO\nvRYkViIUgt0hIdRArCtXruLWK/H9N2PVcz/OnyTt2rVV4XqXIhpyS4M7W6mIhmULpgYl39KMefd2\nbvC4NHmwhzS8p4NaSP3ywww5n35OtZyysieGxB5mRiyYDbU/doh1g0Isaf/C2SMlZ84E2bBhvQwd\nOtQhmtH62u/tvJbMHSu2LJs0vKeLpKedtut+Qs5EiRObKmzLorYkJiIWNFBw6eeff5YLFy6o9pPN\nmjULZHeEhGdBEoERChqhqJIcSny1PPr+mzHy2GO95dFHH5VRo0ZJpyf/JxWr3SiJSclBzbc0Y969\nnRsWScVKllVGhR8Xfi53N7+839GjU0PqiSEk1GzatEmlWCKPGnUXmjdvHpE3RyT4xLpBISa1P+O8\ntG37iLrJRurUiBEjIkb7fa1pwOFDu2Xtiu+ddP+7WcNVtCIMKYREEvFG34ic4FtvvVVmzZolCxYs\nkIceekjuueceVXiFEKsuSLRFCRYk+MGCJFIXJbAawmoOgYMlHC2C8BsVfeHNx+tWBOPavXu32/E5\ntjxyBCGP8N53795NeQa07Y7+c8C+sPCnNZK3Mfjazui8Y1GEc/B0bgf3blceCywuIunzJCQQhg8f\nLrVq1ZIJEybIsmXLpFevXlKvXj05c+YMJ5a4pXSeQhGv34EQa9qfmJikovYct4sU7fel+8nJuWTt\nikXZdL9py94yc+ZMy36WhJhqWIB3YfTo0bJ27Vr5/fff5ccff1T/gPj54osvjOySkKARbQYFPZbw\ns2fPKMu9lUAIIeonICywfv366jce43kNhDAirBA1FBAquG/Xpv9CIS+1oYIFH55Mb9t17drFYyik\nnjH42s7ovJ86dUpFJsz/coTTmJEjiucXfzdBhUFGyudJSKCcPHlSXnrpJZk3b56sXr1aFi9eLPv2\n7ZPChQvLBx98wAkmxA2xpv29evW0pwREmvb70v2mTe9WhSrd7/Os5T5LQnxhKNYQBoW7775bqlat\n6lS0EYVV1qxZI23atDGyW0JMJdpDJh0t/LBy+2u5D3UxI70hhHpbHrnbDguORx55RI0FY8DvvXv3\nqu3KlCkj7733ns8x4D34e9q0z6VJ817Ztnv++ecNzTvmJVeuFClRprKqF4ExJybmktLlqsnf+7ep\nuhFjx46zzOdJSLBBDnGlSpXkrrvusj+XO3du6d27t0yZMoUfACFuoPZ7136AtQiMEbixx3z50n5t\n/TJu3Di1X3fbGdF+t7qflCIlrqooRw7tUkaLb+fOpe6T2DYswIiwY8cOVb01Li7O/jxaRmGRQEg4\niXaDgoZmuYfoOVZJhuUeN+KhqhCtp5iRawghgDBj3BDxS4J9abx4D4Qez3lreeS43aFDh1Qo9YQJ\nE1X+JYS8cuVKsmHjRsm4cEHi4uMlR3y8amN1d/Pebsfw9NNPy0cffWQ/DxRQOnf2pJQsUzHbdtj3\nvFnD/Zp3PI9UDlR8hjHhwJ7NcuFCmuz9c52qUYMFBmopoKZCOD9PQkIF1hL79+9XRaEci0FhLcEO\nU4S4h9rvWfsTEnJKXHyc0n2kGiAqAOkGF7MuSpMHe7nVfnRfwvuh+3gPukzc2qSN5MiRkG071JRD\n/QO9Gq3pPtZpDe/pJIf275At63+Rfbs2Ss7ERBk5cqT07NFDPvuMuk9i2LBw8803qyJLDz74oLRv\n316SkpJk4cKFKh9o3bp15o+SEB3EikHBEb3e/WCiJxLBXQjhhfPpUqRoaXsIoWvBJb0tj7Dd1KlT\nnbwM307/RF2LmrZ8zD6meV+OkIsZGR6LKCHNAlEKjuehtbm8p/UTTtutX79Brr62piz6erR6DouR\n2rVr+Zx3vI7WkmvXOo8NYZ2YR+SRwrgQzs+TkFBRtmxZqVixoopYePzxx6VAgQIqvfL9999XawpC\nSHRpP3T/1MmjUqZ8NdO1/++/dsra5YuUAeGyho+QoiXKqxt5T9oPZ4JjdCLeA+2H7rtuh+Og2CI6\nVaCoJOo/+Jp37TUYPrKybE7aj/lCmif2Qd0nMd1ucteuXdK3b197VwgUX3rzzTdVTlK4Ycup2CIW\nDQquICLAm3c/mMf11ErJsfWT43bwBJjZUtF1DFi4DHyqsTR+oLvHto533d/FZaxoXWmTO+/r5rbN\n5aCP5svPP0zPth2OdfrkUVn9+wL5acFkn62u/JmvcHyehIS63eSRI0ekX79+qgg0ai5UrlxZ3ZQY\n6TJF7Y9eoqE9dCxr/+3NOpveUtFxDA3ubC2Dnm4ije53p+FjJDMzQ5q26K17TaDpPgpEutsO2r/w\n61GyfOks2bJls8+51zNfgLpPIl37dRVvPH36tPz7779Oz5UrV06mT58uBw8elKNHj6pWUVYwKpDY\nIVqLMhpBs/CH+iZUbzEjx4JLn737hPy4cJpaADw9YLw0a/W48gTA+2HGGOANuXA+zWNbx/lfjcxW\n9Klly1aSlua+cOL59LNqAeFuOyw6ChctJRWr3aCraJY/8xWOz5OQYJKenq7+BxxBygPCgVELBZGQ\naGPN1tXEVedhUIDGw6gQqzofydo//8vh8uOCqXbdv7t5L4FbE+mHZowBuu+p+DGer1LzZpXC4Kr9\nWBNUqHa9m/eclW0bf/O4HbS/Rt071PH1FFjUM1/UfRIN6DIsoIgSPAoaSHkYMmRIMMdFiEdoUAg/\nWismeBo8tYhyLWaEcEAsMHZvX6sWFe7aNcGAqacVlLc2VfnyF5HEpFxux5SQM0lyxMepyIOPBndS\nngKEIA4cOMBLm8t45ZXwtZ2eAoveWmqxQCOJdhDh2K5dO/vj5cuXq/aShLhCg0L0aD/qEuVISJC7\n/4sY0HS/SYtehnXfVU+h+0nJKW7Hk5ScW64sfW027cd6BPWYPLWCHP/Jiz6306vb1H4SK+iKO8Y/\nBFpMnj9/XtVTQIQCCi4REkqY8hB+3BVqRCFDRCL4KmaENIfOnTurPENPuY61atWW9PQ0twUg/Slk\nVbpslWwFlpB+Ycu6KE888YTbwpCeCmGijgzqKmjbdenSWVJTRzhth9ZRaInly2tklaJbhIQDrCV2\n7typUh7y58+voiH//PNPfhjEc8pDLqY8RLr2w3CAYopm6r47Pa1+XUNVS8lVm8tcU0OWzB2vus24\naj9qGrnTY6RndunSxb5djhw5DOu+u7FS+0lMGxbQWvLjjz9WC4FChQqpPAtUR/3222+zbYtFuNGQ\nZkIi3aDgq+WildEzdnfFmlDoqGbNGsqy76uIlLc2WfAQNGzWSYUbempFqbeQFY5Rq2YNWfTVSJl7\n4byKOsDCoOdjj9kXLa7FobwVw3Jd5GRlZdmLN+XMmaTCJCOp6BYh4aBatWrSoEEDlf5QuHBhdbMC\n40KpUqWybXvLLbfItGnTwjJOYgGdjzCDQqRqv95xB6L9wdJ9Vz1FSgEMBZd1/1JXiEP7tjppuaP2\nh0r3fR2LkJgs3gjPAiIVZs+erbwOTz75ZLZtUNihSpUqEk5YwCnyFxauWN2goKflolXRO3ZfxYf+\n+GOVuknwVUQKbRWxeED6gz2aYNYI1WXhsb6p2fbrqyCit0JWeLxv3z77tUnPfrwVw3ItFoXCjXnz\nF/mvsGNgYyUkVoo3Yh2B0Odff/1VvvjiC7fOiCJFikjdunX9Hhv+l9Ye3C3JJhSWJKHV/9J5Cqm/\nTx1Ps6zWR4v2+zNuM7Q/mLrvqqcAf+fNm1fXmiSUuu/rWIREuvb7ddW75ppr1A+sjwhtgkeBkEBx\nLMDoCasvMvS0XYr0sXsrPgTr+5YtW6ROnTpOQunOG+IuuuDixUxp1uoxt/t1147KHe6Ohd+VKlXS\nNQ+O7/d0PMc50Ao3GhmrNja92xISTZQuXVr9lC9fXkUxICqSkEgkUrXfn3F70n6tZSQ6u7jqrKse\nh0L3HbfX/kZklD/7cHejb6buA2o/kVgv3ugKIhJgVED15nfffVcVcvzuu+/MHx2JukJM7n60jg5A\nq/bs+mNlIEpaf2N3BQn9LUhk1bG7Kz6EhcG30z9RIYfNmzdXVn14JlD9Hb/xGN1itOfhJYE3BAsX\nWPlXrlwpq1f/ofa7e/s6Q4WRsE9Px9KDP+9nASZCzKNkyZJy7733qhpOqA7/2muvyYwZM3T/75Lo\nci4gUgE/kUKkar+/43bVPej+3Bmfysihl6KWmzS5266ZnvQUWEn3/dkHdZ8Q/RiO0+rTp498+OGH\nct1110liYqKyft52220yZ84clc9MiGt6gxbmqOG4gLC68cAbvjz5/lqzrTp2rfjQsGHD5cg/+6VG\nvTtk8bfjZc+OdaplpKPX45dffpENGzZ69YY4Wu21okaZmRekWIly8s/BXfLDnLG6Chp68rycOHFS\nXn99sOH3O47VXQEmI2MlhDgzfPhwVTW+evXqUqBAAVXP6c0335SffvpJ8uTJw+mKpUKNEbYWiFTt\n93fc0LW2bdvIuHHDVRvG4//+LWuXL1RdHhw1E7XXUOPAm55aRfe97UMbq+P5U/cJCUKNBQ14F26+\n+WbVOqpq1UuFWBAKBcMC/hnbtGkj4YR5ltZcMLgjkhYRnvCVf+hv/p1Vx44ohAceeEDWrl2rihhJ\nXJzExcXJPa2eyPbeeTOHS5MHe8hd93fxuk8N7PvBBx+UNWuw74sqAqJ27Vry1VdfSXJysqHxz5s5\nTJKSkqV7924e812NfHZGx0pINGBGjQUNdIdA4UbUbbrrrrvUc2fPnlXXGW094e/YWGPB+kS6QSHS\ntd9f3XfUO1+6H58jXu5u3tvnfsOp+/7OQSBjJSTWtN9QaAFuLJATqRkVAKo8d+zYUdasWWNsxCTq\ne09HWoqDXjRrNtouQZT27dqkfqNlUdeuXSy5sDAydojqunXrpVmrJ+TpAeOlYdP2YsvKcuv1gPAW\nK1k22/PIx4Q3xBWkVCHCAZEP2Dd+r1+/QT1v1PMC48cNDVsrD4SnTjXe3m/2WAkhzmzdulXlZmtG\nBZA7d27VFo5ridhaH0Qikar9/ur+2rWXohL16H5mRoZuPQ2X7vvah5ljJSTWMJQKASPCjh07BMEO\nsFxqbN++XXehNBJdxHrv6XC1ETKjxZWesaNYKyz1EFPNul+8ZHn5adHnHltI/fPXbtVX2lPupDb2\nfPnyOeV7AuwPvZ4xJvSd9nRu3tpYJSXnlsYPdJOUPPk87sfx/TifUyePSr78RTzmebrmpvozVkJI\n9rUEOkTAE+LoAcFaAq+R6CCa1weRqv3B0P2k5BTJzMxw+5qmp1bQfX+1n7pPSJANC0iDOHHihLJk\ntm/fXpKSkmThwoUyc+ZMWbfOuRALiW6iecHgD1pBQohYKNoImdniSs/YN2/erLwRjtZ9VEeuVb+R\nfDdruFoMaC2kFnz5mRS98mpZ8PUoiU9IsD8PbwgWLqjJggJJ2tiTk3NJenqaoTxVx9xH1zHc2qSN\nGqO3/eD9Xbp0luHDP5X5s1MlM+OCJORMlKyLmfLYY49lm4dIzaklxIqULVtWKlasqCIWHn/8cVVj\n4ffff5f3339frSlIZBML64NI1f5g6D4099SJo9leg/b37NlD3nvvPUvovr/aT90nJMiGBfzDQfT7\n9u0rTz31lFy4cEFq1aolixYt4qI6RoiFBYMRQtVGKFgtrjyVXEEnGEQhuHoIrrjyasm6eFEWzB4p\nczPOK28BhD1nUoosnjNals7N7g1xHfv2jctlwVefefRyoBc1et578sxc9ryMUcfKmTNJjeHuFr18\nVpmGJwJ53nHxOeTu5r3sczn/yxF+e0r0VLIOVtQJIZEIIh7hkOjXr5/6P8b/YuXKlWX69Oly0003\nhXt4xCCxuD6IZO03qvuoqQDDg6b70Nwl8yZLjvi4bNqPFAUr6T40F8Um43Mk+NT+YOi+4zio/URi\n3bAwceJEVWcB4k9ii1hcMFgNs8Py9HhA0AsahYqyeSlmfybFixeTo0f/lYbNOkjlmg1k786N/0Un\n9M7mDfE09u2bV8i8bB6QVKlZs4bUqXOdV8+Mo+dlwIABMmXKVMmVO5/8tXebU6SE45y4njM8FefO\nnpSSZSra53LcuHGq+43j+zx5Stwdw4x5JySaQecHeDFRAI1EPlwfRI72B6r7111XR7VoHDVqtBpL\nxWo3yLIFU+X7r0dl036AQolW030YTcpWqKUMEjlyJHjUfjN1X+/cExKpJBjtPf3111+bPxpiWbhg\nsA5mh+Xp9YBg8Y/0J81LoVVFhsfxgw8+UAubJd9NdIpOQDSTozfE09ibtXpMPnmjqyz59pKXI1eu\nFKlevZqs37BRmuj0zEDcscDA8WbO9J7v6u6c5395ad/3tH7C61yalVNrtueJ3g8SaaDCNLySJLLh\n+iDytD9Q3cfzuAlGe0loIfTTk/br1X3cYFeqVFHWrd/gFEXgS/fffvtt9Xv8+DEGdH+EGjt039tc\nmllLg9pPohlD7SbPnz8vt99+u7z44oty3333Wc7CxpZT5hEtbaGiCTNbXBnZFwo6bdmyRYUsw6Ph\nuC/NQ4E6Cu4s8pf61ddwezwsLtq1ayuTJ0+RtLRzdm9Czz6fKm+Ct3G58wC0bt1KGRpQJErvOX//\nzVgZ9NF8+fmH6T7n0vF8/fVYmPkZ0vtBIrXdJEBryUaNGkn37t1VvaZAx8Z2k6GD64PQYpZumKn7\nerS/Y8cOMmHCRI/HW778d9VdYfqMGZJ27pyKILy18aW0Bmi/t3G56h+cEp06dVTa73hvokf3UZvB\n11wGovtG594T1H5iRe03ZBGYPHmyrF69Wlq0aCE5cuRw6uHarVs3+fDDD42NmlgGeiCsi5lheUY8\nIFhUoICru3Fp26I4oydviKex16hRXS0+jHgT3HkApk5Nlfz582fzcvg654Vfj5KfF33ucy4DyakN\nh+eJEKuxdOlSVa/pm2++UUZHx8VKw4YNZc6cOWEdH3EP1weRrf1m6r4e7R89OlXpO9pbuhv3yJEj\nZdq0zw1FELrTPxwPkRSO+ufrnLdt/E2O/LPf51wGWkuD2k+iHUOGBUQpVK16uYCJIyhCQiIXLhgi\nA7PC8oJVjNBbHuj69eucii6hbz0WS5pRwfU98CY0ur+b8iaY0QrK2znHx8fL70tmSps2j6gbnWCl\nF5g172yDRSKZOnXqyJIlS9y+BqMgsRZcH0SH9gezGKEnLUZEInR+4kRt3LnVuLUoRm/ab0b7Z++6\nn0PGf/KiinbQah1Q+wkJcVcIXBRKlCiRLSRKL6dOnZJjx47ZH5cqVcpjSkWgoUfEN1wwxGaLK7OL\nEum1yIOsLJv999mzZw17E/z1AHhuVZWqqmDv2rVbFaid9vnnYsuyqb7cZhdXCqfniRCrgLBpVH8v\nVKiQXHnlleEeDvEA1wfRpf3B0H09eoS1vKPuA5yD0QhCf/TPm+6jWOTWrdtUCiYcHCtWrJDNm7eo\nx9R+QvzD7xUy0hxeeuklVZgFIC8SYUz+goJvgwcPVu1eDh48qAo4eVoAP/TQQzJ37lyZMWOGtGrV\nyu9jEc9wwRC5mGVRN7MokR5vCG7acSzntIURkjMx0as3wdO4jHhf3J0zFheuxSLnfTlC6ja4R4qX\nLG96eoGVPU+EBBusAbp27aqcDKBZs2Yya9Ysp9RKEl64PrC27gdiNDZb933pEVITpk77PFsxZtwD\neIsgXL50lsdx+at/enUfXTDKXltT7n3oyaCkFlL7STTjl2Fh586d0rdvX1UBHsUbN2/eLD179pR7\n7rlHFWDyhy5duqifrVu3qmIwnkC4NC5IiJAgwVk05M/IxbaREYTZBXvMin7Q4w3xlvKwcHaq2xzM\nnj17qGuFp3HhuS5dOktq6ohsnohevXq6fY/rOcNrivZWTTyEY7bs0NdwO89I9DwREkwQoQSnBMKg\nH374YTlw4IA89thjMmzYMPX/QMILDQrWw+q67ysqIC4+zq2+Tpw4ThV3RF0EVw1r3769ckB6q3fg\nj/b7q/twKATSytsT1H4Szfh1Nfrjjz+kadOmagEAEDa8ceNGWb58ud+GBT3s27dPFYL5/fff5Zpr\nrjF9/0SkdJ5Ccup4Gjs9RBDeivVB+DRvBvAnoiHQokSOnhQsduCJcG3/1KZNGxkxYoT70MWMDOnY\nsaPMnDkumwdFz8IpKytLFs4eKXMzzkvOnEmqNZbec0bElLeQytMnjwYtvSDQeQ+G54mQYLJt2zaV\n/oibBoCaTa+99poq4kjCBw0Kka/70BN/ohkD1R/geDwYC0+cOOHU8rlt2zYqUtGTvnbo0MHetjIU\n2u+P7hcuWoraT0iwDAuoieBanLF48eKydu1aMRt0wezcubNabOAY3sANDKy5ji0xCIlGPBUruph1\nUd2wa96MhIScykOQceGC6TmCrrhr91SlSmXZsmWrylHE4w4d2tt7W3sLXcSNBn788aBgTsaOHSfN\nWj0uDe5srRYDefMXUS0jx40bJ3369PG5H28hlUnJudX+1q1abMn0gmB4nggJJp7WEo51l/RA7TcH\nGhSiQ/cdtdeMqAZ/td913QFHgdb2cebMWR51H/XajGhYoNqvR/cdx0ntJ8Q3fl9pkLowevRo++Nf\nfvlF9u/fb38OaQ0NGjSQQPn0009VYSdcmHwxZMgQZYAg/i8iTmXQCBNJeCpWdOTQHlUMCb2R//5r\np6xdvkiaPNjLlPaD7rwf2nP58uWTN998U7WKatL80vG+nf6JrF27Tpq2fMxt+yc9ofv+eFD27t2r\n5uSq8lVV5wh4GIA/EQaeQjjR8qr29Y3VQsXq6QVmeJ4ICRWoreS4ltiyZYvTcyVLllQRkt6g9gcG\nDQrRo/uetDcY2g/NBdOmTbPXS3K37nBs+axH9/3VsEC135Puz5s1QtVYwDlFQmohtZ9YiTgbQgN0\nMnnyZFVjwRvIiYIFUy9ajQXH4o1//fWXVK9eXb7++mspXbq0PVRy6NCh0rJly2xWQ3deC3SrWHtw\ntyQ79MWOdbItIkSYAhFhQNQrV66iFhKa5+LC+XTp/8SdcnfzXspqP+jpJqpFk/Y6WPzdRFk6d5xs\n3rxJtzi6y+lEPiPAYgLRCCisiLDDhJyJcmvjNnLHPZ3k9efu8Xp8GAyxX+zj7NkzyhPQtWsXv70q\n2bwl/43h7ha9JEeOBL/PWdufNq7ExCSx2bLU9cXoGAmJNqCvV111lboW5TKorz///LM88sgjXre5\n5ZZb1I2LN6j9xuBaILp0H8/hcbC1HykHI0eOUpEStqwsVVyxbIXa0vmpoSHTfbO131X3Ma5KlSra\noz6o/YT4p/1+/Tc/+uij6ifYwMoITyiMFI4nNHDgQFmzZk22LhTwhOKHuIdeiejBnYUdIfqZGReU\nl+DUyaNyPv2cKe0H3eV0okgSFhdlK9SSvTs3qEWN9hq8++fOntJ1fDNC992ND10cTp04KsVKlvPb\ny+AupQAwvYAQc7n55ptVwcZAofb7B9cC0an7IBTaj5twpDvc0+oJB90fId9O/zhkum+29ntKJWSb\ne0KMETbXW3p6uvz9998qOgFoiwz0sq5du7bs2bPHafs8efLI8OHD2W7SD7iIiCwc0wvQgs1T4SXX\nYn3wJmitGhGxkJScEnD7QU85nVjULJg9Uvbv3myPkMCCBr8zMy/ID3PGS2JSLl3HDyR8z9v45s0c\nJsnJuQwXMHQdF9MLCCGRDNcC1sYxvaBMmTJ+6z60L1/+IkHX/nkzh0uj+7s66b7q6PTVaJWKEGzd\nD6b2u46L6QWERJhhAQUftVBIXEi1SIjZs2crw4Ir2IYtJ/XBRURkoYXijRo12im9wFPhJXcW9vfe\ne8/uzah+XUNlvQ+k/aCnnM5L3RvOq7+Rf4jQS3gqEH4ocXFyMTND4uLiVR/oYLU/xHzh/D2NDxEV\nCxbMl0qVKgV8LEIIiVS4FrA20DLUCYH3XUsvgLGgZ48e0q9fP790H9pX6urKAWuvN+3HuuTA3q12\n3YcToVCREpKZeV6NPZi6r80XtZ8QaxM2w8INN9yQLSrBG5s2bQrqeKIBLiIiEy2sr3S5atnSC7wV\nXnK0qDt6M5AniBDhRV+NlLkXziuPAQwUMOTB2q9H5L1VS074r50TCjUhp1Er2oS/Me4dm1fJvFnD\nnY5vZvtDzBeKRSKv0pOHBHlghBASi3AtEBlAy4YPH6GcCY7pBUg7QP0Cf3Rfi2KoXaumqi+gtW00\nU/tF4mTLup/l7ha97cUi9+xYp8ZetkJN+W7mcBXVgPVBMNoeU/sJibLijZEC6jHgAhorxRtZiCny\nizLddncHWfzdhIALLznmBYJDhw7JhAkTZMKEiX63n3r99deVYaNx8552D8SCL1Pl4sVMtRBC5Wlv\nxSKXfDtWRQ7gJt8sj4VjEatzZ0/KjwumSZMWPbN5SIxUwCaEhKZ4Y7CINe13hWuByAH/P5UqVZaM\nzEx7AUZn3R8rmzdv9lv3tfoAZmv//FkjJMuWpYwI3opFLvpmjCybN1HWrFmtiqibBbWfkCgs3kis\nBb0SkY8Wdli0RFlTCi+55gVOnTrV3g7K3/ZTrt4QeCB69eopJ0+elEmTJukqFpmcnOyXUcFdeytP\nYZoly1RUz33/zaXxwcODgq9mekgIIcTqcC0QeUDLkPoIzNZ9PDZb+9u2bWPXfeBJ+ytWu0EVcj59\n+rRuw4Iv3QfUfkIig/hwD4AYW0TgB20j82fkUq0jCyZdzdaREYgWdnj44G574SVH/C285K3IEcIa\n8RteCCwY8Lo3tJxOREusXLlS/UZnljfeeMMeKulYMCqQcSN3El4SRCPUr19f/cZjxzayrmGaqOtw\nT+snZNBH86Vhsw6qaNPgwYPZDpIQElNrAcC1QGQBLcuVK8We0pddP3Mb0v1gab+j7gMztF+v7gNq\nPyGRASMWIgh6JaK7jdRV5aur1k1mFT/yWoAxQG+IY+srM4pFumsf5c674q7tFo7386LPTS0SRQgh\nVl8LaAYFAOcCiRygVd27d5NPP/00m34i5fCxx3ob1rNgab+r9gZaLFKv7mtjofYTYn1YY8HCCwZX\nEKGA6ATARUT0dYWAhwGLAa0rBKz+Xbt28ZgT6St80DEnMZC6Dd7G7FgsEt0gLvxXrNHbuAMdp+ux\n/T0eIcQ4rLEQPmhQiM6uEJ999plcvHhRdTRKTEySHj26Z+sK4U/aQLC031V7EcFQuXIl2bp1m99a\nbGSM1H5CrK/9NCxYeMHgDhoUohetAFPevHlVfqJWiMmXIcJbUSZ3RZjMLHDoWizSsYCUXnbv3q3C\nIJ8eMN6pCvW+XZvko8GdVCimO++Ka8EqQkjwoWEh9NCgEN1Ay/bt26f+9lTs2B/dD7b2uysW6a8W\nG9V9d8cnhAQfFm+MELhgIO7CDr0VPfInfNBdESZ/W0B585C4hkrqCa/0p72Vt1xN12MTQkg0wfVB\nbAAtq1Spktdt/NH9YGu/u/RIf7XYqO4bPR4hJDQwYiFMcMFAjGA0xNGIhd9fD0kgBDuyghBiDoxY\nCD5cHxCzUhusrP3UfUIiB0YsWBQuGEggGC3KZMTC76+HJBAviBneFUIIiWS4PiBmF2O0svZT9wmJ\nPljpLERwwUDMIJDwwUDaVQEcD9WfcfP//PPP+53b6MsLggUL9svcSUJILMH1AbGC7odD+6n7hEQX\nNCwEGS4YSCC4Wvg9tVwKpC1lMNtV+esFYe4kISRW4PqA6NX+UOh+uLSfuk9I9EDDQpDggoEEgjcL\nfyjCB832kATDC0IIIZEI1wfEX+3v06dPSNIFqf2EkECgYcFkuGAgZuDLwh/s8EGzPSTB8IIQQkgk\nwfUBCVT7g50uSO0nhAQCDQsmwQUDMQu93v1ghw+aGRkRyhxRQgixElwfEDO1P9hGeGo/IcQoNCwE\nCBcMxGys4t03s6BiKHNECSHECnB9QPyB2k8IiXRoWDAIFwwkWFjNu2+Wh4StpQghsQDXB8QI1H5C\nSKRDw4KfcMFAgk20evfZUpIQEs1wfUACgdpPCIl0aFjQCRcMJJREs3efraUIIdEE1wfELKj9hJBI\nJs5ms9kkykhLS1M3L2sP7pbkXLkC2hcXDCTcxZyCWQGaEEL81derrrpKXZtyBaivVtZ+PXB9QIIF\ntZ8QEonaz4gFD3DBQKwAvfuEEGItuD4gwYbaTwiJRGhY0LFgAAWTgl+FnxBCCCHWhOsDQgghxDM0\nLPwHFwyEEEIIcYXrA0IIIcQ3MW9Y4IKBEEIIIa5wfUAIIYToJ2YNC1wwEEIIIYTrA0IIISRwYs6w\nQIMCIYQQQrg+IIQQQswj5gwLoHSeQnLqeJr6m0UZCSGEkNiFDgdCCCEkcGLSsKBBowIhhBASm9Cg\nQAghhJhHQqwuIgghhBASe9CgQAghhJhPQiwuIpAGwWgFQgghJHagQYEQQggJHlFtWDh94U/JyJFk\nNygAGhQIIYSQ2DQqaGsBwPUAIYQQYh5RbVgANCgQQgghsQkNCoQQQkhoiGrDQv6MZCmY++pwD4MQ\nQgghIYQGBUIIISS0RLVhoUBSmXAPgRBCCCEhToMsllDA/pgpD4QQQkjwiWrDAiGEEEJiC0QrYnVD\ngwIhhBASOmhYIIQQQkhURSvmSrpcpJEQQgghwSc+BMcghBBCCCGEEEJIlBKVEQs2m039Tk9PD/dQ\nCCGEkKhB01VNZ60EtZ8QQggJn/YnRPPJV6hQIdxDIYQQQqIO6GxKSopYCWo/IYQQEj7tj7NZ0e0Q\nIFlZWXLixAlJTk6WuLg4iTTS0tKkcOHC8u+//0quXMwT5VxaA34vOZdWg9/J0M8llgxYWBQoUEDi\n462VTUntJxq8NpgD59E8OJecy1jQ/qiMWMAJFypUSCIdfMA0LHAurQa/l5xLq8HvZGjn0mqRChrU\nfuIKrw3mwHk0D84l5zKatd9a7gZCCCGEEEIIIYREFDQsEEIIIYQQQgghxDA0LFiQhIQEGThwoPpN\nOJdWgd9LzqXV4HeScxlN8PvMubQa/E5yLq0Iv5fWncuoLN5ICCGEEEIIIYSQ0MCIBUIIIYQQQggh\nhBiGhgVCCCGEEEIIIYQYhoYFQgghhBBCCCGEGIbVAS3Apk2bZNeuXVKuXDmpWrWq/fmzZ8/KnDlz\nnLZFj9EHHnggDKOMHA4cOCA///yz1KtXT8qXL29/HuVEVqxYIYcOHZIaNWqo+SbeWb16tWzfvl3u\nueceyZs3r3pu+fLlsnv3bqftXOeaXGLu3Lly+vRpp+lo2rSp5M+f3/746NGj8ttvv6n+wDfffLMk\nJSVx+lxIS0uTr7/+2uk5zFPz5s11vU6yc+LECfn999/V//WNN94o8fHxTtfKlStXysGDB6V69er8\n3w7S/EOPcubMqa6fefLksb+G68HevXudtr/++uulbNmy/Cp74bvvvpOLFy/Kfffd5/T8P//8o3Qr\nX7580qBBAzXnxDPp6enqenr11Ver7x2AjkHPHMmdO3e2uSYiO3fuVNdPR/C/q82ldo3F9RffzVq1\naqm5JtnBHO3Zs8fpufr169vX775eJ87ge7dmzRr566+/1PexaNGiTq8fPnzYvi7AtTIxMVH8hYaF\nMIJFW9u2beXkyZNSunRptZi44YYbZNasWerDPHLkiLRp00Zatmxpr9ZZoEABGha8kJWVpeYMi4iP\nP/7YviA+f/683H///cqIU6VKFTXX/fv3lxdffDE0H3YEAgMMDAp///23bNmyRSpVqqSeHzFihLrw\nQAw1ihQpwpsPNzz99NNy5ZVXSsmSJe3PwXigGRbmz58vDz30kNSpU0f9v1+4cEGWLFkipUqVCv4H\nHEH8+++/6v+6RYsW9psC3CRohgNfrxNnxo8fL88884xUq1ZNGaux2MBNA4wx+A4++OCDsm7dOvX6\nr7/+Ki+//LK88sornEaT+Oijj+SDDz5QWgTDIm5EvvjiC7nrrrvsr2/YsEEZdTRwHaFhwTPffPON\ntG7dWs2T483ul19+KR07dlTGGyymsZZavHixFCtWjN9nD+D/fdiwYWp9qt0MY+5wjYVexcXFqeeu\nuOIKGhbcsGjRIrW+vPPOO+3P3Xbbbfa5hCG8WbNm6v8e6yqsR9944w21XiDOfPrpp8rBBWegBv53\nNcOBr9eJs9EAa6L9+/dLzZo11f3P0KFD1TofwJjYvn17ue6669T6H//nWI8WL15c/AJdIUh42LFj\nh23VqlX2x0ePHrXlz5/fNmnSJPV49+7d6NhhO336ND8infzvf/+zderUyVamTBnbiBEj7M9/8MEH\nttKlS9uOHTumHi9ZssQWHx9v27ZtG+fWA/fcc49t0KBB6ju4ZcsW+/MdO3a0Pf/885w3HZQvX942\ne/Zst69duHDBVrx4cdtbb72lHl+8eNHWqFEj2yOPPMK5dWH//v3qe3j8+HFDr5PLrFixwpaYmGhb\nunSp/bkff/zRrjOffPKJrUSJEkqPtNdwrdy8eTOn0SRwTUhPT7c/xvW0atWq9scPP/ywrV+/fpxv\nnRw5csRWrlw5W58+fZT2a5w9e9ZWsGBB9Z3Wrrk333yzrVu3bpxbDyxbtsxWq1YtW4sWLZTWa2AN\ngGtsRkYG584HWHs2aNDA4+vQfHxfT548qR7PmzfPlpCQoNb8xJl27drZXnrpJcOvk8tgfYn/6/Pn\nz9uvj9o6IC0tzVa4cGF1rwTwf37bbbep+yl/YY2FMHLNNdcoy5BG4cKFpWDBgioFwhFYjBDi5xoa\nSZzZvHmz8qbDE+TK7Nmz5eGHH1bzC26//XapUKGC8nKQ7IwZM0ZFf8DT4w5YM7/66itlaUc0CPEM\nomRgCYYH2BEtDLJXr17qMULRe/bsqb6TmHuSnWXLlinPumvoo97XiShPJDy60B98L+HtueWWW+yh\n+LhWwisJPQJ4DZ5111QTYhxEhDimPOGzcNV9eIhxjcV1AlEkxDOPPfaY8va6eilxPcC8dunSRT1G\nNFO3bt3Ud5xk58yZM2p+Ro8e7TFd5IcffpB58+bJvn37OIVe0FKZsX4/fvy402v4/iEaBFF14O67\n71ZRit9++y3n1EN0t3YtdLfe9PU6Edm6dauKpHnzzTdVFOLChQtV5AwiacBPP/2koufx/w8Q2dW9\ne3dD10oaFiwAFmwTJkyQRx55RKVE4IKj5a/hZnjKlCnqZhkhU48//ni4h2tJMjMz1U3whx9+qNJF\nXNFqWDiCsFI8T5zBgmHQoEEycuRIt1ODcD7ksU6aNEk6dOggFStWlFWrVnEa3XDvvfeqkGYYaho1\naqRy1o4dO2b/TsLQ5fh9xXfy3LlzKv2EXAb1J3AtnDZtmvofx40uDDII4dfzOrnM2rVrVfg9DAa4\ngYCR4Y477lB51bxWhg7UrsH39X//+5+8++676kfjpptuUou+iRMnSrt27aRy5crZDJPkEphD3Fg8\n8cQT2aYE11iE8eL64HiNRerUqVOnOIUuPP/889KqVSsnh5cGboJxjR03bpwKn4buP/vss5xDN8BQ\niPmZPHmy9OnTR9VPmDp1qtP3kutRfaD+DwyrWG8++uij6lqIGgF6XyeXdR/GbBhhBw4cKEOGDFHp\ny5oxC99J1FtwrPWDayWMDdqaVS+ssWABYP3FjQQWDljgad5K5K99/vnnTh553NTdeuut6gJPLoN/\nkmuvvVbVUfBkeHAtQoJ/MjxPLoMbsc6dO6v8QFjQ3Xl+e/furX4AvqvwsiMvC3UYiDO4yXX0BsE6\n/NJLL8moUaM8fie17yu5TKFChZyuhbC+41qIm2PcePl6nVwGHh1E0WCOEJWAGywYYhDJgBsLXitD\nA/Kr4WVDZAJufOFI0HjqqafUD4ARF9fkTp06ccHsJnION26omeBYfFSD11j9LFiwQHktPd2UlShR\nwukai/UqbuoQ/cmC4s6gVopWLwVA77t27apqLiD/n9dY/cCZqjlUsd6ERx0OLThs9LxOLus+fpo0\naWKvLYfoBURzIXLWzGslIxYsQGpqqlpgbNu2TV2sBw8e7HY7LP7gyUBhQnIZ/FO89dZbam4gfPhB\nGBq86AjbA7iYu3qB8ZgFnJxB2BMMWKgIi3nUupIgFcfdhRqLOVyYcJNCD5B3YAlGVJL2/4vvHjxn\njhdtfCcxpzAqEs8gegtFMD1dC329Hsvge4cbAi3VAZ5IGKvh0dBe57Uy+KA7DAo2ooNR3759VZFm\nd9fQHDlyKMMC1gZaVAm5BDxvKDCKdB7oFTQf2o+/EZWD7zIKljlGLuG7jQWzY2ceIvLkk0+qkHys\nATB/iFxE96fp06e7nR4Uf0NBTF5jfYP/X3jVtagjXmONgbUR5nLjxo3ZUsf0vB7LFPuvWC2iaDXw\nN4qGI+ILr+NvxzRcXCuREqWlkOuFhoUw3xA7AmsRvMRoQ+XudSwqYHxgxXj3OatYoMFAgx+Ek+Mi\nDgs8gKfYsVUS5haLEC2/iFwCFxDMCdJzMI/IwwIw0CAiARcdLNgcwQ0Jbk60dpTkEsirdM2Nxlxp\n/78whAFtjgHC0rBYQ6V+4vlaCcs7jFnaXPp6nVymYcOGylvueLO1Y8cOe+cS/P/DkKiBxQZuHnit\nNAdcEzSN10BIKlIf8L3NyMjIFnqK6wY67yQnJ/Or7OJsQSqZpvvQfGg//obRFhFLiBTT1gHaNRZG\nRxhsyGUQeaDlqmuRNDAuaLVVXK+xmGdcN3iNzY7rXMEpg7WTNleu61FU6V+/fj2vsS4gWgv/x67X\nQqxTEeHl63VyGTgTYFDF/6wG/obhAI4sXBNxj7l06VKnayXWqf62541DBUe/3kFMA3UTUNgFude4\nkcAHCmsxwvoQxot2iUiTQFsaWOKQQ4SbOtwQu6sjQC6DnDZ4gbTCeLhwo6Vf48aN1T8QQtPgscNN\nndY6iWQHqRDIs9LaTWLRizaTSDlB8UuEVKNg5ttvv608HuQyCClF+CM8kcjzxf81Wsni/xxtZUG/\nfv1U/QV8V7EYef/999WCwzGMkogMHz5cLXDRFglFhZC7ijDoP/74Q6VB+HqdXAYLMVwLoTHwUEKD\noDOa0Qs3FLVr11Zhu1gAow4DjIb4/vJaGTinT59WizWEjyN978CBA/LZZ5+pm2DUU8ING4yLeB25\n2rgpwetIq+rRowe/yj6iP6FFjil8KOqIddULL7ygCmAj5QeGctS7IZ5BdB0MWWhNC1ALBAYahFID\nPA+jzcqVK+lUcOPoggcY/8eImEFLRBh08f+t5bOjjgX+x3EdxhrqqquuYvFGF3Cji3nS1puIRMBc\nvffeeyod19frxBm0NMVaCakQuPV/55131D0SaqoBpEIiig6v457pk08+UfdIiGj0BxoWwgxuMlAF\nHhdoeC2QD+xoAUZVY1iP8Q+EntYoUEhLnG9wk4uLtuMNGhYVuOjghgMLZ/xD0QPkHXgrMZe44UWO\nJUC4Lm42YFRAsRf0xa1fv77h/4FoBgsIFGbFzQMMNMj9wwLCEYSd4uKNPGvUqtB6XRNnfvzxR3s0\nEq6FyDl3vBb6ep04/19jgYFQZxQRww2rY69qeCpxrYQHEyHPWKQxisbcaCYYFJF2BsMXbjqQGqHV\nCXB8/corr1TGSRiDiHe+//57ZWDEglgDnmIYGmFAQ2QdrgvQf+IdOLbgqXS8QYNBBumRiLqpUaOG\nWo/yupAdeNLxnfvll19UCiQMtLj5dTTMImoMBkM4FOrWrauuwY6dYsglUDxQW29Co1q0aKHmS+/r\nxBmskeC8wncNRkIUb9aAsQHGL/yfw5mA/293hVx9QcMCIYQQQgghhBBCDMMaC4QQQgghhBBCCDEM\nDQuEEEIIIYQQQggxDA0LhBBCCCGEEEIIMQwNC4QQQgghhBBCCDEMDQuEEEIIIYQQQggxDA0LhBBC\nCCGEEEIIMQwNC4QQQgghhBBCCDEMDQuEkJDz9ddfS+fOnQ299/Dhw1K3bl05c+aMhIpRo0bJJ598\nIqHkxIkT0rhxY0lLSwvpcQkhhJBgcNddd8kff/xh6L0DBgyQTz/9VELFqVOnlAafO3dOQsk777wj\nU6ZMCekxCTELGhYIISHnyJEjsmXLFkPvvXDhglqYZGZmSqgWF/3795cHHnhAQkmBAgWkRIkSITdo\nEEIIIcFg7dq1cvr0aUPv3bVrlxw4cEBCxdChQ6VSpUqSkpIioQRrjZdeeknS09NDelxCzICGBUJC\nALzzI0eO5FxHIJMnT5batWvLVVddFfJjd+nSRXlosrKyQn5sQgghxlm2bJk0aNCAUxiBwIExYsQI\n6dq1a8iPDWMG1hszZswI+bEJCRQaFggJAfDOHzx4kHMdgXz++edy//33h+XYt9xyi0r5+OWXX8Jy\nfEIIIcY4fvy44bB/El6+//57FalQs2bNsBwfaw6sPQiJNGhYICTMoGbAM888Iw0bNpRHHnlEFi9e\nnK2ewO+//y4dO3aU22+/XYXIOeb87d27V3m277jjDnnhhReUh12rX7B9+/ZsHpMvv/xSunXrZn98\n9uxZGThwoDRp0kQefPBBJzHz9X5tfL/++qsa34033ui2JsD+/fuV5f/OO++UPn36ONVH0DNGd6xZ\ns0YeffRRtc8hQ4Y4pUZ4OyeA4w8aNEiaNm0qLVu2lN9++82j12LFihXqHDW0c160aJG0b99ebr31\nVnWs8+fPy8cff6w+R4Qy4n2BvAfExcXJddddJz/99JPXuSCEEBJZTJ06VZo3by6NGjVSeuCo66gn\n8OGHH6p8e+T5Q8dcDczDhg1Trz300ENKWxzrFzz11FMybdo0p+2hM+vWrXO6eW7Xrp16/umnn5aj\nR4/aX/P1fm18b731ltx9993Ku+8OPA8dxhh/+OEHp9f0jNGVjIwMef3119WctW3bVtavX+/0urdz\n0l7HugHrJewH+3PHzz//7KT72jm/++67au2A4+Ozwzpkw4YNaiy33XabvPbaa05rESPvAfXq1VOf\nt81m8zgXhFgRGhYICSNYSNSvX1/++ecfefXVV+WGG26Qe++9V+bNm+dUT+CJJ55QgvTyyy/LnDlz\nlEhpN8i4mU9ISFB1AK644grp1auXvX4B9u/qMcFN7tatW9XfENWbb75ZNm7cqG74Ibh9+/ZVxQr1\nvF8bH46JG2PUA0hKSnLaHjf5N910kxJInGOhQoWkX79+TnPg7Rie6Nmzp1pswSiDBVrv3r11nRPG\njPHAmPDkk0/aX3dnEPn777/VzT9qHWho54zPAoul5557ToYPH67SJXbv3q0+m8qVK6vPUVsoGnmP\nBo4N4xEhhJDo4L333pNnn31WWrRooRwCcCjcc889TvUEoBfQH+hmlSpVpFmzZqqoL3jzzTflf//7\nn3Tv3l05ErDN0qVL7fULYLDHusIR3NBqr0+aNMl+Aw79wXEQIafdaPt6P8YH3cQaxFMNIhhFYPSH\n46NTp05qe238eo7hDqQGIhLklVdeUekC0PJ9+/bpOqdx48ZJ69at1ZoLY8bz77//vtvj7Nmzx0n3\ntXPG+xITE9V8582bV30m+AywXzh9xo4dK6mpqQG9B+DYJ0+edJovQiICGyEk6Fx//fW2gQMHZnt+\n9OjRtquuusqWkZFhf65Pnz62W265Rf29f/9+mKttK1eutL8+atQoW/369dXfI0eOtFWqVMmWlZVl\nf71Tp07qeGDNmjW2pKQkp2OOGDHC1qBBA/X35MmTbeXKlbNlZmbaX589e7atYsWKut6vjW/p0qUe\nzx3jvfbaa20XL160P9exY0fdY3RFO+acOXPsz2F+4uPjbf/++6/Pc5o4caKtRIkStrS0NPvrmH/H\n8Wls375dHevw4cPZjr9q1Sr7c0888YTtuuuusz/GvlJSUmwrVqww/B6N7t27q/kihBASOUB3XLVN\no0iRIrYJEybYHx86dMiWM2dO2y+//KIet2vXztayZcts71m8eLH6u3DhwrYvv/zS/trWrVuVxixZ\nskQ9btKkie2DDz5wen/+/PltP/30k/q7VKlStqlTp9pfwxoCawmMWc/7Mb67777b6/ljvDNmzLA/\n3rJli19jdAXHdF0X3Hnnnba+ffvqOifoPtZMjqSnp7s9VosWLWzPPfdctuO3atXK/vjAgQPqfL7/\n/nv7c4MGDVLvDeQ9YOfOnWo7fC8IiSQSwm3YICSWgcW+Vq1aKuJAA+F3EydOdNrummuuceoWAEs2\n+PPPP1UOIELmNerUqaO748KmTZvk33//leuvv97+HDz3O3fu9CsEr1q1ah5f27Fjh/LMx8dfDpBC\neL+viARfYB8a2D/Gi3H7Oie8jvcmJyfbX3ecf0eKFi2qfh87dkxFgzhSvnx5+9/58+d3+oxwrvBM\naJ9TIO/BscuWLatzVgghhFgZeNwRoo9wd43ixYtLqVKl1JoAXnjgqA+aZkAf8H5oHHRPo2LFirq7\nF6DTEborIJoAkRMahw4dUnpthu5r5+io02Z0WHDcn/YYc+brnDBvqHPlmnbpGmHpqP3QXldcNdzd\nc750X897cGysCQoXLuxxLgixIjQsEBJGIBquOYB4rFdMChYsKKtXr3Z6DgsOjZw5c8rFixfVDbVm\nfHAMM0RawrXXXus2PxLb+3q/Ro4cOTyOEcdwTXVwPGe9x3AF53nllVfaRRjvx7z5OicYCFzDLz0B\nwS9XrpwyRmDhFg6QQ9qqVauwHJsQQoi55MuXT+meow6i8w90TI/2wwANYzg08Oqrr1bPIYXOMZ0P\n+3fM24fGaq/nzp1b3VC/+OKLKsXCkZIlS/p8vx7dxxixD4xRM4z7M0ZPOK5vHNdLvs4pT5486nVo\nv+vr7oCDRkufDJfuV69eXc0RIZEEaywQEkZQPBAF+7SiRhBNFGS67777dL8fLa2WL19ut84jj1AD\nOYi4mcY2AAsX5CFqIC8S1n4cF5ES+ClWrJgqFqnn/XpAPuGPP/5oL5CIMY4fP173GD2B/E2tDePb\nb7+tFgtYZPk6J1RbRnEox6JRyHH0tKBB3ityV8PBX3/9pfJHUZSLEEJI5IMbchQ8RI0E1N8BH330\nkfJQ62lPCaMCijYOHTpU3YxreugYZYhohyVLltg1EoZ27Vg4PozVX3/9tYoigEbiRhoFBTWjvrf3\n6wFjRNFGFC7UjAfQab1j9MRXX31lj8hEJMLMmTOVRvs6J7yOehYolKjVLUBRaWzvDnw+eG+4ahxg\nzYG1EyGRBg0LhISIkSNH2m90HW94UcQPhRkrVKigboxxg4wiP3pAGgTEGpWFIaYIC8SPFtoPrwEq\nH8MAgf3iNRQJ1IBnf8qUKdKjRw8pU6aMEnp0LMDNvp736wFWdywuUIUZXn+cN4pUahg9BrwP8ESU\nLl1a9XtG+ggWZr7OCa+j2COKPuI5RD3A6OEpJBJFlrB/T9Wjgwm6WeC7UaRIkZAfmxBCSGDgRtlV\n96GHcCDAiA79gUcfhgEYuxFxpwcUSsaNL4r8Qedwsw2vvab90DekG0LjsP+VK1cqr73j+6F5eH/V\nqlXVcdEJQYsC9PV+vWPEuKDTGOO2bdvsqQBGj4FijChyjPUO1hboNqUVjvR1TngdqaQYD96PLlta\nhIYrWFeg4xS0P9TAyfHtt9/67IxFiBWJQ6GFcA+CkGgH4unYYtHRMIBQN9y0ogYADA1Ib9DA8/Cu\nI5dSCztE7iJyBSGcGtg3nkPYPowSqL0AS74GFjAIGUReH/7Gj2NoPy4DyE+EZ8FdPr+n97sbnyfQ\nHQIeeIwRXgDsD+Kud4zu5gT7ROcGGA8cazjoOSd4SVD5GQsdX+Gn6OQA7wsWMe7OGXOfnp6uzk0D\n2+AxDCdG34Pznzt3rt/GHEIIIeEFOgctdpe/rxm6oV/Qb1z3HTUM3YLQScDxxhedjlCHATfHmsZh\nO2gYdAVrB3jxtdoM0D50JcDx8J61a9eq1xxv3qGhiIrDjb9r/QNv73c3Pnc4jhE666hxesfoOicw\nFOA9MLhrc+GIt3MCqGeAlAh36wZHkMKJrlFIhcR2rueMNQRSUWvUqKGe1zpa4XOHo8jd56jnPWjj\nCaPRmDFjvM4tIVaEhgVCIhx4OdB2MVeuXGoRg1BKtKLCTTAxBxhzYOhwLLgUbJCPCsOHnnxQQggh\nsQNudmGQQJFi3KzC+z9//nwVFeBYzJkEBm7wccPvKaIxGCCVE04mx+gOQiIFFm8kJMJBjiUs8wj7\ng3UcofvoGU3MA54gx0iSUABPC40KhBBCXEEXCaQAIPINNQTguf/iiy9oVDAZpFuEGi1ygZBIhBEL\nhERJHidCA5EXiDxLQgghhEQ3SC+Ec0FLrSCEkHBCwwIhhBBCCCGEEEIMw64QhBBCCCGEEEIIMQwN\nC4QQQgghhBBCCDEMDQuEEEIIIYQQQggxDA0LhBBCCCGEEEIIMQwNC4QQQgghhBBCCDEMDQuEEEII\nIYQQQggxDA0LhBBCCCGEEEIIMQwNC4QQQgghhBBCCBGj/B8SvNdXiKe2/AAAAABJRU5ErkJggg==\n"
          }
        }
      ],
      "source": [
        "from sklearn.inspection import DecisionBoundaryDisplay\n",
        "from sklearn.neighbors import KNeighborsClassifier\n",
        "from sklearn.pipeline import make_pipeline\n",
        "\n",
        "configurations = [\n",
        "    (1, \"uniform\"),\n",
        "    (5, \"uniform\"),\n",
        "    (25, \"uniform\"),\n",
        "    (5, \"distance\"),\n",
        "]\n",
        "weight_names = {\"uniform\": \"uniforme\", \"distance\": \"distance\"}\n",
        "classes, y_encoded = np.unique(y_train, return_inverse=True)\n",
        "\n",
        "fig, axes = plt.subplots(2, 2, figsize=(11, 7), sharex=True, sharey=True)\n",
        "for ax, (n_neighbors, weights) in zip(axes.ravel(), configurations):\n",
        "    sklearn_knn = make_pipeline(\n",
        "        StandardScaler(),\n",
        "        KNeighborsClassifier(n_neighbors=n_neighbors, weights=weights),\n",
        "    )\n",
        "    sklearn_knn.fit(X_train, y_encoded)\n",
        "    DecisionBoundaryDisplay.from_estimator(\n",
        "        sklearn_knn,\n",
        "        X_train,\n",
        "        response_method=\"predict\",\n",
        "        multiclass_colors=\"Set2\",\n",
        "        alpha=0.25,\n",
        "        ax=ax,\n",
        "    )\n",
        "\n",
        "    for label, name in enumerate(classes):\n",
        "        selected = y_encoded == label\n",
        "        ax.scatter(\n",
        "            X_train[selected, 0],\n",
        "            X_train[selected, 1],\n",
        "            color=plt.get_cmap(\"Set2\")(label),\n",
        "            edgecolor=\"black\",\n",
        "            s=20,\n",
        "            label=name,\n",
        "        )\n",
        "\n",
        "    ax.set_title(f\"k={n_neighbors}, poids='{weight_names[weights]}'\")\n",
        "    ax.set_xlabel(\"Longueur du bec (mm)\")\n",
        "    ax.set_ylabel(\"Profondeur du bec (mm)\")\n",
        "\n",
        "handles, labels = axes[0, 0].get_legend_handles_labels()\n",
        "fig.legend(handles, labels, loc=\"upper center\", ncols=len(classes))\n",
        "fig.tight_layout(rect=(0, 0, 1, 0.93))\n",
        "plt.show()"
      ],
      "id": "cell-decision-boundaries"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Avec $k=1$, des observations d’entraînement individuelles contrôlent de\n",
        "petites régions, ce qui produit une frontière très irrégulière.\n",
        "Augmenter $k$ lisse la frontière, car un voisinage plus grand doit\n",
        "s’accorder. La pondération par la distance permet aux observations\n",
        "proches de conserver une plus grande influence locale.\n",
        "\n",
        "# Régression\n",
        "\n",
        "La recherche des voisins ne change pas pour la régression. Nous\n",
        "remplaçons le vote des classes par une moyenne pondérée des cibles\n",
        "numériques :"
      ],
      "id": "3456a3b2-eb80-4303-b07e-2dafb5303555"
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {},
      "outputs": [],
      "source": [
        "def regression_prediction(targets, distances, mode=\"uniform\"):\n",
        "    weights = voting_weights(distances, mode)\n",
        "    return float(np.average(targets, weights=weights))"
      ],
      "id": "regression-aggregation"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Puisque le résultat est une moyenne des cibles observées dans le\n",
        "voisinage, la régression KNN ordinaire n’extrapole pas au-delà de leur\n",
        "étendue.\n",
        "\n",
        "# Complexité et limites\n",
        "\n",
        "Pour chaque requête, cette implémentation directe calcule les distances\n",
        "en $\\mathcal{O}(ND)$ et effectue un tri complet en\n",
        "$\\mathcal{O}(N\\log N)$. Elle conserve également le jeu d’entraînement de\n",
        "taille $\\mathcal{O}(ND)$ en mémoire.\n",
        "\n",
        "Parmi les autres limites importantes :\n",
        "\n",
        "- les distances sont sensibles à l’échelle des attributs et à la\n",
        "  métrique choisie;\n",
        "- les voisinages deviennent moins informatifs dans les espaces de grande\n",
        "  dimension;\n",
        "- un petit $k$ peut être sensible au bruit, tandis qu’un grand $k$ peut\n",
        "  masquer la structure locale;\n",
        "- le déséquilibre des classes peut dominer le vote d’un voisinage.\n",
        "\n",
        "# Expériences suggérées\n",
        "\n",
        "1.  Comparez l’exactitude sur le jeu de test pour plusieurs valeurs de\n",
        "    `n_neighbors`.\n",
        "2.  Modifiez les quatre configurations de la figure des frontières de\n",
        "    décision.\n",
        "3.  Retirez la standardisation et observez comment les résultats\n",
        "    changent.\n",
        "4.  Remplacez la distance euclidienne par la distance de Manhattan.\n",
        "5.  Créez une requête identique à un exemple d’entraînement et examinez\n",
        "    la pondération par la distance.\n",
        "6.  Comparez cette implémentation au `KNeighborsClassifier` de\n",
        "    scikit-learn."
      ],
      "id": "5e765f0a-81df-456b-b7da-4129024e364d"
    }
  ],
  "nbformat": 4,
  "nbformat_minor": 5,
  "metadata": {
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3 (ipykernel)",
      "language": "python",
      "path": "/Users/turcotte/.virtualenvs/csi4106/share/jupyter/kernels/python3"
    }
  }
}